USF Libraries
USF Digital Collections

Attitudinal factors related to driving behaviors of young adults in Belize

MISSING IMAGE

Material Information

Title:
Attitudinal factors related to driving behaviors of young adults in Belize an application of the precaution adoption process model
Physical Description:
Book
Creator:
Hoare, Ismael A
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Accidents, Traffic   ( mesh )
Motor Vehicles   ( mesh )
Automobile Driving   ( mesh )
Attitude   ( mesh )
Risk-Taking   ( mesh )
Risk Factors   ( mesh )
Risk perception
Risk-taking attitudes
Risky driving behaviors
Knowledge of road laws and signs
Motor vehicle crashes
Dissertations, Academic -- Community & Family Health -- Doctoral -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: Young adults' risk-taking attitudes, risk perception, and knowledge of road laws and signs influence their driving behaviors. The adoption of risky driving behaviors increases young adults' risk of motor vehicle crashes. The purpose of this study was to increase the understanding of the factors that lead to increased risks of MVC-related mortality and morbidity for young adults in Belize, to provide support for the development of evidence-based programs, and, more importantly, to investigate the relationships involving young adults' risk-taking attitudes, risk perception, and knowledge of road laws and signs and their relation to driving behaviors. The Precaution Adoption Process Model provided the theoretical foundation for this study and was used as the framework to investigate the variables of interest. This study used a nonexperimental, cross-sectional research design to examine the relationships between the latent variables. A convenience sample of 532 students enrolled at the University of Belize participated in this study. Data were collected through the completion of the Driving Behavior Survey. Structural equation modeling was used to examine the strength and direction of relationships among these latent variables and provide a better understanding of the relationships among these latent variables. The study found that the majority of students were in the final stages of the Precaution Adoption Process Model and were exhibiting the safest behaviors. However, the risk-taking attitudes significantly contributed to the manifestation of risky driving behavior and to a lesser extent so did risk perception. The study's findings suggest that interventions should focus on lowering young adults' risk-taking attitudes and raising risk perception to reduce risky driving behaviors.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2007.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Ismael A. Hoare.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 202 pages.
General Note:
Includes vita.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001988967
oclc - 307589669
usfldc doi - E14-SFE0002190
usfldc handle - e14.2190
System ID:
SFS0026508:00001


This item is only available as the following downloads:


Full Text

PAGE 1

Attitudinal Factors Related to Driving Behaviors of Young Adults in Belize: An Application of the Precaution Adoption Process Mode l by Ismael A. Hoare A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Community and Family Health College of Public Health University of South Florida Co-Major Professor: Robert J. McDermott, Ph.D. Co-Major Professor: Wayne W. Westhoff, Ph.D. Robert F. Dedrick, Ph.D. Julie Baldwin, Ph.D. Heather G. Stockwell, Sc.D. Dale O. Ritzel, Ph.D. Date of Approval: October 26, 2007 Keywords: Risk Perception, Risk-Taking Attitudes, R isky Driving Behaviors, Knowledge of Road Laws and Signs, Motor Vehicle Cra shes Copyright 2007, Ismael A. Hoare

PAGE 2

Dedication I dedicate this dissertation to my wife, Olda, who offered encouragement, support, and love especially on those dark days when I felt like there was no reason to continue; and to my children, Alyssa, Ismael and Kieran who i nvoluntarily journeyed with me and provided the motivation and encouragement to comple te my studies. I dedicate this dissertation to the memory of my pa rents, Fulgencio and Paula Hoare who inculcated in me from early in life the d esire and ambition to achieve the highest level of education; to my brothers and sist ers who have always supported me in my academic pursuits, especially Armando who is als o completing his dissertation at the University of South Florida; to my brothersand si sters-in-law whose support facilitated my doctoral studies; and to my parents-in-law, Froi la and Valdemar Zetina, who have taken me into their family as one of their own. I also dedicate this work to my friends in Tampa, M ichael Brennan and Giovanna Brennan, Clare Guild and Jason Guild, Carol William s Oladele, and Dr Tyra Dark who shared their time and friendship with us.

PAGE 3

Acknowledgements Many individuals have contributed to the successful completion of this dissertation. I wish to acknowledge the professiona l expertise and assistance of my dissertation committee: Dr. Robert McDermott, Dr. W ayne Westhoff, Dr. Robert Dedrick, Dr. Heather Stockwell, Dr. Julie Baldwin a nd Dr. Dale O. Ritzel. I wish to particularly acknowledge and thank Dr. Robert McDer mott, my co-major professor who provided professional guidance, excellent advice, t imely feedback, and friendly encouragement throughout the life of this dissertat ion. I also thank Dr. Wayne Westhoff for his constant encouragement, professional sugges tions, and for his initial suggestion and support to pursue doctoral studies at the Unive rsity of South Florida. I thank Dr. Elizabeth Gulitz and Dr. Rony Francois for their en dless encouragement and support. I am grateful to University of Belize for supporting my doctoral studies. I acknowledge the assistance of the University of Belize’s faculty wh o facilitated the survey, and the students who willingly participated in this study.

PAGE 4

i Table of Contents List of Tables..................................... ................................................... ...............................v List of Figures.................................... ................................................... ............................vii Abstract........................................... ................................................... ..............................viii Chapter 1: Introduction............................ ................................................... .........................1 Country Background................................ ................................................... .............1 Statement of the Problem: Global Impact........... ................................................... ..2 Global Impact..................................... ................................................... .......2 Regional Impact: Latin American and Caribbean Coun tries.......................5 Motor Vehicle Crashes in Belize................... ................................................... .......8 Need for the Study................................ ................................................... ..............10 Implications for Public Health.................... ................................................... ........11 Purpose of Study.................................. ................................................... ...............12 Conceptual Model.................................. ................................................... .............13 Research Questions................................ ................................................... .............15 Overview of Study Methods......................... ................................................... ......16 Delimitations..................................... ................................................... ..................16 Limitations....................................... ................................................... ...................17 Definitions....................................... ................................................... ....................19 Chapter 2: Literature Review ...................... ................................................... ...................21 Study Background.................................. ................................................... .............21 Theoretical Foundation ........................... ................................................... ...........22 Rationale for the Use of Stage Theories .......... .........................................22 Precaution Adoption Process Model ................ .........................................23 Adaptation of PAPM ............................... ..................................................2 6 Developmental Characteristics of Young Adults (18to-24-Years Old) ..............27 Growth and Development ........................... ..............................................27 Physical Developmental Characteristics ........... ........................................28 Cognitive Developmental Characteristic ........... .......................................29 Developmental Characteristics and Driving Behavior .............................30 Risk Perceptions.................................. ................................................... ................31 Risk Perceptions and Behavior .................... .............................................31 Risk Perceptions and Optimism Bias ............... .........................................32 Risk Perceptions and Age Differences ............. ........................................34

PAGE 5

ii Risk Perceptions and Driving Skills .............. ...........................................36 Risk-Taking Attitudes ............................ ................................................... ............37 Attitudes and Behaviors .......................... ..................................................3 7 Dimensions of Risk-Taking Attitudes .............. ........................................38 Knowledge and Behavior............................ ................................................... ........42 Process to Obtain a Belizean Driver’s License..... .....................................44 Knowledge of Road Signs and Driving Behavior...... ................................45 Knowledge of Road Laws and Behavior............... ....................................47 Summary........................................... ................................................... ..................49 Chapter 3: Methods............................................ ................................................... .............50 Purpose of Study.................................. ................................................... ...............50 Conceptual Model.................................. ................................................... .50 Research Questions................................ ................................................... .52 Research Design .................................. ................................................... ...52 Population and Sample ............................ ................................................... ..........53 Population Demographics........................... ...............................................53 Sample Description................................ ................................................... .53 Minimum Sample Size............................... ................................................55 Exclusion Criteria................................ ................................................... ...56 Measures ......................................... ................................................... ...................56 Questionnaire Description......................... ................................................56 Modified Young Driver Attitude Scale.............. ........................................57 Survey Modifications.............................. ................................................... 58 Reliability of Scores from Instruments............ ..........................................59 Reliability of Scores from Pilots and Dissertation .....................................60 Confirmatory Factor Analysis of Driving Behavior S urvey......................64 Validity of Scores from Instruments............... ...........................................66 Socio-demographic Variables....................... ................................................... ......67 Age............................................... ................................................... ...........67 Ethnicity......................................... ................................................... .........67 Sex............................................... ................................................... ............67 Enrolment Status.................................. ................................................... ...67 Crash Experience.................................. ................................................... ..67 Data Collection .................................. ................................................... ................68 Procedure......................................... ................................................... .......68 Pilot Study....................................... ................................................... ....................70 Purpose and Components............................ ...............................................70 External Panel Review............................. ..................................................7 1 Mini-pilot Test................................... ................................................... .....72 Field Testing..................................... ................................................... ......72 Data Analysis .................................... ................................................... .................73 Data Entry........................................ ................................................... .......73 Univariate and Bivariate Analysis ................ ............................................74 Multivariate Analysis............................. ................................................... .74

PAGE 6

iii Research Question 1: To What Extent are the Specif ic Stages of the Precaution Adoption Process Model (PAPM) observ ed in the Study Population?.................................. ................................................... .75 Research Question 2: What is the Relationship Betw een Selected Demographic Factors and Risky Driving Behavior?.... .............................77 Research Question 3: What is the Relationship Betw een Young Adults’ Risk-Taking Attitudes and Risky Driving Beh aviors?.................77 Research Question 4: What is the Relationship Betw een Young Adults’ Knowledge of Road Laws and Signs and Risky Driving Behaviors?......................................... ................................................... ......78 Research Question 5: What is the Relationship Betw een Young Adults’ Risk Perceptions and Risky Driving Behavior s?..........................78 Chapter 4: Results............................................ ................................................... ...............79 Introduction...................................... ................................................... ...................79 Pilot Study Results............................... ................................................... ...............79 External Panel.................................... ................................................... .....79 Mini-Pilot Test at Sacred Heart Junior College.... .....................................81 Field Test at Sacred Heart Junior College ........ ........................................81 Main Study Results ............................... ................................................... .............85 Survey Summary.................................... ................................................... .85 Univariate Analyses Results....................... ................................................... ........85 Population Demographics........................... ...............................................85 Driver and Crash Experience....................... ..............................................87 Precaution Adoption Process Model Staging Variable s............................89 Normality......................................... ................................................... .......90 Bivariate Analyses Results........................ ................................................... .........92 Gender Differences................................ ................................................... .92 Research Question Results......................... ................................................... .........98 Research Question 1: To What Extent are the Specif ic Stages of the Precaution Adoption Process Model (PAPM) observ ed in the Study Population?.................................. ................................................... .98 Multivariate Analyses: Structural Equation Modelin g (SEM) Results................101 Research Question 2: What is the Relationship Betw een Selected Demographic Factors and Risky Driving Behavior?.... ...........................104 Research Question 3: What is the Relationship Betw een Young Adults’ Risk-Taking Attitudes and Risky Driving Beh aviors?...............107 Research Question 4: What is the Relationship Betw een Young Adults’ Knowledge of Road Laws and Signs and Risky Driving Behaviors?......................................... ................................................... ....108 Research Question 5: What is the Relationship Betw een Young Adults’ Risk Perceptions and Risky Driving Behavior s?........................109 Additional SEM Results............................ ..............................................110

PAGE 7

iv Chapter 5: Discussion......................................... ................................................... ..........111 Introduction...................................... ................................................... .................111 Research Summary.................................. ................................................... .........111 Discussion of Results............................. ................................................... ...........113 Research Question 1: To What Extent are the Specif ic Stages of the Precaution Adoption Process Model (PAPM) observ ed in the Study Population?.................................. ..................................................1 13 Research Question 2: What is the Relationship Betw een Selected Demographic Factors and Risky Driving Behavior?.... ...........................114 Research Question 3: What is the Relationship Betw een Young Adults’ Risk-Taking Attitudes and Risky Driving Beh aviors?...............116 Research Question 4: What is the Relationship Betw een Young Adults’ Knowledge of Road Laws and Signs and Risky Driving Behaviors?......................................... ................................................... ....117 Research Question 5: What is the Relationship Betw een Young Adults’ Risk Perceptions and Risky Driving Behavior s?........................118 Conclusions....................................... ................................................... ................118 Strengths and Limitations of Study................ ................................................... ...120 Strengths......................................... ................................................... ......120 Limitations....................................... ................................................... .....121 Data Collection Lessons........................... ................................................... ........122 Questionnaire..................................... ................................................... ...122 Logistics......................................... ................................................... .......123 Implications for Public Health.................... ................................................... ......123 Implications for Future Research.................. ................................................... ....125 References......................................... ................................................... ............................129 Appendices......................................... ................................................... ...........................151 Appendix A: IRB Approval Letter................... ................................................... 152 Appendix B: IRB Modification Approval Letter...... ...........................................154 Appendix C: Permission Letter Sacred Heart Junior College.............................156 Appendix D: Permission Letter University of Belize ..........................................157 Appendix E: Driving Behavior Survey First Draft. ...........................................158 Appendix F: Driving Behavior Survey Final Draft. ..........................................171 Appendix G: External Panel List................... ................................................... ...185 Appendix H: External Panel Review Guide........... .............................................186 Appendix I: Research Question Table............... ..................................................1 88 Appendix J: Pilot Test Review Guide............... ................................................... 190 Appendix K: SEM Output............................ ................................................... ....191 About the Author................................... ................................................... .............End Page

PAGE 8

v Lists of Tables Table 3.1 Reliability Scores for the Construct: Ri sky Driving Behavior..................62 Table 3.2 Reliability Scores for the Construct: Ri sk-Taking Attitudes ....................62 Table 3.3 Reliability Scores for the Construct: Ri sk Perceptions..............................63 Table 3.4 Reliability Scores for the Construct: Kn owledge of Road Laws & Signs.............................................. ................................................... ..........63 Table 3.5 A Summary of Confirmatory Factor Analysi s Standardized Factor Loadings .......................................... ................................................... .......65 Table 3.6 List of Courses Surveyed on Both Campus ..............................................69 Table 4.1 Mean, Standard Deviation, Skewness and K urtosis Values for the Construct: Risky Driving Behavior.................. .........................................83 Table 4.2 Mean, Standard Deviation, Skewness and K urtosis Values for the Construct: Risky Driving Behavior.................. .........................................83 Table 4.3 Mean, Standard Deviation, Skewness and K urtosis Values for the Construct: Risk Perceptions........................ ...............................................84 Table 4.4 Mean, Standard Deviation, Skewness and Ku rtosis Values for the Construct: Knowledge of Road Laws and Signs ....... ...............................84 Table 4.5 Participants’ Demographic Characteristic s ...............................................86 Table 4.6 Participants’ Driving and Crash Characte ristics .......................................88 Table 4.7 PAPM Staging Questions Frequency Distrib ution Values........................89 Table 4.8 Normality Values of Driving Behavior Sur vey Scales..............................91 Table 4.9 Effect Size Values of Driving Behavior S urvey Scales.............................94 Table 4.10 Correlation Matrix for Risky Driving Beh avior Subscales and Age ........96

PAGE 9

vi Table 4.11 Correlation Matrix for Risk-Taking Atti tudes Subscales and Age............................................... ................................................... ...........96 Table 4.12 Correlation Matrix for Risk Perceptions Subscales and Age.....................97 Table 4.13 Correlation Matrix for Knowledge of Road Laws and Signs Subscales and Age.................................. ................................................... 97 Table 4.14 Proportions of Young Adults in Each Stag e of the Precaution Adoption Process Model............................. ...............................................99 Table 4.15 Structural Model Fit Indices............ ................................................... .....102 Table 4.16 Direct Effects of Demographic Factors on Risk-Taking Attitudes [RTA], Risk Perception [RP], and Knowledge of Road Laws and Signs [KLS]........................................ ................................................... ..105 Table 4.17 Indirect Effects of Demographic Factors on Risk-Taking Attitudes [RTA], Risk Perception [RP], and Knowledge of Road Laws and Signs [KLS]........................................ ................................................... ..106

PAGE 10

vii Lists of Figures Figure 1 Conceptual Model.......................... ................................................... .........14 Figure 2 Measurement Model......................... ................................................... ......51 Figure 3 Precaution Adoption Process Model Staging Algorithm...........................76 Figure 4 Precaution Adoption Process Model Staging Algorithm results.............100 Figure 5 Final Structural Model.................... ................................................... ......103

PAGE 11

viii Attitudinal Factors Related to Driving Behaviors of Young Adults in Belize: An Application of the Precaution Adoption Process Mode l Ismael Hoare, M.P.H. ABSTRACT Young adults’ risk-taking attitudes, risk perceptio n, and knowledge of road laws and signs influence their driving behaviors. The ad option of risky driving behaviors increases young adults’ risk of motor vehicle crash es. The purpose of this study was to increase the under standing of the factors that lead to increased risks of MVC-related mortality an d morbidity for young adults in Belize, to provide support for the development of e vidence-based programs, and, more importantly, to investigate the relationships invol ving young adults’ risk-taking attitudes, risk perception, and knowledge of road laws and sig ns and their relation to driving behaviors. The Precaution Adoption Process Model pr ovided the theoretical foundation for this study and was used as the framework to inv estigate the variables of interest. This study used a nonexperimental, cross-sectional research design to examine the relationships between the latent variables. A conve nience sample of 532 students enrolled at the University of Belize participated in this st udy. Data were collected through the completion of the Driving Behavior Survey. Structu ral equation modeling was used to examine the strength and direction of relationships among these latent variables and provide a better understanding of the relationships among these latent variables.

PAGE 12

ix The study found that the majority of students were in the final stages of the Precaution Adoption Process Model and were exhibiti ng the safest behaviors. However, the risk-taking attitudes significantly contributed to the manifestation of risky driving behavior and to a lesser extent so did risk percept ion. The study’s findings suggest that interventions should focus on lowering young adults ’ risk-taking attitudes and raising risk perception to reduce risky driving behaviors.

PAGE 13

1 Chapter 1: Introduction Country Background Belize, a former British Colony, lies in the Caribb ean coast of Central America. Belize is bordered on the north by Mexico and on th e west and south by Guatemala. Belize is a nation of 22,700 km2 including surrounding cayes. Its widest point span s 109 kilometers and its length spans approximately 280 k ilometers. Belize has a population estimated at 282,600. Approximately 80% of the popu lation is 40 years or younger and less than 10% is 55 years or older (Central Statist ical Office [CSO], 2005). The 2000 population census identified the major ethnic group s as Mestizo, Creole, Ketchi, Yucatec and Mopan Maya, Garifuna and East Indians (CSO, 200 1). Other ethnic groups form the remainder of the population. Belmopan City, built in 1970, is the capital of the country and is the location for all the government ministry’s main offices. Belize exercises a parliamentary democracy based on the Westminster Model and gained its indep endence from Great Britain on September 21, 1981. The government comprises the Ho use of Representatives (elected officials) and the Senate (appointed officials). Th e major party forms the government and a few elected members form the cabinet led by the p rime minister. The country is subdivided into six administrative districts with e ach having a town board or a city council as part of the major municipality. The boar d or city council has administrative jurisdiction only for that town or city, e.g. Coroz al Town Board for Corozal Town but not for Corozal District. Each district comprises sever al villages administered by a village

PAGE 14

2 council with no coordination with the town board or city council. Village councils, town boards, and city councils are not authorized to mak e any laws. Belize has a limited road infrastructure comprising a Northern Highway running from Corozal District to Belize District, a Western Highway running from Belize to Cayo District, the Hummingbird Highway running from Belm opan City to Stann Creek District, and the Southern Highway running from Sta nn Creek to Toledo District. The total length of the highway system is approximately 330 miles (CSO, 2004). Villages have been established alongside each of the major h ighways. Each highway comprises two lanes with either one running in opposite direc tions. Speed limits are 55 miles per hour on the highways, 45 mph through villages, and 25 mph through the towns. Enforcement of speed limits is, however, rare. Statement of the Problem Global Impact. Motor vehicle crashes (MVCs) are a major cause of f atalities and injuries and a globally recognized public health pr oblem (Jacobs, Aeron-Thomas, & Astrop, 2000; Kopits & Cropper, 2003; Murray & Lope z, 1997a, 1997b, 1997c; World Health Organization [WHO], 2004a). In 2000, the est imated MVC mortality rate for the world was 20.8 per 100,000 population with a rate o f 30.8 for males and 11.0 for females (WHO, 2004a). WHO (2004a) reported that an estimate d 1.26 million people died in 2000 from MVCs worldwide, with 85% to 90% of deaths occurring in low and middle income countries (Peden, McGee, & Sharma, 2002). Mu rray and Lopez (1997a, 1997c) projected that MVCs fatalities will be the sixth le ading cause of deaths and the second leading cause of disability-adjusted life years los t in developing countries by 2020. Developed countries have studied the causes and eff ects of MVCs and have implemented

PAGE 15

3 measures to reduce the incidence (Odero, Garner, & Zwi, 1997; Soderlund & Zwi, 1995). However, low and middle income countries have lagge d in addressing the effects of MVCs, by failing to implement comprehensive interve ntions shown to be effective in reducing injury and deaths (Nantulya & Reich, 2003; Odero et al., 1997; O’Neill & Mohan, 2002; Soderlund & Zwi, 1995). Global attenti on to this health problem has gained momentum in both developed and developing co untries. The World Health Organization has taken the lead to promote awarenes s and address the impact of MVCs. WHO celebrated World Health Day in 2004 with the th eme “Road Safety” to emphasize the importance of addressing the impact o f MVCs and its global threat to health and contribution to global mortality, morbid ity and disability (Murray & Lopez, 1997b; WHO, 2004a). The celebration of World Health Day brought attention to the impact of MVCs and highlighted various related stat istics. Recent estimates on the number of MVC-related deaths range from 750,000 to 880,000 persons for 1999 with 85% of these deaths occurring in low and middle inc ome countries (Jacobs et al., 2000). Jacobs et al, (2000) also estimated worldwide MVC-r elated injuries at 23 to 34 million persons annually. This injury estimate nearly doubl es previously estimated figures (Jacobs et al., 2000). In the next 10 to 20 years, MVC deaths are projected to increase by 1 to 1.3 million persons and injuries are expected to reach as high as 50 million annually (Jacobs et al., 2000; Murray & Lopez, 1997c). By 2 020, WHO (2004b) projects that MVC deaths could increase by 65% worldwide, with an 80% increase observed in low and middle income countries if interventions do not increase or improve. Interestingly, these projections for the year 2020 differ significantly between low and middle income countries and high income countri es. For example, high income

PAGE 16

4 countries’ fatality rates are projected to be less than 8 per 100,000 versus 20 per 100,000 in low and middle income countries (Jacobs et al., 2000; Kopits & Cropper, 2003; Murray & Lopez, 1997c; WHO, 2004b). Another differe nce can be observed in the type of most vulnerable road user; in high income countr ies, most vulnerable road users are mainly car occupants, whereas in low and middle inc ome countries, pedestrians and cyclists are the most vulnerable road users (Jacobs et al., 2000; WHO, 2004b). Apart from the mortality and morbidity, MVCs produc e an additional economic burden on countries. The estimates ranged from 0.3% to 4% of gross national products (GNP) (Jacobs et al., 2000; Kopits & Cropper, 2003; WHO, 2004b). Widely accepted formulas provide a crude estimate of the economic i mpact of MVCs by using the value of 1% of the gross national product (Jacobs et al., 20 00; WHO, 2004b). However, recent studies suggest that a more realistic value would b e 2% of GNP for highly motorized countries (high income countries) and 1% of GNP val ue for less motorized countries (low and middle income countries) (Jacobs et al., 2000; Kopits & Cropper, 2005; WHO, 2004). By using this formula, the crude economic co st from MVCs is estimated at $518 billion US dollars worldwide (Jacobs et al., 2000; WHO, 2004b). Of the $518 billion, low and middle income countries incur an estimated $65 billion in MVC-related costs (Jacobs et al., 2000; WHO, 2004b). This amount exceeds annu al financial assistance that the low and middle income countries receive, thus placing a significant burden on their development (WHO, 2004b). The most recent figures f or Latin America and the Caribbean countries (LACs) showed a cost estimated at $18.9 billion for 1997 (Jacobs et al., 2000; WHO, 2004b).

PAGE 17

Regional Impact: Latin American and Caribbean count ries. The economic and social costs of MVCs in LACs provide a glimpse of t he impact of MVCs. In 2002, the Pan American Health Organization (PAHO) (2004a) rep orted that over 128,000 persons died due to MVCs in the Americas. Of the reported d eaths in 2002, the LACs of Brazil, Colombia and Mexico accounted for 46% of the fatali ties (PAHO, 2004a). Although these countries accounted for the majority of fatal ities, smaller population countries have shown disproportionate mortality rates. Mortalities rates in the LACs range from 15 per 100,000 population in high-income countries to 18.1 per 100,000 population in low and middle income countries. In Caribbean countries, th ese rates can range from 26 (Guadaloupe) to 64.1 (St. Lucia) per 100,000 popula tion (Le Franc & Alleyne, 2004; PAHO, 2004a). The disproportionate mortality rates in the Caribbean exemplify the need for concern and the urgency to address the steady i ncrease in the number of MVC-related deaths in LACs. In 2002, with a reported 30,859 MVC-related deaths, Brazil ranked first in Latin America and the Caribbean, and fifth worldwide (PAH O, 2004a). For the same year, PAHO (2004) also reported a 17.7 per 100,000 mortal ity rate for Brazil, a rate lower than other Latin American and Caribbean countries. Vasco ncellos (1999) reported that 340,000 persons were injured or killed due to MVCs in 1995, with 39% of these occurring in urban areas. Pedestrians and cyclists comprised the vulnerable road users and accounted for 60% to 70% of all fatalities (Vas concellos, 1999). Vasconcellos (1999) also identified possible reasons for the observed i ncreasing trends of motor vehicle fatalities and injuries. The MVCs were attributed t o multiple causes, such as poor traffic management, lack of enforcement of traffic regulati ons, poor road conditions and

PAGE 18

6 maintenance, and the absence of a coordinated effor t to address MVC-related deaths and injuries (Vasconcellos, 1999). These challenges req uire a coordinated effort to reduce the mortality and morbidity rates attributed to MVCs. Mexico faces a similar dilemma. With over 17,500 MV C-related deaths and a mortality rate of 14.3 per 100,000 population in 20 00, Mexico’s vulnerable road users are primarily pedestrians. Pedestrians comprise 54% of all MVC-related fatalities (Fraser, 2005; Hjar-Medina, Carillo-Ordaz, Flores-Aldana, A naya, & Lpez-Lpez, 1999; Hjar, Kraus, Tovar, & Carillo, 2001; Hjar, Vazquez-Vela & Arreola-Risa, 2003; Hjar, Arredondo, Carillo, & Solrzano, 2004). However, Me xico’s official mortality rate masks the variation observed within the country. Fo r example, a mortality rate of 28.7 per 100,000 population is reported in Baja California a s compared to 7.9 reported in Chiapas (Hjar et al., 2003). In addition to the high morta lity rates, there are about 13 injuries for every fatality noted (Hjar-Medina et al., 1999; H jar et al., 2004). The mortality and morbidity rates are observed in adults in their hig h work productivity years. Those most affected come from the uninsured populations in Mex ico and are in lower economic class populations that can least afford the loss of a wag e earner (Hjar et al., 2004). Uninsured persons with MVC-related injuries incurred an avera ge out-of-pocket cost of $80.00. This out-of-pocket cost is significant considering that the minimum daily salary in Mexico is $4.00 (Hjar et al., 2004). A large percentage of a ffected persons are not part of the formal economy, do not have a steady income source, and are the sole income earner, thus compounding the financial effect of MVC-relate d injuries (Hjar et al., 2003). Between 1991 and 1995, Colombia’s reported number o f deaths and injuries increased two-fold and three-fold, respectively (Po sada, Ben-Michael, Kahan, & Richter,

PAGE 19

7 2000). Of all deaths in 1995, 80% were males and 55 % were younger than 35 years (Posada et al., 2000). The fatalities were mostly a n urban phenomenon with 76% of deaths occurring in urban areas (Posada et al., 200 0). PAHO (2004a) reported 8,272 deaths and a mortality rate of 19 per 100,000 popul ation for 2002. In 1986 over 64,000 MVCs were reported; this reported figure increased to 231,974 recorded MVCs in 2000 with 60% occurring in just three cities (Rodrguez, Fernndez, & Velsquez, 2003). The increases recorded in 1995 and 2000 can be partiall y explained by the passage of the Traffic Accident Mandatory Insurance Policy in 1993 that introduced mandatory reporting of MVCs as a requirement for insurance re imbursement (Posada et al., 2000; Rodrguez et al., 2003). Other explanations include the rapid increase in the number of motor vehicles, poorly designed and maintained road ways, ineffective enforcement of regulations and ineffective speed control or traffi c management measures (Fraser, 2005; Posada et al., 2000; Rodrguez et al., 2003). The Caribbean region has experienced a similar rapi d increase in MVC fatalities as observed in Latin America after 1992 (WHO, 2004b ). Even though the Caribbean has a low number of motor vehicles in comparison to Lat in America, the relative risk of MVC death is significantly higher in the Caribbean (Jacobs et al., 2000). St. Bernard and Mathews (2003) examined MVC cases occurring in 2000 obtained from the database of the Traffic and highway Patrol Unit of Trinidad and Tobago. In Trinidad and Tobago, MVCs were largely an urban phenomenon paralleling t hose observed previously in LACs. They also found that the vulnerable road user s comprised mainly pedestrians, passengers and drivers, who accounted for 93% of al l fatalities and 95% of all injuries in 2000 (St. Bernard & Mathews, 2003). St. Bernard and Mathews (2003) were unable to

PAGE 20

8 determine the underlying factors that led to the MV Cs in Trinidad and Tobago. The inability to identify specific underlying factors c an be attributed to the poor datacollecting infrastructure. Jamaica faced similar ch allenges of unavailability of data sources, leading to the implementation of an emerge ncy-based injury surveillance system in 1998 (Ashley & Holder, 2002; Holder, 2002). McDo nald (2002) reported an estimated 400 MVC related deaths with a mortality rate of 18 per 100,000 population, but he was unable to identify the underlying causes leading to MVCs deaths. In an observational study conducted in Jamaica prior to the passage of the seat belt law, 21.1% of drivers and 13.6% of front seat passengers voluntarily wore sea tbelts (Crandon, Branday, Simeon, Rhoden, Thompson, & Carpenter, 1996). This low seat belt usage rate may provide a reason for the 20% general surgery admission and 77 % trauma mortalities associated with patients involved in MVCs (Crandon, Carpenter, & McDonald, 1994). The limited quantity of available studies emphasizes the need t o conduct further studies that identify factors contributing to the negative impact of MVCs in the Caribbean and support evidenced-based interventions.Motor Vehicle Crashes in Belize Apart from national and PAHO reports, just one nonpeer reviewed journal article about MVCs was found. Kim (1993) reported that male s were 2.6 times more likely than females to suffer from MVC injury and identified th e 21 to 25 age group as the one most at risk. Kim (1993) reviewed police reports from 19 90 to 1992 and found the data to be deficient in content. In 2002, MVCs were the leadin g cause of death in Belize (National Health Information and Surveillance Unit [NHISU], 2 003). Available mortality and morbidity data show that MVC mortality rates rose f rom 10.7 per 100,000 population in

PAGE 21

9 1993 to 31 per 100,000 in 1999 (PAHO, 1998, 2002). During the period of 1990 to 1998 (excluding 1992 due to unavailable data), males acc ounted for 84% of the deaths from MVCs (WHO, 2004a). MVCs accounted for 49% and 62% o f deaths from all external causes of death for the period 1993 to 1996 and 199 6 to 1999, respectively (PAHO, 1998, 2002). Silvi (2004) reported that Belize had the hi ghest male-to-female death ratio of 5.4 per 100,000 population relative to 12 countries dur ing 1985 to 2001, but did not identify whether these were adjusted rates or not. Proportio nally, Belize reports one of the highest mortality rates in LACs in 2002 with 30.1 per 100,0 00 population (PAHO, 2004a). Mortality rates for males increased from 14.4 per 1 00,000 population in 1993 to 55 per 100,000 in 1999, whereas female rates changed from 6.9 per 100,000 population to 7.4 per 100,000 for the same period, (PAHO, 1998, 2002) In 1998, two age groups, 0 to 14 years and 15 to 39 years, represented the majority of MVC fatalities (70%), with 16 and 54 de aths, respectively (National Health Information and Surveillance Unit [NHISU], 2003). D ata from the Joint Intelligence Coordinating Center of the Police Department (2005) show that in 2003 2,508 MVCs were documented with 68 fatalities and 2,622 in 200 4 with 61 fatalities. Hospitalizations due to MVCs for the same period are unavailable. Although the MVC mortality and morbidity rates sign ificantly impact the health of Belizeans, the estimates may need to be adjusted by 25% to account for general underreporting that occurs in developing countries (Kopi ts & Cropper, 2005). Further studies are needed to identify the various factors that lea d to or increase the risk of MVCs. There is a paucity of information, data or published repo rts on MVCS epidemiological,

PAGE 22

10 economic, and risk factor data. This scarcity of da ta hinders the development of interventions that effectively address MVCs in Beli ze. Need for the Study As mentioned, limited studies are available to prov ide the true cost of the impact or the underlying factors leading to the increased levels of MVCs in LACs. This limitation is observed in studies related to the av ailability of MVC related mortality and morbidity data and statistics, inconsistency in app lication of classification codes, identification of vulnerable road users, lack of un iformity in MVC related definitions, identification of risk factors and the development and application of appropriate countermeasures (Forjuoh, 2003; Jacobs et al., 2000 ; Kopits & Cropper, 2005; Nantulya & Reich, 2002, 2003; Odero et al., 2003; Odero, Kha yesi & Heda, 2003; Posada et al., 2000). To identify MVC research conducted in LACs, Hjar (2004) compiled 186 studies and documents only 37% of which were papers publish ed in journals. The rest comprised interviews with experts, abstracts of scientific me etings, grey papers and documents from non-governmental organizations and international ag encies. Hjar (2004) did not indicate whether these documents were readily accessible. Th e absence of critical research about MVCs poses significant obstacles in developing rese arch-based interventions and programs. Even with the limited research conducted in LACs, c ommon trends have been identified. Pedestrians are the most vulnerable roa d users in Brazil, Colombia and Mexico (Hjar-Medina et al., 1999; Hjar et al., 2001; Pos ada et al., 2000; Vasconcellos, 1999). Commonly found conditions leading to increases in M VC-related death and injuries in LACs included poor traffic management systems, lack of enforcement of laws, poor road

PAGE 23

11 conditions and lack of speed limit enforcement (Hj ar-Medina et al., 1999; Hjar et al., 2001; Kim, 1993; Posada et al., 2000; Vasconcellos, 1999, WHO, 2004b). The impact of MVC on the health of Belize’s populat ion and the scarcity of published research in this field suggest an urgency to conduct studies. These studies could support findings common to regional countries or determine whether interventions developed and evaluated in high-income countries ar e applicable in Belize. Currently, the most common causes of MVCs for 2000 to 2003 in Beli ze are those reported by the National Police Headquarters and include: inattenti on/misjudgment, reversing turning error, negligent pedestrians/cyclist, failure to gi ve way and failing to obey traffic rules (CSO, 2004). Flores (1999) reported similar causes for MVCs occurring in Belize during 1998. However, these reported causes are related to charges that are applied to the offender and do not provide sufficient detail to id entify the true cause. Failing to obey traffic rules does not provide any detail as to whi ch rule/law in particular has not been obeyed, further suggesting that research is needed.Implications for Public Health WHO (2004b) has recognized the impact of MVCs world wide and declared the 2004 World Health Day to promote awareness, encoura ge discussion and mobilize action to address MVCs. The Ministry of Health in Belize a lso has recognized the urgency in developing intervention programs to address the eno rmous challenge in maintaining a healthy young population. What needs to be addresse d is the collection of data relating to MVCs injuries in Belize. A systematized approach fo r addressing injuries, especially those related to MVCs, in Belize is practically non existent. The interventions applied in Belize do not appear to be based on studies providi ng necessary data or theoretical basis

PAGE 24

12 for their application. The approach of implementing interventions without in-depth investigation as to whether they are appropriate to the Belizean population may not appropriately incorporate the factors affecting or impacting drivers’ behavior and other factors contributing to MVCs in Belize. Effective i nterventions need to incorporate the drivers’ and passengers’ behavioral components to i mpact the negative consequences of MVCs. This study’s investigation of young adults’ p erceptions of risk, risk-taking attitudes and knowledge, and their relationship to risky driving behaviors contributes to the understanding of the impact of these factors on causes and support possible solutions to reduce MVCs. Because a lack of adequate epidemiological and soci oeconomic data on MVCs at the national level impedes effective national and i nternational response (WHO, 2004d), this study adds significantly to the body of knowle dge on MVCs in Belize. Findings from this study support a systematized approach in the d evelopment and implementation of intervention programs addressing the effects of MVC s. Purpose of the Study Young adults are over represented in mortality and morbidity rates in both high income countries (HICs) and low income countries (L ICs) (Afukaar, 2003; Afukaar, Antwi & Ofosu-Amaah, 2003; Flores, 1999; Forjuoh, 2 003; Nantulya & Reich, 2003; National Committee for Injury Prevention and Contro l, 2005; Odero et al., 2003; Rodriguez et al., 2003; Smith, 1993; St. Bernard & Matthews, 2003). The principal investigator in this study acquired data to increas e the understanding of the factors that lead to increased risks of MVC-related mortality an d morbidity for young adults in Belize, and to provide support for the development of evidence-based programs.

PAGE 25

13 Specifically, the purpose of this study was to inve stigate the relationships involving young adults’ risk-taking attitudes, risk perceptio ns, knowledge, and driving behaviors. This study used a quantitative research design that explored these four constructs using the Precaution Adoption Process Model (PAPM) as a t heoretical framework to evaluate factors influencing driving behaviors of young adul ts at the University of Belize. Conceptual Model The conceptual model for this study is depicted in figure 1. The conceptual model is based on the premise that young adults’ ri sk-taking attitudes, risk perceptions, and knowledge are related to their driving behavior s. Furthermore, the adoption of safe or risky driving behaviors influences their risk of MV Cs, which may lead to increased mortality and morbidity risks.

PAGE 26

Risk Perception Risk-taking Attitudes Knowledge of road laws and signs Risky Driving Behavior Figure 1. Conceptual Model unaware of issue unengaged by issue deciding about acting decided not to act decided to act maintenance acting

PAGE 27

Research Questions This study provided data to answer the following q uestions: Research Question 1: To what extent are the specific stages of the PAPM observed in the study population? Research Question 2: What is the relationship between selected demograph ic factors and risky driving behavior? Research Question 3: What is the relationship between young adults’ risk -taking attitudes and risky driving behaviors? Research Question 4: What is the relationship between young adults’ know ledge of road laws and signs and risky driving behaviors? Research Question 5: What is the relationship between young adults’ risk perceptions and risky driving behaviors?

PAGE 28

16 Overview of Study Methods This study used a cross-sectional correlational de sign to collect primary data from young adults enrolled at the University of Belize. In this study, participants were restricted to the 18-to-24-year-old student populat ion at two campuses of the University of Belize located in Belmopan City and Belize City. The restriction criteria comprise age, education, enrollment at the university and exposur e to commuting. Participants were enrolled in either bachelor or associate degree pro grams at the university. Data were collected through the use of a self-administered qu estionnaire. The questionnaire was completed during class sessions. The questionnaire took an estimated 45 minutes to complete. The restriction criteria helped to contro l for a number of factors and provided a more homogenous population. Data collected in this study were the first known attempt to record and understand factors that contribute to or are related to, the risks of MVC injury and death in Belize. Previous studies have p rovided sparse details on epidemiological data related to MVC related injurie s and deaths. Delimitations The delimitations section describes parameters for the study and the population to which the study results may be generalized (Heppner & Heppner, 2004; Pyrczak & Bruce, 2000). The applicable delimitations of this study are described below Data for this study were collected from young adult s: in the age range from 18-to 24-years representing a n age group of Belize’s population at risk for MVC deaths and inju ry, enrolled at the University of Belize during the 200 6 to 2007 academic year,

PAGE 29

17 previously enrolled at various feeder tertiary leve l institutions in Belize, and may represent the student populations at other tertiary level institutions in Belize, and representing a portion of tertiary level students w ho commute to the university sites in Belmopan City and Belize City. The university has been selected because its databa se is of better quality and student data are accessible. The results of this st udy may be generalizable to young adults enrolled at the University of Belize. The results m ay also be generalizable to students enrolled in other tertiary level institutions in Be lize, because the university student population is derived from these feeder institution s. Findings from this study may be generalizable to students who are drivers or passen gers in vehicles commuting to the University of Belize campuses in Belmopan City and Belize City. Limitations Limitations describe methodological weakness or fac tors that potentially weaken the validity or interpretation of the study’s resul ts (Heppner & Heppner, 2004; Pyrczak & Bruce, 2000). This study has several limitations th at are described in the following paragraphs. Participation in this study was voluntary and based on self-reporting from the participants. The survey instrument collected data on issues that may be sensitive to social desirability bias. The self-reporting may in crease the possibility of social desirability bias that has been found in studies ut ilizing questionnaires and interviews. Participants in this study were limited to young ad ults ages 18-to-24-years-old who were enrolled at the University of Belize during the 200 6 to 2007 academic year. The

PAGE 30

18 participants in this survey may differ from the you ng adults in the general population who are not enrolled in a tertiary level institution or who are employed or unemployed. The study data were collected during a two-month pe riod at the University of Belize and provides a snapshot of the participants’ experience. This experience may be influenced by recent MVCs that have received major publicity in the Belize media. This heightened publicity may provide increased particip ation in the completion of questionnaire items as a result and may influence t he responses provided. Even though the questionnaire was lengthy and parti cipation was voluntary, the number of participants that declined to participate was not as high as the anticipated rate ranging from 20 to 50 percent. Demographic data suc h as gender, age, student status, were not collected from the two participants who de clined to complete the survey. Hence a determination of whether differences existed betw een those who participate and those who do not was not carried out. The cross-sectional design of this study does not a llow for changes that occur over time, and therefore, the findings may be limit ed in their application. The crosssectional design of this study limits the conclusio ns that can be drawn and the results are not appropriate for the establishment of cause and effects of the variables in this study. This cross-sectional study is correlational in natu re. The analysis is guided by sound theory but any causal relationship inferred does no t meet the rigorous requirements of an experimental study. Therefore, conclusions and infe rences drawn from the results must be restricted to the nature of correlational data.

PAGE 31

19 Definitions District – A district is a geographical region tha t represents a subdivision of the country. In Belize, the country is divided into six districts. Fatality – A person involved in a motor vehicle cra sh who was killed outright or who died within 30 days after the crash (WHO, 2004b). Knowledge – For this study, knowledge refers to th e cognizance of road rules in Belize, risks of drinking and driving, and effectiveness of seat belts. Risky driving behavior – Risky driving behaviors a re those driving practices that increase the possibility of a negative health outco me (Ulleberg & Rundmo, 2002). Some examples of these risk behaviors include, but are not necessarily limited to speeding, distracted driving, aggressive driving an d not adhering to traffic laws. Risk perception – Risk perception refers to the su bjective experience of risk in potential traffic hazards (Deery, 1999). Risk-taking attitude – For this study, risk-taking attitude is defined as dimensions that affect preferences towards risk-taking in traf fic (Ulleberg & Rundmo, 2002; Ulleberg & Rundmo, 2003). Tertiary level institutions – Two-year institution s that provide associate degree level education in Belize.

PAGE 32

20 University of Belize – Belize’s only national unive rsity that has been authorized to offer degree programs and consists of four facultie s: Faculty of Education and Arts, Faculty of Managemen t and Social Science, Faculty of Nursing and Allied Healt h, and Faculty of Science and Technology Vehicle – For this study, vehicle refers to a mech anized mode of transportation such as cars, pickup trucks, motorcy cles and trucks. Vulnerable road user – A term applied to those most at risk in traffic Young adult – Individuals whose age ranges from 18 to 24 years.

PAGE 33

21 Chapter 2: Literature Review Study Background MVCs have been identified as contributing significa ntly to the high morbidity and mortality rates in Belize (Joint Intelligence Coord inating Center of the Police Department [JICCPD], 2005; NHISU, 2003; PAHO 2002, 2004a). Cu rrently in Belize, a centralized depository for data on morbidity and mortality due to MVCs does not exist, nor is there a coordinating body tasked with researching and apply ing its findings to reduce fatal and nonfatal injuries related to MVCs. Presently, stati stical information is collected by three agencies under the auspices of three separate gover nment ministries. The fragmented collection of data contributes to an uncoordinated approach to promote interventions that are designed to reduce MVCs in Belize and their sub sequent health effects. Even with the accepted importance of MVCs, Belize has not identif ied or charged any particular institution or agency with the responsibility to pu rsue in-depth research into the causes of, and solutions to address MVCs. Research that thorou ghly identifies, addresses, and analyzes the numerous factors contributing to the M VCs in Belize is urgently needed. Successful interventions addressing the MVC-related mortality and morbidity utilize measures that include engineering, educational, and legislative principles. Research into the factors influencing mortality and morbidity wil l provide the basis for selecting, developing and implementing intervention programs t hat incorporate engineering (Evans, 2003; Evans, Fielding, Brownson et al., 2001; Gross man & Garcia 1999; Retting, Ferguson, & McCartt, 2003), educational (Grossman & Garcia, 1999; Rivara Thompson

PAGE 34

22 & Cummings, 1999; Shults, Elder, Sleet et al., 2001 ) and legislative (DeYoung, 1999; DeYoung, 2000; Figuerido, Rasslan, Bruscagin, Cruz, & Rocha, 2001; McArthur & Kraus, 1999; Peck & Voas, 2002; Shepherd, 2001) mea sures previously proven successful in reducing fatal and nonfatal injuries. Of course, any research targeting MVCs in Belize mu st utilize a structured approach that will provide the foundation for possi ble interventions. This present study addresses a specific aspect of MVCs in Belize by fo cusing on the behavioral factors that contribute to driving behaviors, and by extension, contribute to the high rates of MVCrelated mortality and morbidity. The following sect ions of this chater explore the theoretical foundation for this study that provides the underlying principle for utilizing the Precaution Adoption Process Model (PAPM) as the framework to investigate the variables of interest. This presentation is followe d by an analysis of the developmental characteristics specific to 18-to-24-year-olds. The following sections also explore and discuss the variables of interest that provide the basis for the research questions, i.e., young adults’ risk-taking attitudes, risk perceptio ns, knowledge, and driving behavior, and provide the rationale to investigate the relati onships involving these variables. Theoretical Foundation Rationale for the Use of Stage Theories. Traditional theories of health behaviors, such as the Theory of Reasoned Action, the Theory o f Planned Behavior, the Health Belief Model and Social Cognitive Theory, have been used to address behavior by exploring the various factors that contribute to th e actual behavior (DiClemente, Crosby, & Kegler, 2002; Glanz, Rimer, & Lewis 2002; Schwarz er, 1999; Weinstein, Rothman, & Sutton, 1998; Weinstein & Sandman, 2002). These the ories seek identification of

PAGE 35

23 variables that impact action and combine them to pr edict behavior (Weinstein, 1988; Weinstein & Sandman, 2002). These theories have bee n used successfully to investigate and address factors affecting health behavior. Howe ver, in studies addressing unintentional injuries, few theories and planning m odels have been utilized to reduce or mitigate the effects of unintentional injuries (Noa r & Zimmerman, 2005; Trifiletti, Gielen, Sleet, & Hopkins, 2005). This lack of utili zation implies that the application of theories in studies related to unintentional injuri es is needed. Precaution Adoption Process Model. A theory that can be applied to the field of unintentional injury prevention is the Precaution A doption Process Model (PAPM). PAPM is a stage theory that has been applied previo usly to injury prevention and safety practices of families (Trifilletti, 2003). PAPM cou ld provide a new approach to these behaviors, such as adoption of protective health be havior against osteoporosis (Blalock et al., 1996), radon safety (Weinstein & Sandman, 1992 ; Weinstein & Sandman, 2002) and safety practices (Trifilletti, 2003). PAPM proposed and later revised in 2002 by Weinstein and Sandman (Glanz, Rimer, & Lewis 2002; Rutter & Quine, 2002; Weinstein & Sandman, 2002) is a stage theory that may be appl ied to address MVC issues. The PAPM arose from Weinstein’s (1988) critique of cont inuum theories where he proposed four constructs that supported stage theories as an alternative to continuum theories. The development of the PAPM was supported by Weinstein’ s research on home radon testing and the decision process that determined whether th e homeowner tested for radon or not. The model proposed that decisions followed a sevenstage process: unaware of issue (stage 1), unengaged by issue (stage 2), deciding a bout acting (stage 3), decided not to act

PAGE 36

24 (stage 4), decided to act (stage 5), acting (stage 6) and maintenance (stage 7) (Weinstein & Sandman, 1992; Weinstein & Sandman, 2002). The original version of the PAPM offered four assum ptions that supported this model. One of the assumptions states that stages re present meaningful distinctions among individuals and would require documentation of this difference (Weinstein & Sandman, 1992; Weinstein & Sandman, 2002). This distinction is important for the development of stage-based interventions targeting individuals in the various stages. The distinction between someone who has decided not to act and some one who is unaware of the issue is one that can determine the content and focus of int ervention programs. The second assumption is that the factors that pre dict movement between stages differ at each stage in the PAPM (Weinstein & Sandm an, 1992; Weinstein & Sandman, 2002). In other words, the variables that determine whether a person becomes engaged in the issue differ from those that determine whether the person acts on the decision. Therefore, a different set of predictor variables i s expected for each stage of the PAPM. Thirdly, the assumption that perceptions of person al susceptibility have a strong influence on decisions about actions indicates that optimistic biases have to be overcome (Weinstein & Sandman 1992; Weinstein & Sandman, 200 2). This optimistic bias usually impedes individuals from making an accurate assessm ent of the level of personal risk they are facing. This perceived level of optimism d eters individuals from feeling personally threatened by the risks of not adopting the precaution. Lastly, the fourth assumption is that the behaviors and opinions of others have a strong influence on hazard responses (Weinstein & S andman, 1992; Weinstein & Sandman, 2002). The adoption of certain types of pr ecautions is influenced by other

PAGE 37

25 individuals rather than as a result of independent analysis and decision. This adoption scenario is especially true for certain types of pr ecautions that have few available information resources, limited availability of reso urces, prolonged time of benefit appearance and proximity of personal risk assessmen t. Stage theories offer several advantages over contin uum theories. A stage theory suggests an ordering where persons are expected to progress through the stages to arrive at the endpoint of action or maintenance of behavio r (DiClemente, Crosby, & Kegler, 2002; Glanz, Rimer, & Lewis 2002; Rutter & Quine, 2 002; Weinstein & Sandman, 2002). However, the progression through these stage s does not necessarily conclude with action or maintenance, nor does it imply that it is irreversible. The achievement of the variables in the stages determines this progression Another element of a stage theory points out that people in the same stage face commo n barriers to change (DiClemente, Crosby, & Kegler, 2002; Glanz, Rimer, & Lewis, 2002 ; Rutter & Quine, 2002; Weinstein, & Sandman, 2002). The commonality of bar riers within the stage suggests that program developers would utilize them as part of their programs to encourage movement through the stages. Finally, people in dif ferent stages face different barriers to change (DiClemente, Crosby, & Kegler, 2002; Glanz, Rimer, & Lewis, 2002; Rutter & Quine, 2002; Weinstein, & Sandman, 2002). If the ba rriers were similar throughout the stage process, then the concept of stage would be r edundant. Therefore, barriers encountered in the seven stages are expected to dif fer from each other.

PAGE 38

26 Adaptation of PAPM. I propose that these seven stages can be adapted t o explore MVCs by using the following schema: Stage 1 unaware of issue (MVCs) Stage 2 unengaged by issue (MVCs) Stage 3 deciding about acting (adopt safe or risky driving behavior) Stage 4 decided not to act (adopt risky driving be havior) Stage 5 decided to act (adopt safe driving behavio r) Stage 6 acting (practice safe driving behavior) Stage 7 maintenance of safe driving behavior. These seven stages can be applied to driving behavi ors and investigate the relationship with risk-taking attitudes, risk perce ptions, and knowledge of young adults. In particular, it is hypothesized that the effects of these three constructs will impact movement from stages three, four, and five within t he PAPM leading to the adoption of risky driving behaviors. Risk perceptions, risk-tak ing attitudes, and knowledge of young adults affect driving behavior and influence young adults’ decisions to engage in risky driving behaviors (Assum, 1997; Deery, 1999; Ullebe rg, 2002; Ulleberg & Rundmo, 2002; Ulleberg & Rundmo, 2003). Prior to expanding on the aforementioned variables of interest, the developmental characteristics of youn g adults are discussed in the subsequent section.

PAGE 39

27 Developmental Characteristics of Young Adults (18-t o-24-years old) Growth and Development. Humans transition through various stages starting from birth to adulthood. Specific changes occur as human s develop through these stages. In some instances, consensus has been reached on the v arious components that constitute a specific growth phase (Berk, 2004; Cameron, 2001; G oldscheider & Goldscheider, 1999; Huebner, 2000). This study focuses on young adults between the ages of 18 and 24 years and relates its findings specifically to this age g roup. The transition from adolescence to young adulthood has raised considerable debate in d etermining the point at which this transition occurs, including whether the transition is delineated by specific milestones or highlighted by underlying characteristics (Arnett, 2000; Nelson & Barry, 2005). Commonly accepted transition milestones may include physical, self-concept, emotional, sexual, psychological, and cognitive dev elopmental characteristics or may take the form of societal assigned roles or respons ibilities (Arnett, 2000; Berk, 2004; Cameron, 2001; Goldscheider & Goldscheider, 1999; H uebner, 2000; Malina & Bouchard, 2004; National Research Council and Insti tute of Medicine [NRCIOM], 2005; Shanahan, Porfeli, Mortimer, & Erickson, 2005). Sha nahan et al. (2005) refer to five distinct markers that define the transition into ad ulthood, namely, completion of studies, the start of a career, leaving home, marriage, and parenthood. The adoption of these roles signifies that the youth have abandoned the identit y of adolescence (Arnett, 2000; Nelson & Barry, 2005; NRCIOM, 2005; Shanahan et al., 2005) For the purpose of this study, the focus is on the transition milestones that identify physical and cognitive developmental changes signaling common characteristics of young a dults 18 to 24 years of age rather than the adoption of societal roles as outlined by Shanahan et al., (2005). The reason for

PAGE 40

28 choosing these characteristics rests on their capac ity to contribute to the actualization of behavior. Physical Developmental Characteristics. Individuals are expected to follow a typical maturation process but are not expected to adhere to a strict timeline. This expectation is based on the premise that progressio n through the maturation process differs from one individual to the next (Arnett, 20 00; Berk, 2004; Cameron, 2001; Goldscheider & Goldscheider, 1999; Huebner, 2000; M alina & Bouchard, 2004; Shanahan et al., 2005). For example, males and fema les differ in their changes as they progress through the maturation process. However, t ypical changes are expected over time. Several physical characteristic that 18-to-24-yearolds are expected to have achieved include attaining full adult stature and t he completion of the maturation process. Body structures should have reached maximum capacit y and the initiation of senescence should be ongoing (Berk, 2004; Cameron, 2001; Golds cheider & Goldscheider, 1999; Huebner, 2000). Athletic skills including strength, speed, endurance, and motor performance that increased dramatically during earl y teen years are now peaking (Berk, 2004; Cameron, 2001; Goldscheider & Goldscheider, 1 999; Malina & Bouchard, 2004). Decline in athletic ability and motor performance c an be largely attributed to a change into a less active lifestyle rather than on biologi cal degeneration (Berk, 2004; Cameron, 2001; Goldscheider & Goldscheider, 1999; Malina & B ouchard, 2004). Developments of secondary sex characteristics are expected to have reached full maturity. At this stage in the maturation process, the individual has reached full growth in physical characteristics.

PAGE 41

29 Cognitive Developmental Characteristic. Similar to the physical developmental changes, cognitive developmental transformations ca n be observed as individuals move from adolescence into young adulthood. The young ad ults’ thinking pattern, abstract conceptualization, meta-cognition and reasoning ski lls change dramatically when compared with the early teen years. Broad changes a re observed in the thinking process that deviate from knowledge acquisition to applicat ion of knowledge for problem solving purposes. These thought processes go beyond Piaget’ s four operational stages. These post-formal thought processes allow young adults to move away from the acceptance of absolute truths to the recognition of multiple trut hs. In other words, the thought processes shift from Perry’s concepts of dualistic thinking into relativistic thinking stage (Arnett, 2000; Berk, 2004; King, 1978; Love & Guthr ie, 1999). Perry’s theory emphasizes the transition from accepting truths to constructing knowledge to fit context (Berk, 2004; King, 1978; Love & Guthrie, 1999). Si milar to Perry’s theory, Schaie (1983) posits that young adults’ cognitive developm ent evolves from the acquisitive stage into the achieving stage. Schaie’s theory of post-f ormal thought addresses the transition from immediate planning to long-term planning and, therefore, highlights the importance of applying knowledge to problem solving and linkin g problems to context (Berk, 2004; Schaie, 1983). A third theory that is consistent wi th this developmental stage is Labouvie-Vief’s (1996; 1999) portrait of adult cogn ition. Labouvie-Vief proposes a pragmatic approach in the thinking process (Berk, 2 004; Labouvie-Vief, 1996; LabouvieVief, 1999). Logic is used as a means by which to a rrive at solutions that embrace realistic and sometimes ambiguous explanations. Aga in the thinking process here centers on the application of gained knowledge rather than acquisition.

PAGE 42

30 A common thread can be observed within the three th eories, that is, the complexity of the thinking process broadens as the individual matures into a young adult. Acceptance of absolute truths give way to the conce ptualization of other possibilities and the ease of using cognitive tools to arrive at othe r potential solutions to life’s problems. The development of advanced reasoning and abstract thinking skills allows the young adult to explore probable resolutions, identify und erlying principles, offer hypothetical scenarios, and depart from accepted truths. Cogniti ve development changes in the young adult foster increased autonomy as well as more def ined career goals and expectations. Developmental Characteristics and Driving Behavior. Young adults have attained certain physical and cognitive development stage ch aracteristics that provide them with the capacity to operate a motor vehicle effectively Young adults’ physical development stage allows for the motor skills, reflexes and eye -hand coordination necessary to drive safely on the roadways. Their cognitive development fosters independent decisionmaking processes. Road rules and regulations are in terpreted not only as clear cut guidelines but are also interpreted to fit the cont ext of the driving environment. The interpretation of the road rules and regulations tr anslates into driving behavior that may involve safe or unsafe driving practices. The clear manifestations of these interpretations are of keen interest to researchers studying the im pact of factors affecting driving behavior.

PAGE 43

31 Risk Perceptions Risk Perceptions and Behavior. Road users’ behaviors are implicated as a major cause of MVCs and have been investigated to identif y the specific mechanism or role they play in MVCs. Research has identified risk per ception, risk-taking attitudes and being cognizant of traffic laws as factors that hav e an impact on behavior (Evans, 2004; Ulleberg, 2002; Ulleberg & Rundmo, 2002; Ulleberg & Rundmo, 2003). Early research focused on identifying the effect and mechanisms by which the various dimensions of risk perception influence driving behavior (Brnstr m, Kristjansson, & Ulln, 2005; Brown, 2005; Evans, 2004; Iversen, Rundmo & Klempe, 2005; Rundmo & Iversen, 2004; Ulleberg & Rundmo, 2003). Some of these inves tigations scrutinized the differences in risk perception among age groups to determine and explain any identifiable difference (Brnstrm et al., 2005; Brown, 2005; Ev ans, 2004; Iversen et al., 2005; Rundmo & Iversen, 2004; Ulleberg & Rundmo, 2003). T hese studies teased out the numerous factors that contribute to the differences in risk perceptions and built on earlier pioneering studies on risk perceptions. Further research on risk perception revealed that d riving behaviors were affected by variables, such as optimism bias (DeJoy, 1989), age differences (Finn & Bragg, 1986; Jonah, 1986; Trnkle, Gelau & Metker, 1990), crosscultural differences (Sivak, Soler, Trnkle, & Spagnhol, 1986; Sivak, Soler & Trnkle, 1989), gender differences (Evans, 2004; Laapotti & Keskinen, 2004; Mathews & Moran, 1 986), driving experience and exposure (Jonah, 1986; Sagberg & Bjrnskau, 2006; S venson, 1978; Trnkle, Gelau,& Metker, 1990), and seatbelt usage (Svenson, Fischho ff, & MacGregor, 1985). Subsequent investigations identified the dimensions of risk pe rception that significantly influenced

PAGE 44

32 driving behavior to be optimism bias (DeJoy, 1989; Jonah, 1986; Weinstein, 1980, 2003), age differences (Finn, & Bragg, 1986; Jonah, 1986; Trnkle, Gelau & Metker, 1990), and driving experience and exposure (Jonah, 1986; Sagbe rg & Bjrnskau, 2006; Svenson 1978; Trnkle et al., 1990). Risk Perceptions and Optimism Bias. Weinstein investigated the concept of optimism bias in a series of published studies star ting in 1980s. The first part of his twopart study on optimism bias explored unrealistic op timistic beliefs people held that negative life events were more likely to happen to others and positive life events would more likely happen to them. The study also examined possible factors that contribute to development of these beliefs (Weinstein, 1980). Thi s study was conducted using a college student sample that rated a list of positiv e and negative life events. Weinstein (1980) found that students rated their chances of e xperiencing positive events as higher than their comparison group, M = 15.4%, t (255) = 6 .8, p < .001 and, conversely, their chances of experiencing negative events as lower th an their comparative group, M = 20.4%, t (255) = 13.9, p < .001. Correlation analys is revealed different patterns for the comparison of variables and comparative ratings. D egree of desirability and perceived probability were positively correlated with mean ra tings of positive events, .45 p < .01 and .74 p < .001, respectively. Personal experience was positively correlated with mean ratings of negative events, .42 p < .01. Perceived controllability and stereotype salience were negatively correlated with mean ratings of neg ative events, .67 p < .001 and .76 p < .001, respectively. In either case, students rated their chances of experiencing positive life events as higher than their comparison group and, c onversely, provided lower ratings for negative life events. Stereotypic salience and perc eived controllability seemed to

PAGE 45

33 contribute to the student’s optimistic ratings of l ife events. Once students perceived that the life event was controllable and were committed to the outcome, their optimism was bolstered by comparing themselves with inappropriat e standards or stereotypes. The second part of the study confirmed the initial find ings. Weinstein concluded that the students perceived that their actions, lifestyle an d personality placed them at an advantage when compared to their peers. Weinstein n oted that this perception persisted even when the students’ risks of experiencing negat ive life events were deemed as being high. This finding meant that young people are incl ined to perceive their actions as being better and more attractive than those of their peer s. Other studies explored Weinstein’s concept of optim ism bias and found it to be applicable in larger representative community sampl es (Weinstein, 1987, 1989, 1998, 2003; Weinstein, Klotz, & Sandman, 1988; Weinstein, Sandman, & Roberts, 1990). Even though Weinstein’s findings were obtained from stud ies encompassing a multitude of life events, the key concepts of risks perception and op timism bias have been found to be applicable and relevant to young adults’ driving an d other health behaviors (Brnstrm et al., 2005; Brown, 2005; Chambers & Windschitl, 2004 ; Deery, 1999; Dejoy, 1989; Harre, Foster, & O'Neill, 2005; Weinstein, 2003; Williams 2003). Dejoy (1989) investigated the link between risk per ceptions and optimism bias in a sample comprised of college age students. In this study, Dejoy (1989) examined the mechanisms that led to the inflated beliefs of poss essing superior driving skills and abilities, and lowered perceived risks of MVCs. Par ticipants ages 18 to 36 were asked to rate 10 different MVC scenarios. Six constructs of optimism bias and their effect on risk perception were explored to determine their relatio nship with specific MVC scenarios.

PAGE 46

34 The constructs of controllability of the crashes an d ease of imagining the individual involved in a crash were significantly correlated t o optimism. Multiple regression analysis indicated that the construct of controllab ility of the crashes, F (1,8) = 75.33, p < 0.001 significantly influenced optimism. Further an alysis showed that the individuals with high levels of optimism indicated that they we re more skillful ( r = -.37, p < 0.001), considered themselves to be safe drivers ( r = -.45, p < 0.001) and less like to be involved in a MVC ( r = .45, p < 0.001). Although the younger drivers were able t o identify the driving risky situations or behaviors, they did not identify the risk as applying to them, but rather, to other drivers in their group. Simila r to the concepts explored by Weinstein and others (Brnstrm et al., 2005; Brown, 2005; Ch ambers & Windschitl, 2004; Deery, 1999; Dejoy, 1989; Harre, Foster, & O'Neill, 2005; Weinstein, 1987; 1989; 1998; 2003), these findings suggested that optimism bias influen ces risk perception of young drivers and, by extension, their driving behaviors. Risk Perceptions and Age Differences. Finn and Bragg (1986) compared how risk perception differed when assessing driving situatio ns in young male drivers 18-to-24 years of age as compared with older male drivers 38 -to-50-years of age. They reported that young drivers perceived their risks of being i nvolved in a crash as significantly lower than their older counterparts after reviewing drivi ng situations that included tailgating, driving at night, speeding, driving on snow covered roads, and driving after drinking (Finn & Bragg, 1986). Finn and Bragg (1986) showed that not only were the younger drivers’ risk perceptions lower than the older comp arison group, but the younger drivers’ perceptions of being involved in a crash were lower than their own peers. This lowered perception of being involved in a crash seemed to c ontradict the study’s finding that the

PAGE 47

35 young and older groups both perceived that younger drivers are most at risk of MVCs. However, the contradiction is indicative of drivers ’ optimism bias (DeJoy, 1989; Jonah, 1986; Weinstein, 1980; 2003) and perceived superior driving skills (Mathews & Moran, 1986). More research is needed to understand this a pparent discrepancy. Mathews and Moran (1986) obtained similar results when they inv estigated the relationship between perceived risks and perceived driving skills in two groups with age ranges of 18 to 24 and 35 to 50. Their study showed that younger drivers p erceived their risk of being involved in a crash as lower than their peers, rated their d riving skills as superior to their peers, and viewed their reflexes to be better than older drive rs skills (Mathews & Moran, 1986). The results suggested that younger drivers believed tha t they possessed the same skills and abilities of more experienced older drivers. Thus, younger drivers estimated their risk of MVCs as being substantially lower than their peers and the older group (Deery, 1999; Mathews & Moran, 1986; Williams 2003). The over-rat ed driving skills of drivers in the 18-to-24-year age group seemed to affect driving be havior and adoption of safe driving practices. Based on their findings and the literatu re at the time, Mathews and Moran (1986) posited that drivers’ knowledge of their abi lity had an effect on their risk perception which in turn influenced their driving b ehavior. Trnkle, Gelau and Metker (1990) found that young m ale drivers between ages 18 and 21 consistently rated the risk of a crash lower than their older comparison group of 35 to 45 years, a finding consistent with the liter ature at the time. They also found that females consistently rated risk of crashes as highe r than their male counterparts. The findings from this study led Trnkle et al. (1990) to conclude that younger male drivers were more accepting of risky driving situations, we re rating risky driving situations much

PAGE 48

36 lower than the other groups, and may have poorly de veloped driving skills. The findings suggested that education programs should target the low risk perception and acceptance of high-risk situations to reduce the risk of crash es and suggested the need for further research in risk perception to determine whether th ese conclusions were accurate. A study by Leung and Starmer (2005) reinforced the conclusion that young drivers have lowered risk perceptions than mature d rivers (Deery, 1999; Finn & Bragg, 1986; Frick, Rehm, Knoll, Reifinger, & Hasford, 200 0; Sagberg & Bjrnskau, 2006; Williams 2003) and overrated driving skills (Jonah, 1986; Mathew & Moran, 1986; Svenson 1978; Trnkle et al., 1990). Leung and Star mer (2005) used an experimental design to illustrate how alcohol influences risk pe rception. They showed that cognitive processes are essential for accurate risk perceptio n, and thus, can influence driving behavior. Leung and Starmer (2005) concluded that m ajor differences existed between young and mature drivers when comparing driving beh avior related to risk perception. Risk Perceptions and Driving Skills. Risk perceptions or the subjective experience of risk in potential traffic hazards (Deery, 1999) can influence how drivers interpret and decide on driving behaviors. Deery’s proposed model posits that novice drivers use different cognitive processes than experienced driv ers to assess hazards and decide on risky driving behaviors. Three central differences are observed. Deery (1999) concluded that novice drivers do not recognize and identify h azards as efficiently as experienced drivers. Novice drivers have a narrow scope of visu al perception that expands with driving experience to allow for a more holistic per ception and identification of hazards. Secondly, novice drivers detected lowered risks in specific traffic hazards than more experienced drivers. In other words, they were unab le to identify the subsequent elevated

PAGE 49

37 risk of traffic hazards than their more experienced counterparts. The last notable difference between novice and experienced drivers c an be observed in the determination of risk perception. Even though novice drivers perc eived their risks of accidents rather precisely, they still held the optimistic belief th at their chances of being involved in a crash were much lower than their peers. Concepts similar to Deery’s (1999) were incorporate d into a campaign promoting safe driving behavior in two counties in Norway (Ru ndmo & Iversen, 2004). Rundmo and Iversen (2004) evaluated the campaign’s effecti veness and examined the interactions among perception, behavior, and personality. They found that differences in risk perceptions that were incorporated into educational campaigns involved specific cognitive processes that influenced driving behavio r. The study demonstrated a change in risk perception and an indirect effect on driving b ehavior in their sample of 18-to-24year-old students (Rundmo & Iversen, 2004). Partici pants in this study were able to perceive risks much more than at the inception of t he campaign and reported fewer instances of risky driving behaviors. Rundmo and Iv ersen’s (2004) study provides another piece of evidence linking risk perception a nd its influence on driving behavior. Risk-Taking Attitudes Attitudes and Behaviors. The link between attitude and behavior has been explored since the early 1900s. The link between at titude and behavior is based on the assumption that conceptually, attitude influences, induces, or molds behavior. Kraus (1995) catalogued the trends of research findings t o inform current attitude-behavior correlations research. Initial research questioned this basic premise to the extent of refuting the link between attitude and behavior (Kr aus, 1995). The consensus that refuted

PAGE 50

38 the attitude-behavior link was challenged in the ea rly 1970s. Fishbein and Ajzen (1972) argued that the prior research contained methodolog ical flaws that failed to identify the link between attitude and behavior statistically. T he flaws centered on the failure to link the appropriately identified attitude measure to it s corresponding behavior (Fishbein & Ajzen, 1972). Once the appropriate measures for att itude and its corresponding measures for behavior were identified, the correlations betw een attitude and behavior were found to be greater than .40 (Ajzen & Fishbein, 1977). Kraus ’s (1995) meta-analysis found that reported attitude-behavior correlations ranged from -0.10 to 0.91. Kraus (1995) concluded that prior studies showed that attitude a nd behavior were highly correlated once the appropriate corresponding measures for eac h concept were utilized. The observed correlations suggested that attitudes sign ificantly contributed to the determination of behavior but could not be isolated as its sole determinant. However, the existing evidence is strong enough to support resea rch that seeks to identify the specific attitude-behavior correlations (Ajzen, 1988; Ajzen & Fishbein, 1977; Assum, 1997; Fishbein & Ajzen, 1972; Kraus, 1995; Parker, 2002; Ulleberg & Rundmo, 2002; Ulleberg & Rundmo, 2003; Whissell & Bigelow, 2003). Dimensions of Risk-taking Attitudes. Assum (1997) studied the relationship between correct or right attitudes, incorrect or wr ong attitudes, and behavior and their relationship with accident risk. The 7,425 responde nts of a random sample for this study were selected from the Norwegian driver’s license r egister. A survey measured general attitude related to road safety and road traffic be havior (Assum, 1997). The study found a significant difference between drivers who had corr ect or right attitudes and those who had incorrect or wrong attitudes towards traffic sa fety and speeding. Although Assum’s

PAGE 51

39 conclusion was that a direct link between the attit udes measures and accident risk was not significant, his study supported the link between a ttitudes and behavior by presenting evidence that a significant difference in behavior is observed between drivers who had the right or correct attitudes and those who did no t. This study highlights the need to apply the appropriate measures for attitude and its corresponding measures for behavior (Ajzen, 1988; Ajzen & Fishbein, 1977; Assum, 1997; Fishbein & Ajzen, 1972; Kraus, 1995). In effect, the match between attitude and be havior was explored by using concise, narrowed measures of risk-taking attitudes and link ing them to clearly defined risk driving behaviors. The relationship between risk-taking attitudes with risky driving behaviors has been established (Iversen, 2004; Malfetti, Rose, De Korp & Basch, 1989; Parker, 2002; Ulleberg & Rundmo, 2002; Ulleberg & Rundmo, 2003; W est & Hall, 1997). For this study, the term “risk-taking attitudes” is defined as dimensions that affect preferences towards risk-taking in traffic (Ulleberg & Rundmo, 2002; Ulleberg & Rundmo, 2003). Risk taking refers to driving in a way that does no t contravene traffic laws but increases the risk of being involved in a crash (West & Hall, 1997). Ulleberg and Rundmo (2002) explored the premise that by addressing risk-taking attitudes a change in driving behavior would be achieved. Dimensions of risk-taking attitu des were measured by using the Young Drivers Attitude Scale (YDAS) developed by Ma lfetti et al. (1989). The survey was administered in Norway to 4,500 adolescents and young adults ranging in age from 16 to 23 years (Ulleberg & Rundmo, 2002). Risk-taki ng attitudes explained 50% of the variance of risk-taking behavior (Ulleberg & Rundmo 2002). Ulleberg and Rundmo (2002) found that lower risk taking attitudes were correlated with less risk taking

PAGE 52

40 behavior. Ulleberg and Rundmo (2003) also investiga ted the relationship of personality, risk taking attitudes, risk perception and risky dr iving behavior using multiple regression methods. The standardized path coefficient ( = .79) indicated the size of the direct effect of risk-taking attitudes on risk-taking behavior. A weak effect was detected for the risk perception-risky behavior component. Ulleberg and Rundmo (2003) suggested that the weak effect of the risk perception-risky behavior r elationship may be explained by the weak measures that were utilized in their study. Th eir suggestion implies that risk perception may still be significantly related to ri sky driving behavior in the same realm as risk-taking attitudes. The Ulleberg and Rundmo (200 3) study also suggests that attitudes about speeding may contribute significantly to risk y driving behavior. Whissell and Bigelow (2003) investigated this association using an attitudinal scale to identify the link between speeding violations and reported crashes. A significant correlation r (158) = .40, p < 0.01, was found between driving attitudes and spe eding (Whissell & Bigelow, 2003). However, their small convenience sample of 283 univ ersity students makes it difficult to generalize the findings to the young adult populati on and suggests that further studies are needed to investigate this relationship. Iversen (2004) investigated the relationship betwee n risk-taking attitudes and risky driving behavior in a random sample of Norweg ian drivers. The participants completed two surveys. The second survey was done 1 2 months after the first and focused on three attitudinal dimensions encompassin g rule violations and speeding, careless driving, and drinking and driving. Risk-ta king attitudes were correlated to risky driving behavior. Using structural equation modelin g analysis, the three dimensions of risk-taking attitudes explained 52% of the total va riance of risky driving behavior. An

PAGE 53

41 important finding of this study was that persons wi th attitudes towards risk-taking actions seem to carry out risky driving behaviors. Iversen (2004) suggests that the dimensions of risk-taking attitude in this study seemed to predic t future risky driving behavior. The study suggest that the attitudinal dimensions encom passing rule violations and speeding, careless driving, and drinking and driving may dete rmine future risky driving behaviors and may form an important aspect of safety campaign s focusing on curbing risky driving behaviors (Iversen 2004; Iversen & Rundmo 2004). Fu rther studies comprising less heterogeneous groups may provide a better understan ding of how these dimensions affect specific age subgroups. Risk-taking attitudes and their component dimension s have been investigated to determine their individual and combined influences on risky driving behavior (Assum, 1997; Greening & Stoppelbein, 2000; Iversen 2004; I versen & Rundmo, 2004; Iversen, Rundmo & Klempe, 2005; Malfetti et al., 1989; Ulleb erg & Rundmo, 2002; West & Hall, 1997; Whissell & Bigelow, 2003; Yagil, 1998). Previ ous studies indicate that a strong link exists between these two variables. Fishbein a nd Ajzen (1972) and Kraus (1995) identified methodological flaws in the studies, whi ch did not find any correlation between attitudes and behaviors, and suggested the requisit e need for well-designed research incorporating the measures that match the specific levels of attitudes and behaviors. Seven risk-taking attitude dimensions seemed to con tribute to the understanding of how risky driving behaviors are determined. The seven r isk-taking attitude dimensions include speeding, safe driving, riding with an unsafe drive r, concern for others, concern for oneself, drinking and driving, and safety belts (As sum, 1997; Greening & Stoppelbein, 2000; Iversen 2004; Iversen & Rundmo, 2004; Iversen Rundmo & Klempe, 2005;

PAGE 54

42 Malfetti, Rose, DeKorp & Basch, 1989; Pinksky, Labo uvie, Pandina & Laranjeira, 2001; Ulleberg & Rundmo, 2002; Vanlaar & Yannis, 2005; We st & Hall, 1997; Whissell & Bigelow, 2003; Yagil, 1998). These results have bee n obtained from diverse populations in Brazil, Canada, Israel, New Zealand, Norway, Uni ted Kingdom and the United States. Although the environmental, infrastructural, social legal and cultural settings differ, the similar results obtained from studying risk–taking attitudes and risky driving behaviors suggest that the concepts may be applicable to the Belize young adult population. The findings from such diverse populations can inform t he design of a study investigating the effects of risk-taking attitudes on driving behavio r in Belize and provide a platform to expand on these studies to investigate whether simi lar results will be obtained with the young adult population. Knowledge and Behavior In developing countries, the alarming rates of MVC -related deaths and injuries are fueled by certain conditions. These conditions incl ude a lack of road infrastructure, scarcity of regulating legislation, a dearth of edu cational, engineering and legislative interventions designed to mitigate the negative eff ects of motorization, the populace’s inexperience and adaptation with increased motoriza tion, and increasing motorization of developing countries (Evans, 2004; WHO, 2004b). In developing countries, the absence of adequate measures effectively addressing the imp act of MVCs may be rooted in the inexperience to develop and implement a coordinated approach to this health problem. As a developing country, Belize faces similar challeng es, for example, divided responsibilities with addressing the effects of MVC s, absence of dedicated funding, and inadequate resources. In contrast, developed countr ies have well-established

PAGE 55

43 governmental and non-governmental agencies tasked w ith researching MVCs, developing policies, and designing, testing, implem enting and evaluating educational, engineering and legislative interventions targeting MVC-related injury and deaths. A strongly recommended component of effective inter vention to address MVCs is to utilize educational strategies (Dinh-Zarr et al., 2 001; Task Force on Community Preventive Services, 2001). Educational strategies increase the awareness of an d encourage adherence to motor vehicle laws, safety measures and risks of MV Cs (Dinh-Zarr et al., 2001; Task Force on Community Preventive Services, 2001; WHO, 2004b). Educational interventions’ objectives are based on the underlyi ng assumption that safe driving behavior may be a consequence of combined and conti nued learning opportunities (Cottrell, Girvan, & McKenzie, 2002). The education al interventions may implement programs that are designed to increase the driver’s knowledge of safe driving behavior, MVC risks, road rules and regulations, to name a fe w (Hedlund, Shults, & Comptom, 2003; Masten & Hagge, 2004; McKnight & Peck, 2003). The purpose of increasing drivers’ knowledge rests on the principle that know ledge influences behavior. Graduated driver licensing (GDL) is one such program. GDL pro grams strive to promote safer driving behavior by extending the period that the n ovice driver is able to gain and apply knowledge of safe driving practices, road laws and driving experience (Hedlund, Shults, & Comptom, 2003; Masten & Hagge, 2004; McKnight & P eck, 2003). Other components of GDL programs include exit tests, hazard percepti ons, speed restrictions, and extended learner’s permit holder’s period (Ferguson, 2003; H edlund, & Comptom, 2004; 2005; Rice, Peek-Asa & Kraus, 2004). The following sectio ns will provide a description of

PAGE 56

44 obtaining a Belize driver’s license, the link of kn owledge of Belizeans road laws and road signs and driving behavior, and knowledge of motor vehicle crash risks and driving behavior. Process to Obtain a Belizean Driver’s License. The Department of Transport is the body that is responsible for the registration, issuance and control of all vehicles and driver’s licenses in Belize (Attorney General’s Min istry [AGM], 2003). Persons can apply for a Belizean driver’s license once they mee t the prerequisite criteria. The criteria to obtain full driving privileges include being 17 years of age or older, obtaining a medical check up, obtaining a 70% passing mark on a written exam, and successfully completing a practical exam (AGM, 2003). The writte n exam tests the applicant’s understanding of the rules of the road, road signal s and road signs (AGM, 2003). The written exam is based on a 46-item handout sheet co ntaining a list of information covering topics related to the Motor Vehicle and Ro ad Traffic Act [MVRTA] and road usage (AGM, 2003). The written and practical exam c an be taken at any of the district’s office. Drivers’ licenses are valid for one calenda r year and are renewable on the holder’s birth date. Licensed drivers can renew their licens es annually without having to perform any written or practical exam again (AGM, 2003). Th e process of obtaining a driver’s license is not an intensive process and is applied at the discretion of the transport officer (AGM, 2003). The use of discretion by the transport officer may lead to subjective interpretation and application of the MVRTA legisla tion as well as the issuance of licenses to unqualified drivers.

PAGE 57

45 Knowledge of Road Signs and Driving Behavior. Drivers in Belize do not go through a stringent process to obtain a driver’s li cense. The written and practical driving exams are applied at the discretion of the transpor t officer. Section 31 (3) of the MVRTA states that the written test “shall include a test of the applicant’s knowledge of the rules of the road, road signals and road signs” (AGM, 2003). The handout sheet provided to applicants contains five items providing written in formation on road signs (Department of Transport [DOT], 2004). One refers to the speed limits for various vehicles. The second one refers to the legality of the road signs Three of the handout items refer to the “no entry”, “keep right” and “yield” road signs (DO T, 2004). It is important to note that the handout sheet only describes the road signs con tent and their purpose but diagrammatic samples of these road signs are not pr ovided. The failure to provide more information and samples of road signs belies their important contribution to traffic safety and forces the driver to learn the meaning through experience. Most importantly, the three road signs that are described do not reflect the undetermined number o f road signs used in the roads of Belize. Hence, the unnecessary challenge of independently interpreting road signs is presented to novice driv ers as part of their learning process. The independent interpretation of road signs may lead t o driving behavior that is contrary to the intended road sign message. Road signs are extensively used as an integral part of road designs, as well as an important component of roads’ safety design (Al-Mad ani, 2000; Al-Madani & Al-Janahi, 2002a; 2002b). Road signs convey information to dri vers by using either alphanumeric messages or symbols (Crundall & Underwood, 2001; Jo rgensen & Wentzel-Larsen, 1999). The information conveyed alerts drivers of road conditions and possible hazards,

PAGE 58

46 or provides recommendations that are necessary for safe driving (Charlton, 2004, 2005; Crundall & Underwood, 2001; Van Houten & Retting, 2 001). Road sign effectiveness is affected by a driver’s understanding of its message (Al-Madani, 2000; Al-Madani & AlJanahi, 2002a, 2002b; Charlton, 2004, 2005; Crundal l & Underwood, 2001). Comprehension of road signs is imperative for the m essage to be useful to the driver (AlMadani, 2000; Al-Madani, & Al-Janahi, 2002a; 2002b; Charlton, 2005). Al-Madani and Al-Janahi (2002a; 2002b) surveyed drivers in five A rabian Gulf countries to determine their comprehension of road signs and the factors a ffecting their interpretations. They found that drivers accurately identified and interp reted between 50% and 60% of the roads signs. They suggested that the low comprehens ion rate was a reflection of the ineffective learning system associated with the dri vers’ licensing process and recommended an overhaul of the system for increasin g driver’ comprehension of road signs. Several evaluation methods to test the effectivenes s of road signs have been developed. Early research used the roadblock paradi gm to assess drivers’ recollection of road signs they had recently passed on the road (Ch arlton, 2005; Fisher, 1992; Johansson & Backlund, 1970; Jorgensen & Wentzel-Larsen, 1999) Investigators that used the roadblock paradigm stopped drivers a short distance after passing road signs and questioned the drivers to determine their recollect ions of the road signs (Johansson & Backlund, 1970). Studies utilizing the roadblock pa radigm showed that drivers had poor recollections of roads signs they had passed (Johan sson & Backlund, 1970; Jorgensen & Wentzel-Larsen, 1999). The poor recollection was de emed to represent the ineffectiveness of road signs and suggested that re sources should be invested in other

PAGE 59

47 safety measures (Crundall & Underwood, 2001; Fisher 1992). Fisher (1992) challenged this accepted conclusion, and found that drivers un consciously adjusted their driving after passing road signs alerting them of a road hazard. The findings suggested that the effectiveness of road signs should be assessed by e valuating their capacity to alert drivers of road hazards rather than by assessing drivers’ r ecollection of the content of the road signs (Fisher, 1992). In response to the poor perfo rmance of the roadblock paradigm, Crundall and Underwood (2001) proposed the priming paradigm to explain the warning potential of road signs and their subtle effects on driving behavior. Drivers unconsciously responded to the warnings from road signs by adjust ing their driving behavior to fit with the data provided (Charlton, 2004, 2005; Crundall & Underwood, 2001; Van Houten & Retting, 2001). These recent studies have reinforce d the important contributions of road signs as a component of road safety strategies. Knowledge of Road Laws and Behavior. The handout sheet provided to applicants for a Belize driver’s license contains 21 items re lated to the MVRTA and 20 items on “Do’s & Don’ts” of road use (DOT, 2004). The topics covered on the information sheet include an item on speed limits, overtaking practic es, right of way at a stop sign, age requirements for licensing, and two items related t o obsolete practices. The DOT handout (2004) does not provide detail about laws related t o safe driving behaviors or road laws (e.g., seatbelt use/law, speed, or driving under th e influence of alcohol or drugs). The list of “Do’s & Don’ts” of road use does not cover topic s concerning laws pertaining to safe driving behaviors or road laws (e.g., seatbelt use/ law, speed, or driving under the influence of alcohol or drugs) (DOT, 2004). The pra ctice of providing a 46-item handout to new drivers’ license applicant is in direct cont radiction to emphasis given to driver

PAGE 60

48 education as an effective intervention to reduce MV Cs (Carstensen, 2002; Hatakka, Keskinen, Gregersen, Glad, & Hernetkoski, 2002; Hed lund & Comptom, 2005; Mayhew & Simpson, 2002). Apart from not providing a compre hensive overview of the MVRTA, the alarming fact is that this DOT handout is the i nformation that is provided to applicants regardless of the category of vehicle th ey plan to drive (i.e., motorcycle, trucks, cars, farm vehicles, and so on). A novice d river would have to gain knowledge either through driving experience or from other sou rces. This type of learning process and a knowledge base filled with inaccurate information may foster the development of poor driving habits. Rather than developing safe drivin g behaviors, uninformed drivers may focus more on driving skills, capability and experi ence and may give more weight to their abilities (Asiamah, Mock, & Blantari, 2002; Zhang, Huang, Roetting, Wang, & Wei, 2006). Zhang et al. (2006) suggested that these dri ving behaviors and attitudes may be reflective of a poor knowledge base of road laws. Similarly, Asiamah et al. (2002) found that Ghanaia n drivers attributed crash risks to vehicle and road infrastructure rather than to b ehavioral factors including those associated with alcohol use, and concluded that a m ore aggressive campaign to raise the level of awareness of the MVC risk associated with alcohol use was needed. In addition to the publicity of MVC risk, promotion of laws rel ated to alcohol use and driving was needed as a component for interventions addressing MVCs. Ferguson and Williams (2000) also provide support for the need to increas e awareness of zero tolerance laws to impact driving behavior. A survey of 17-to-20-yearold drivers illustrated that 31% to 56% of the drivers were aware of the zero tolerance laws in their state. Ferguson and Williams (2000) suggest that educational campaigns are needed to raise awareness of

PAGE 61

49 these laws before compliance can be expected to occ ur. Masten and Chapman (2004) studied three instructional methods to test their e ffectiveness to improve drivers’ knowledge of laws and attitudes and found significa nt improvements in knowledge and attitudes. Instructional methods have been used suc cessfully to improve drivers’ selfawareness and general knowledge of proper road use (Eby, Molnar, Shope, Vivoda, & Fordyce, 2003). Without a doubt, the process of edu cating new drivers in Belize needs to be restructured to reflect the gains made through d evelopment of effective drivers’ education methods. Summary The literature supports the importance of conductin g a study to understand the relationships among risk-taking attitudes, risk per ceptions and knowledge, and driving practices that can contribute to safer driving beha viors in Belize. Findings from this study may contribute significantly to the development of intervention programs in Belize. Conceptually, the three independent variables are l inked to driving behavior, and therefore, suggest that a significant portion of va riance may be explained by these variables. The use of stage theories in the investi gation of decision models appears to be applicable to investigating factors contributing to MVCs in Belize, particularly, in relation to risk-taking attitudes, risk perceptions and knowledge, and driving behavior of young adults at the University of Belize. This stud y attempts to fill that gap in knowledge in three ways: (1) by investigating the effects of the three latent variables on driving behavior, (2) by providing information on driving b ehavior, and (3) by applying the PAPM to investigate factors affecting driving behav ior of young adults attending the University of Belize.

PAGE 62

50 Chapter 3: Methods Purpose of Study The purpose of this study was to increase understan ding of the factors associated with the risks of MVC-related mortality and morbidi ty for young adults in Belize. Specifically, this study investigated the relations hips involving young adults’ risk-taking attitudes, risk perceptions, knowledge, and driving behaviors. The study used a quantitative research design to explore these four constructs using the Precaution Adoption Process Model (PAPM) as a theoretical fram ework to evaluate factors influencing driving behaviors of young adults at th e University of Belize. Conceptual Model. The theoretical framework for this study was depic ted in figure 1 (See Chapter 1). This framework was based on the premise that young adults’ risk-taking attitudes, risk perceptions, and knowle dge influence their driving behaviors. Furthermore, the adoption of safe or risky driving behaviors was related to their risk of MVCs, which explained the increased mortality and m orbidity risks experienced by young adults. The conceptual model was further expa nded to include the separate dimensions that were used to analyze the possible r elationships among the variables of interest. A full depiction of the analysis diagram of the conceptual model can be viewed in Figure 2.

PAGE 63

51 Risk Perception Risk-taking Attitudes Knowledge of Road Laws & Signs Speeding Knowledge of Road Laws Knowledge of Road Signs Speeding Concern for Others Drinking and Driving Riding With an Unsafe driver Cognition-based Perception Distracted Driving Not Adhering to Traffic Laws Figure 2. Measurement Model Risky Driving Behavior Aggressive Driving Concern-based Perception Emotion-based Perception

PAGE 64

52 Research Questions. The following questions were explored in this stud y: 1. To what extent are the specific stages of the PAPM observed in the study population? 2. What is the relationship between selected demograph ic factors and risky driving behavior? 3. What is the relationship between young adults’ risk -taking attitudes and risky driving behaviors? 4. What is the relationship between young adults’ know ledge of road laws and signs and risky driving behaviors? 5. What is the relationship between young adults’ risk perceptions and risky driving behaviors? Research Design. This study employed a cross-sectional survey resea rch design. The cross-sectional research design was used to exa mine the patterns of relationships among the independent variables, risk-taking attitu des, risk perceptions, and knowledge, and the dependent variable, risky driving behaviors The study design permitted the examination of the strength and direction of relati onships among these variables within the young adult population at the University of Bel ize and illustrated patterns in these relationships. The cross-sectional research design did not allow for the discovery of cause and effect relationships because no experimental de sign was employed. A cross-sectional research design did, however, provide a better unde rstanding of the relationships among the variables that would provide the basis for futu re studies. Initial approval from the Office of Research Compliance Institutional Review Board (IRB) at the University of South Florida (USF) (IRB # 104876) (See Appendix A) was obtained prior to conducting any data collection. A waiver of written consent wa s also granted as part of the initial approval. A modification approval was obtained afte r the questionnaire had been

PAGE 65

53 reviewed through the pilot testing of the initial q uestionnaire (IRB # 104876G) (See Appendix B). As part of the IRB approval process a letter of support that granted permission to conduct the pilot study was obtained through the Office of the President of Sacred Heart Junior College (See Appendix C). An ad ditional letter expressing support and granting permission to conduct the full study w as obtained through the Office of the Provost of the University of Belize (See Appendix D ). Population and Sample Population Demographics. Belize, a country located in Central America, is divided into six districts: Corozal, Orange Walk, B elize, Cayo, Stann Creek and Toledo. Belize has 291,800 inhabitants with nearly 61% of t he population under the age of 25 years as estimated by the 2000 population census (C SO, 2001, 2005). Mestizo/Spanish, comprising 48.7% of the population, is the largest ethnic group (CSO, 2001, 2005). Creoles (24.9%) are the second largest ethnic group Mayas (Ketchi, Mopan, and Yucatec) comprise 10.6% of the population (CSO, 200 1, 2005). Ethnic groups such as Garifuna (6.1%), Mennonite (3.6%), East Indian (3%) along with other minor ethnic groups complete the distribution in the population (CSO, 2001; 2005). The majority of the population resides in rural areas. The largest urban city is Belize City with a population of 60,800 (CSO, 2001; 2005). The most po pulated district is the Belize district with a population of 87,000; the least populated di strict is the Toledo District with a population of 27,600 (CSO, 2001; 2005). Sample Description. For this study, the target population consisted of students enrolled at the University of Belize (UB) in the fa culties of Education and Arts, Management and Social Science, Nursing and Allied H ealth, and Science and

PAGE 66

54 Technology. UB has four campus sites respectively l ocated in Belize City, Belmopan City, Central Farm and Toledo. UB’s major campus is in Belmopan City. The UB recently moved to Belmopan City and it is anticipat ed that eventually, all major educational operations will be relocated there. The recent move to Belmopan City has led to an increase in the commuting population attendin g UB, and that fact has increased the exposure to road traffic experiences of these stude nts. The population of interest for this study was stude nts within the age of 18-to-24 years. A sample from the student population of each of the faculties was surveyed. This population was chosen because it provided an identi fiable and accessible group of individuals in this age group. Students enrolled at the University of Belize were likely to include individuals who: 1. were legal adults in the age range of 18-to 24-year s representing an age group of Belize’s population at risk for MVC deaths and injury, 2. were qualified for full driving privileges (18 year s), 3. had reached legal drinking age status (18 years), 4. had access to motor vehicles, 5. were similar to Belize’s tertiary level student pop ulation and demographic composition, and 6. were accessible to study. Interventions can be designed and implemented to fo cus on this population, and the structured environment provides a platform to condu ct research. For the second semester of the academic year 2006-2 007, 2,471 students were enrolled at the University of Belize. Student ages ranged from 15 to 55 years; 56.5%

PAGE 67

55 were fulltime students; and 63.1% were females. The ethnic composition of UB’s student population is unknown as this information is not co llected as part of the registration process. Of the enrolled students, the 18-to-24-yea r-old students totaled 1,276, and form 61% of the student population. These young adults w ere enrolled in both Bachelor and Associate degree programs at UB. The questionnaire was applied to a convenience sample of the entire young adult population ages 18 -to-24 years. The sample for this study was drawn from the student population of the academic year 2006-2007. The survey was completed by UB students enrolled in the Associate and Bachelor degree programs at both the Belmopan City and Belize City campus. Minimum Sample Size. Structural Equation Modeling (SEM) requires rather large samples for analysis ( http://www.fleshandbones.com/readingroom/pdf/946.pd f ). SEM sample sizes are difficult to calculate in advance by using exact equations. Recommended sample sizes are estimated based on the number of p arameters. These parameters are estimated on the number of measured variables in th e model. Sample sizes are usually estimated by multiplying the number of parameters p er variables by a factor of 10 ( http://www.fleshandbones.com/readingroom/pdf/946.pd f ). A sample size of 200 to 400 is commonly recommended for SEM analysis ( http://www2.chass.ncsu.edu/garson/pa765/structur.ht m ). The analytic model for this study measured 26 parameters. This number is below the recommended maximum of 91 parameters that can be measured for this model. Bas ed on the 26 parameters, a minimum sample size of 260 would be recommended for this st udy. The number of completed questionnaires targeted was 550.

PAGE 68

56 Exclusion Criteria. Students not within the age range of 18-to-24 year s were excluded from this study. Students enrolled at the UB Toledo University Center also were excluded from this study. Students enrolled at the UB Toledo University Center comprised students from the southernmost districts of Belize and are not representative of the diversity that can be found at the Belmopan and Belize City campuses. Students attending UB’s Regional Language Center (RLC) were not included in the sample. RLC students come from various countries to study Engli sh as a second language. Their ESL program typically is one academic year in length. T he RLC students are not representative of student populations in Belize and they do not take courses with the general student UB population. RLC students may hav e been exposed to different transportation experiences in their respective coun tries that may confound findings in this study. Due to these differences, they were not incl uded in this study. Measures Questionnaire Description. This study employed a self-administered questionnaire to collect data for analysis. Student s enrolled in courses selected through the sampling process completed the questionnaire. T here was no single, pre-existing instrument available to survey the four constructs of interest: risk-taking attitudes, risk perceptions, knowledge of road laws and signs, and risky driving behavior. Instruments were available that focused on one or two construct s only. The instrument that was used for this study combined questionnaire items from si x instruments to measure the constructs of interest. The questionnaires used wer e the modified Young Driver Attitude Scale (YDAS) by Ulleberg and Rundmo (2002) based on the original developed by Malfetti et al. (1989); the Risk Perceptions Survey developed by Ulleberg and Rundmo

PAGE 69

57 (2003); Knowledge of Road Laws, adapted from Whitin g, Dunn, March and Brown (1998); Department of Transport [DOT] written test (DOT, 2004); and Motor Vehicles And Road Traffic laws of Belize (Attorney General’s Ministry, 2003 Ed.), Risky Driving Behaviors, adapted from Ulleberg and Rundmo (2002) and socio-demographic questions, adapted from the CSO (2001) census questionnaire. T he questionnaire for this study comprised six sections. Each section was designed t o obtain information on the following elements: 1) risk-taking attitudes; 2) risk percept ions; 3) knowledge of road laws and signs; 4) risky driving behaviors; 5) the PAPM stag ing questions; and 6) sociodemographic data. Modified Young Driver Attitude Scale. Malfetti et al. (1989) developed an instrument to assess risk-taking attitudes of young drivers and their relationship with risky driving behaviors. The questionnaire tested s even dimensions of risk-taking attitudes and behaviors in a group of U.S. and Cana dian students. These seven dimensions included speeding, safety belt use, safe driving, drinking and driving, riding with an unsafe driver, myself, and concern for othe rs. The entire questionnaire had a total of 70 items. Upon further testing by Ulleberg and R undmo (2002), five dimensions were selected out of the seven dimensions: speeding, uns afe driving, riding with an unsafe driver, drinking and driving, and concern for other s. Measurement of these five dimensions encompassed a total of 19 questions, and formed the risk-taking attitudes component of the questionnaire. Eight questions cov ering three dimensions of risk perceptions were used from a survey developed by Ru ndmo and Iversen (2004). The three risk perception dimensions included emotion-b ased risk perception, cognition-based risk perception and concern-based risk perception. The section of the questionnaire

PAGE 70

58 addressing knowledge of road laws and signs was ada pted and developed from Whiting, Dunn, March and Brown (1998), Department of Transpo rt written test (DOT, 2004), and Motor Vehicles And Road Traffic laws of Belize (Att orney General’s Ministry, 2003 Ed.) and included a total of eight questions covering tw o dimensions: knowledge of traffic signs and knowledge of road laws. Risky driving beh aviors comprised four dimensions and the nine questions were drawn from a questionna ire developed by Ulleberg and Rundmo (2002). The placement of participants in the PAPM algorithm was done using a total of six questions. The last section of the que stionnaire included six questions that sought to obtain socio-demographic data for compari son purposes. The first draft of the questionnaire for this study comprised 99 items. A sample of the first draft of the questionnaire for this study can be found in Append ix E. Survey Modifications. The YDAS instrument was designed for both paper an d computer application. The risk perceptions survey d eveloped by Rundmo and Iversen (2004), and risky driving behaviors adapted from Ul leberg and Rundmo (2002) were designed for mailing to participants. The questionn aire for knowledge of road laws adapted from Whiting et al., (1998) was designed fo r face-to-face completion. The final instrument was designed for in-class completion by study participants. A sample of the final draft of the questionnaire for this study com prising 90 questions can be found in Appendix F. The process of modification of the firs t draft of the questionnaire is described in detail in the pilot testing section of this chapter.

PAGE 71

59 Reliability of Scores from Instruments. This section discusses the reliability measures that indicate the reproducibility of the s urvey instruments’ data and their application in this study (Litwin, 1995). Reliabili ty was assessed by examining the internal consistency reliability of the domains tha t are used to measure the variables of interest. To determine internal consistency of the survey instruments and scales, the psychometric measure used was internal consistency reliability (Cronbach, 1951). This psychometric measure is applied to determine whethe r the various items are measuring the domain of interest. Reliability estimates were calculated for the quest ionnaire items by conducting internal consistency and test-retest reliability (M alfetti et al., 1989). The calculated values for internal consistency were estimated and Cronbac h’s for speeding was .88, .89 for riding with an unsafe driver, .89 for drinking and driving, and .70 for concern for others. The test-retest value for speeding was .82, .78 for riding with an unsafe driver, .75 for drinking and driving, and .76 for concern for other s. Ulleberg and Rundmo (2002) estimated both Cronbach’s and Loevinger’s H for each dimension of interest. Loevinger’s H determines conformity of a group of items to Mokken ’s criteria and validates their use as a scale of a unidimensional latent variable The obtained values for speeding were Cronbach’s = .84 Loevinger’s H = .56; for unsafe driving, Cronbach’s = .63 Loevinger’s H = .41; for riding with an unsafe driver Cronbach’s = .84 Loevinger’s H = .48; for drinking and driving, Cronbach’s = .76 Loevinger’s H = .58; and finally, for concern for others, Cronbach’s = .62 Loevinger’s H = .40. Internal consistency reliability was calculated for the questionnaire items testing risk perception by using Cronbach’s and Loevinger’s H values (Rundmo & Iversen,

PAGE 72

60 2004). The calculated values for the domains measur ing risk perceptions were as follows: emotion-based perception, Cronbach’s = .89 Loevinger’s H = .71; cognitive-based perception, Cronbach’s = .67 Loevinger’s H = .54; and concern, Cronbach’s = .81 Loevinger’s H = .70 (Rundmo & Iversen, 2004). Reliability of Scores from Pilots and Dissertation. This section discusses the reliability of scores that indicate the reproducibi lity of the survey instrument’s data (Litwin, 1995). Reliability scores were obtained fr om the pilot testing of the questionnaire through a test-retest procedure. The reliability of scores obtained from the main study’s data was also calculated. The internal consistency reliability estimates for the different scale scores of the questionnaire wer e examined with data obtained from the pilot study and the main study. To determine intern al consistency of the scores from the survey instruments and scales, Cronbach’s was used. Internal consistency reliability scores were measured for the questionnaire items te sting the constructs of the main study ( Risk-Taking Attitude, Risk Perception, Knowledge of Road Laws and Signs and Risky Driving Behavior ) by using Cronbach’s and item-to-total correlations values. Tables 3.1 through 3.4 provide the Cronbach’s and item-to-total correlations values obtained from the pilot testing and the main study. The Cronbach’s values, for the construct Risky Driving Behavior ranged from .583 to .791 for the main study, .556 to .792 for P ilot 1, and .703 to .826 for Pilot 2 as shown in Table 3.1. The Cronbach’s values, for the construct Risk-Taking Attitude ranged from .234 to .613 for the main study, .328 t o .697 for Pilot 1, and .262 to .642 for Pilot 2 as shown in Table 3.2. The Cronbach’s values, for the construct Risk Perception ranged from .550 to .720 for the main study, .517 to .712 for Pilot 1 and .524

PAGE 73

61 to .805 for Pilot 2 as shown in Table 3.3. The Cron bach’s values, for the construct Knowledge of Road Laws and Signs ranged from .413 to .629 for the main study, .430 to .467 for Pilot 1, and .388 to .592 for Pilot 2 as s hown in Table 3.4. The Item-to-Total correlations values obtained from the pilot testing and the main study for the four constructs showed similar ranges that were acceptab le for both the pilot testing and the main study. The Cronbach’s and item-to-total correlations values obtained fro m the pilot testing and the main study are not compared with va lues obtained from previous studies. The number and types of items used in the pilot tes ting and the main study are different from those of the original scales, and therefore, n o comparison is possible.

PAGE 74

62 Table 3.1 Reliability Scores for the Construct: Risky Driving Behavior Scale Name Items Pilot1 ( n = 47) Pilot 2 ( n = 32) Main Study ( n = 532) Cronbach Item-to-Total Correlations Cronbach Item-to-Total Correlations Cronbach Item-to-Total Correlations Speeding 5 .792 .428 to .691 .777 .326 to .679 .672 .142 to .579 Distracted Driving 6 .773 .267 to .725 .826 .374 to .791 .791 .391 to .677 Aggressive Driving 9 .722 .240 to .557 .786 .364 to .649 .678 .263 to .461 Not Adhering to Traffic Laws 7 .556 .031 to .437 .703 .220 to .718 .583 .190 to .410 Table 3.2 Reliability Scores for the Construct: Risk-Taking A ttitudes Scale Name Items Pilot1 ( n = 47) Pilot 2 ( n = 32) Main Study ( n = 532) Cronbach Item-to-Total Correlations Cronbach Item-to-Total Correlations Cronbach Item-to-Total Correlations Riding with an Unsafe Driver 4 .328 .021 to .337 .642 .348 to .461 .389 .182 to .261 Speeding 3 .697 .424 to .629 .554 .140 to .511 .613 .375 to .454 Concern for Others 4 .422 .065 to .385 .262 -.032 t o .327 .234 .046 to .148 Drinking and Driving 5 .651 .016 to .634 .476 .013 to .459 .513 .129 to .493

PAGE 75

63 Table 3.3 Reliability Scores for the Construct: Risk Percepti ons Scale Name Items Pilot1 ( n = 47) Pilot 2 ( n = 32) Main Study ( n = 532) Cronbach Item-to-Total Correlations Cronbach Item-to-Total Correlations Cronbach Item-to-Total Correlations Cognition-based Perception 3 .517 .000 to .647 .52 4 .080 to .570 .550 .118 to .598 Concern Perception 3 .524 .000 to .628 .748 .367 to .727 .720 .304 to .676 Emotion-based Perception 3 .712 .354 to .748 .805 400 to .823 .711 .322 to .647 Table 3.4 Reliability Scores for the Construct: Knowledge of Road Laws and Signs Scale Name Items Pilot1 ( n = 47) Pilot 2 ( n = 32) Main Study ( n = 532) Cronbach Item-to-Total Correlations Cronbach Item-to-Total Correlations Cronbach Item-to-Total Correlations Knowledge of Road Laws 9 .467 .000 to .390 .592 .09 9 to .481 .629 .173 to .452 Knowledge of Road Signs 4 .430 .104 to .396 .388 .0 59 to .372 .413 .182 to .332

PAGE 76

64 Confirmatory Factor Analysis of Driving Behavior Su rvey. This section discusses the confirmatory factor analysis [CFA] conducted on the variables of the indicators. The items were pooled to form the indicators that were used for the constructs of interest namely, Risk-Taking Attitude, Risk Perception, Know ledge of Road Laws and Signs, and Risky Driving Behavior. The construct Risky Driving Behavior was comprised of the following indicators Speeding, Distracted Driving, Aggressive Driving, and Not Adhering to traffic laws. The construct Risk-Taking Attitude was comprised of the following indicators Riding with an unsafe driver, Speeding, Concern for others, and Drinking and driving. The construct Risk Perception was comprised of the following indicators Cognition-based perception, Concern perc eption, Emotion-based perception. The construct Knowledge of Road Laws and Signs was comprised of the following indicators Knowledge of Road Laws and Knowledge of Road Signs (See Appendix I) for a complete listing of items comprising each indicat or. A summary of the factor loading ranges are shown in Table 3.5. The loadings were significantly different from zero. The standar dized factor loadings indicate that a considerable amount of unexplained variance. The fi nal set of items was selected after careful analysis of the original items. Some items were removed from the list due to theoretical redundancy and statistical significance

PAGE 77

65 Table 3.5 A Summary of Confirmatory Factor Analysis Standardi zed Factor Loadings Construct Indicator Items Standardized Factor Loading Speeding A2-A5 .19 to .65 Distracted Driving A6-A11 .41 to .78 Aggressive Driving A12-A14, A16, A18A20 .32 to .60 Risky Driving Behaviors Not Adhering to Traffic laws A20-A27 .18 to .68 Riding with an Unsafe Driver C1, C6, C16 .34 to .60 Speeding C2, C8, C12 .48 to .65 Concern for Others C3, C5, C11, C13 .10 to .41 Risk-Taking Attitudes Drinking and Driving C4, C7, C10, C14, C15 .18 to .69 Cognition-based Perception D4, D8, D12 .34 to .74 Concern perception D3, D7, D11 .14 to .77 Risk Perception Emotion-based Perception D1, D9, D10 .17 to .93 Continued next page

PAGE 78

66 Table 5 (continued) Knowledge of Road Laws K1-K6, K8-K10 .35 to .85 Knowledge of Road Laws and Signs Knowledge of Road Signs K11-K14 .34 to .64 Crash Experience Crash Experience Cr1-Cr4 .35 to .6 9 Validity of Scores from Instruments. The respective validities of the instruments used are examined in this section to determine the extent in which the items used are measuring their intended domain (Litwin, 1995). Val idity measures are important in establishing the appropriateness of the survey inst ruments used in this study. Malfetti et al. (1989) compared the five dimensions of interest with Mann Inventory subscales as an instrument of recognized validity. The values obta ined suggested that concurrent validity of the two measures was established. Similarly, to determine discriminant validity, Ulleberg and Rundmo (2002) examined the intercorrel ations between subscales and also found them to be satisfactory. The values obtained from the various studies on rel iability and validity suggest that the instruments selected for this study have a dequate psychometric properties along the dimensions of reliability and validity. A Flesc h-Kincaid grade level test was conducted to measure readability, coherence, and co mprehensiveness of the instruments. The result of the questionnaire Flesch-Kincaid grad e level test indicated a readability grade level of 5.3.

PAGE 79

67 Socio-demographic Variables Age. The students selected to participate in this study were selected from the general student pool limited to the age range of 18 -to-24 years. The age range provided a homogenous sample and assisted in providing a contr ol of research. Students were asked to provide their age as part of the questionnaire. Ethnicity. Participants were asked to select the ethnic group with which they identified, from an array of choices on the questio nnaire. The ethnic selections were taken from the 2000 population census (CSO, 2001). The 2000 population census is the most recent survey delineating the various ethnic g roups in Belize (CSO, 2001). The selections were limited to the following ethnic gro up categories: 1) Chinese 2) Creole 3) East Indian 4) Garifuna 5) Maya 6) Mennonite 7) Mes tizo/Spanish 8) Other (CSO, 2001). This information was used for descriptive purposes in this study. Sex. Students were asked to identify their sex as part of the questionnaire. The sex variable was measured as a dichotomous variable and was listed as either male or female. Enrolment status. Enrolment status was measured as a dichotomous var iable, fulltime or part-time. Full-time status is determined b y a minimum 12 credit hour enrolment in courses at the University of Belize. Crash Experience. Four items of the questionnaire requested informat ion on the respondents’ crash experience. The participants pr ovided information on whether they had been involved in a motor vehicle crash. The par ticipants also provided information on whether they had experienced injury as a result of MVC experience.

PAGE 80

68 Data Collection Procedure. The survey was completed by a non-random, convenie nce sample of UB students enrolled in courses in Associate and Ba chelor degree programs. The selected courses were offered by the faculties of Education and Arts, Management and Social Science, Nursing and Allied Health, and Science and Technology at the Belize City and Belmopan City campuses. The courses and the number of sections that were surveyed are listed in Table 6. These courses were selected in c onsultation with the Registrar of the University of Belize. The selection of courses was based on the premise that the selected courses would have the largest number of students w ithin the study’s age range. An estimated 1,265 students were expected to be enroll ed in these courses. However, the survey was conducted during an extended drop/add pe riod at the beginning of the second semester of the 2006-2007 academic year. This exten ded drop/add period probably impacted the actual class enrolment. Therefore, the actual number of students present when the questionnaires were distributed in the sel ected courses totaled 775.

PAGE 81

69 Table 3.6 List of Courses Surveyed on Both Campus Belmopan City Campus Belize City Campus Course Code Number of sections Course Code Number of sections BIOL402 1 ACTG201 1 BIOL403 1 ACTG202 2 CMPS360 1 CMCN209 1 EDUC323 1 ENGL112 2 ENGL090 1 ENGL299 1 ENGL111 2 FNAN221 1 ENGL112 8 MATH101 1 ENGL299 3 MATH104 1 MATH121 3 MGMT202 1 MATH340 1 MGMT285 2 PHAR109 1 MGMT304 1 MGMT373 2 TOUR233 1 TOUR321 1 Teachers for the selected courses and sections rece ived a letter requesting their permission to conduct the survey and an information al sheet describing the study and the questionnaire Appendix F. The teachers agreed to ap portion 45 minutes of their class for the researcher to administer the questionnaire. The questionnaire was administered to the

PAGE 82

70 students at the beginning of the selected class. Th e researcher introduced himself to the class. Before the questionnaire was distributed, th e researcher read the instructions found on the second page of the questionnaire. The resear cher collected the completed questionnaire as soon as the participants filled th em out. The length of time taken to complete the administration of the questionnaire to ok less than the requested 45 minutes that was estimated during the pilot phase of this s tudy. Participation in the study was voluntary and no identifying data were collected. T he number of students who received the questionnaires in the identified courses totale d 775, of which, only two refused to participate in the study. The students who refused to participate did not provide any reason for their non-participation. Of the 773 comp leted questionnaires, only 532 were within the study’s age range. A total of 532 comple ted questionnaires were collected, more than the recommended 260 minimum sample. The r esponse rate was more than the estimated 50%. The administration of the questionna ire was conducted during the second semester of the academic year 2006–2007. The first course was surveyed on January 16 and the last course was surveyed was on February 13 2007. Pilot Study Purpose and Components. Prior to carrying out the main study, a pilot stud y was conducted to identify problems with the questionnai re content and design, readability, administration process, data entry procedure, and d ata analysis strategies (Heppner & Heppner, 2004; McDermott & Sarvela, 2001). The pilo t study was also conducted to estimate the time it would take for students to com plete the questionnaire. The pilot study consisted of three components comprising an externa l panel review by professionals experienced in research, mini-pilot test and a fiel d test with a target sample consisting of

PAGE 83

71 students from Sacred Heart Junior College (SHJC) (H eppner & Heppner, 2004; McDermott & Sarvela, 2001). The entire pilot study was conducted between June and October 2006. External Panel Review. The external panel comprised professionals with res earch experience in injury prevention, young adults’ heal th risk behavior, road safety and survey design (See Appendix G). The individuals ide ntified as potential members of the external review panel were contacted via email. Of a total of 10 professionals contacted, five agreed to participate in the external panel. A review guide and a research question table were sent to the external panel along with fi rst draft of the questionnaire (See Appendix H for the external panel review guide, App endix I for the Research question table and Appendix E for the first draft of the que stionnaire). Included in the review guide, the researchers were asked to make suggestio ns, comments and recommendations on the questionnaire and to submit additional surve y questions. The recommended changes were incorporated into second draft of the questionnaire. The second draft of the questionnaire was resubmitt ed to the external panel for further examination. The expert panel that reviewed the first draft of the questionnaire agreed to review the second draft. The expert panel submitted further comments and suggested revisions on the questionnaire. These com ments and suggestions were incorporated into the third draft of the questionna ire. Communications with the expert panel were carried out via email throughout the rev iew process.

PAGE 84

72 Mini-pilot Test. After the completion of the expert panel review, a mini-pilot test was conducted with a convenience sample of seven st udents enrolled at Sacred Heart Junior College [SHJC]. SHJC is a feeder institution for the University of Belize. Students from SHJC enroll into Associate and Bachelor’s degr ee programs at UB and are expected to bear similarities with the UB student population The seven student volunteers were given a review guide to provide comments on the que stionnaire (See Appendix J for the Pilot Test review guide). The students were asked t o comment on whether the directions were concise and clearly understood, whether the qu estions and responses were appropriate, and whether the format of the question naire was easy to follow. In addition, they were asked to provide other comments, if warra nted. The results of the mini-pilot test were incorporated into the final draft of the questionnaire (See Appendix F for the final draft of the questionnaire). In addition to t he review guide, the students were asked to complete the questionnaire to estimate the amoun t of time it would take to fill it out in class. Field Testing. With the expert panel review and the mini-pilot te st completed, the final component of the pilot study, the field test, was conducted to estimate the time it would take for students to complete the questionnai re, to identify possible implementation challenges, to assess data entry and data coding strategies, and to conduct preliminary analysis in preparation for the full da ta collection process. SHJC students were expected to have similar characteristics as st udents enrolled at UB. The researcher requested and received permission from the lecturer of two English subject courses to facilitate participation of students in the field t esting of the questionnaire.

PAGE 85

73 The field test was completed in two phases using a convenience sample of students enrolled in two English subject courses at SHJC. These two phases were completed by students enrolled at SHJC. Two English subject courses were selected to participate in the field test. The courses were sel ected with the assistance of the Dean of Sacred Heart Junior College. Forty-seven students participated in the first phase and 32 students participated in the second phase. The decr ease in the number of students participating in the second phase was due to absenc es rather than refusal to participate. The students participating in the two phases of the field study were within the age range of 18 to 24 years. The students participating in th e field testing were not enrolled or form part of the student body of UB. Preliminary analysis on the data obtained through t he field testing was conducted to examine internal consistency reliability using C ronbach’s coefficients. Basic univariate analysis was performed on the field test data to determine mean, standard deviation, skewness and kurtosis. The results of th ese analyses obtained from both phases were compared. Preliminary analysis was conducted u sing the SPSS 15.0 for Windows software program. Data Analysis Data Entry. Data were entered into an electronic database entr y form. The software Microsoft Office Word 2003 was used to cre ate the database form. The database file was imported into a statistical analysis softw are program, SPSS 15.0 for Windows. The SPSS 15.0 for Windows software program was used to conduct univariate, and bivariate. Multivariate analysis was conducted usin g Muthn and Muthn M plus

PAGE 86

74 (version 4.21) statistical software program. The th ree types of analysis are described in more detail below. Univariate and Bivariate Analysis. Univariate analysis consisted of frequency distributions and the construction of frequency tab les for all descriptive data, including demographic information. Descriptive statistical an alysis reported, where appropriate, basic statistics on mean, kurtosis, data distributi on, standard deviations, frequencies, variance, missing values, and normality/skewedness. This analysis provided basic information to support the bivariate and multivaria te analysis. Univariate analysis was done on the independent variables (risk-taking atti tudes, risk perceptions, and knowledge) and the dependent variable (driving behaviors). Bivariate statistical analysis was done to identify focal relationships between the independent variables and the dependent variables a nd included chi square analysis and Pearson’s product-moment correlations. Multivariate Analysis. The analytic approach most appropriate for latent (unobserved) independent and dependent variables is Structural Equation Modeling [SEM] also known as path analysis with latent varia bles, structural equation analysis, covariance structure models, path modeling and late nt variable analysis of structural equations (Hatcher, 1994; Maruyama, 1998). This stu dy used constructs that are measured indirectly by multiple indicator variables (Hatcher, 1994; Maruyama, 1998). The latent constructs were used to develop the meas urement model and then to develop the structural model seen in Figure 2. The SEM analytic method is well-suited for this stu dy as the variables used are latent variables that are measured indirectly throu gh two or more indicators. SEM was

PAGE 87

75 used to test the relationship between and among the variables. The relationships of the variables and their respective dimensions are illus trated in figure 2. SEM was applied to examine the relationship among the independent vari ables (risk-taking attitudes, risk perceptions, and knowledge) and the dependent varia ble (driving behaviors), and explained the variance of the variables as presente d by the model (Hatcher, 1994; Maruyama, 1998), as well as to provide the basis fo r understanding the relevance of these relationships. SEM allowed for the determination of the effect of each independent variable separately. To answer the research questions in this study, the following analyses were conducted. Question 1: To what extent are the specific stages of the Precaution Adoption Process Model [PAPM] observed in the study populati on?. The questions B1 through B8, found in Section B of the questionnaire, were used to develop an algorithm to place the responses into the PAPM stages. The algorithm used to place the participants in the various PAPM stages is shown in Figure 3.

PAGE 88

76

PAGE 89

77 Question 2: What is the relationship between selec ted demographic factors and risky driving behavior? The demographic factors of interest were collecte d in section F of the questionnaire. The questions F1 to F8 were u sed to collect data on general demographic factors. The participant’s Crash Experience was assessed using questions F9 to F13. Cronbach’s and item-to-total correlations values were obtaine d for the section addressing participant’s crash experience. Confirmatory Factor Analysis [CFA] was conducted on this indicator. CFA was also condu cted on the four indicators forming the Risky Driving Behavior construct. The four indicators for the Risky Driving Behavior construct were Speeding, Distracted Driving, Aggres sive Driving and Not Adhering to Traffic Laws. The questions A1 to A27 were used to collect data for the four indicators of the Risky Driving Behavior construct. The CFA was part of the recommended two-step proces s in SEM analysis to determine its relationship with risky driving behav ior (Buhi, 2007; Hatcher, 1994; Hoyle, 1995; Maruyama, 1998). The results from the CFA ana lysis were used to build the measurement model that examined the relationship be tween Crash Experience and Risky Driving Behavior Question 3: What is the relationship between young adults’ risk-taking attitudes and risky driving behaviors?. The questions C1 to C16 of the questionnaire were used to collect data for the construct, Risk-Taking Attitude Cronbach’s and item-to-total correlations values were obtained for the section a ddressing Risk-Taking Attitude The four indicators for the Risk-Taking Attitude construct were Riding with an Unsafe Driver, Speeding, Concern for Others and Drinking and Drivi ng. CFA was also conducted on

PAGE 90

78 these indicators as well as on the indicators of th e Risky Driving Behavior construct. The results from the CFA analysis were used to build th e measurement model that examined the relationship between Risk-Taking Attitude and Risky Driving Behavior Question 4: What is the relationship between young adults’ knowledge of road laws and signs and risky driving behaviors? The questions E1 to E14 of the questionnaire were used to collect data for the con struct, Knowledge of Road Laws and Signs Cronbach’s and item-to-total correlations values were obtaine d for the section addressing Knowledge of Road Laws and Signs The two indicators for the Knowledge of Road Laws and Signs construct were Knowledge of Road Laws and Knowledge of Road Signs. CFA was also conducted on these indicators a s well as on the indicators of the Risky Driving Behavior construct. The results from the CFA analysis were used to build the measurement model that examined the relationshi p between Knowledge of Road Laws and Signs and Risky Driving Behavior Question 5: What is the relationship between young adults’ risk perceptions and risky driving behaviors?. The questions D1 to D12 of the questionnaire were u sed to collect data for the construct, Risk Perception Cronbach’s and item-to-total correlations values were obtained for the section a ddressing Risk Perception The three indicators for the Risk Perception construct were Cognition-based Perception, Concernbased perception and Emotion-based Perception. CFA was also conducted on these indicators as well as on the indicators of the Risky Driving Behavior construct. The results from the CFA analysis were used to build th e measurement model that examined the relationship between Risk Perception and Risky Driving Behavior

PAGE 91

79 Chapter 4: Results Introduction This section provides the results obtained from the pilot study, comprising an external panel review, mini-pilot test and field te st, and the main study’s survey results. The main study’s survey results will be used to add ress the proposed five questions: (a) To what extent are the specific stages of the PAPM observed in the study population?, (b) What is the relationship between selected demograph ic factors and risky driving behavior?, (c) What is the relationship between you ng adults’ risk-taking attitudes and risky driving behaviors?, (d) What is the relations hip between young adults’ knowledge of road laws and signs and risky driving behaviors? and (e) What is the relationship between young adults’ risk perceptions and risky dr iving behaviors? The answers to these five questions will increase the understanding of t he factors associated with the risks of MVC-related mortality and morbidity for young adult s in Belize. Pilot Study Results External Panel. Of the ten professionals contacted, six agreed to participate in the external review panel. Ultimately, five persons par ticipated in reviewing the Driving Behavior Survey in the two-phase process. The five professionals, who did not participate, declined due to time constraints and w orkload. In the initial phase, the five member external panel received an electronic copy o f the questionnaire titled “ Attitudinal Factors Related to Driving Behaviors of Young Adult s in Belize: An Application of the

PAGE 92

80 Precaution Adoption Process Model, ” an evaluation guide and a research question table The external panel reviewed the questionnaire using the evaluation guide that sought to obtain contribution on the survey questions that we re related to clarity, age appropriateness, suitable response options and form at. The external panel was also asked to recommend additional items. The external panel was satisfied with the clarity, age appropriateness of the survey instrument, ease of navigation and with the survey format. However, the panel made five groups of recommendations in this initial review. A change in the title of the survey was recommended and the titled was changed t o “ Driving Behavior Survey. ” The second major change centered on the instruction sec tion, which was reworded to address certain research protocols. Thirdly, sections head ing of the questionnaire were reworded to reduce the potential of response bias. Fourthly, additional survey items were recommended along with changes to some of the initi al items. Finally, the external panel recommended changes in the response options for two of the sections. The recommended changes were made and the updated questionnaire was sent back to the panel for another review. The five persons, who made up the initial review pa nel, agreed to participate in the second external review panel. An evaluation gui de was sent along with the updated questionnaire. The second external review panel rec ommended minor editorial changes that mainly focused on the formatting of the survey items options. The recommendations were accepted and the changes were made to the ques tionnaire.

PAGE 93

81 Mini-pilot Test at Sacred Heart Junior College. After completing the two phases of the external panel review, a mini-pilot test was conducted at the Sacred Heart Junior College [SHJC]. Seven students participated in thi s study. The students were 18-to 24years old who were enrolled fulltime at SHJC. The students were provided with an evaluation guide and were asked to use the guide to review the questionnaire to determine whether the questions were clear, age appropriate, and whether the response options and format were easy to follow. The students completed the review and provided mino r editorial changes to the questionnaire. Overall, their review indicated that the questionnaire format was easy to navigate, age appropriate and the questions and res ponses were understood. The editorial changes recommended by the students were made to th e final draft of the questionnaire. The students were asked to fill out the questionnai re after they had completed the review. This procedure provided an estimate of the time nee ded to complete the questionnaire. Forty minutes were needed to complete the questionn aire. Field Test at Sacred Heart Junior College. The field test portion of the pilot study consisted of two parts; Pilot1 (N=47) and Pilot2 (N =32). The results of the field test were used to examine internal consistency reliability us ing Cronbach coefficients and itemto-total correlations (See Tables 3.1-3.4 p.63-64). Basic univariate analysis was also performed on the field test data to determine mean, standard deviation, skewness and kurtosis. The results of these analyses are found i n tables 4.1-4.4. The skewness and kurtosis values for the construct Risky Driving Behavior ranged from -0.50 to 1.12 and -1.02 to 1.63 for Pilot 1, a nd -0.25 to 1.26 and -0.76 to 2.26 for Pilot 2 (Table 4.1). The skewness and kurtosis valu es for the construct Risk-Taking

PAGE 94

82 Attitude ranged from 0.07 to 0.24 and -0.79 to 0.13 for Pil ot 1, and -0.36 to 1.22 and 0.92 to 1.35 for Pilot 2 (Table 4.2). The skewness and kurtosis values for the construct Risk Perception ranged from -0.32 to 0.86 and -0.27 to 0.61 for Pi lot 1, and 0.00 to 1.14 and -1.05 to 2.05 for Pilot 2 (Table 4.3). The skew ness and kurtosis values for the construct Knowledge of Road Laws and Signs ranged from -0.67 to 0.24 and -0.56 to 0.41 for Pilot 1, and -0.92 to 0.23 and -0.54 to 2.00 fo r Pilot 2 (Table 4.4). The mean and standard deviation values obtained from the field t esting for the four constructs showed similar ranges that were acceptable for both phases of the field testing.

PAGE 95

83 Table 4.1 Mean, Standard Deviation, Skewness and Kurtosis Val ues for the Construct: Risky Driving Behavior Scale Name Items Pilot1 (n = 47) Pilot 2 (n = 32) Risky Driving Behaviors M SD Skewness Kurtosis M SD Skewness Kurtosis Speedinga 5 2.32 0.93 0.07 -1.02 2.27 0.80 0.22 -0.69 Distracted Drivinga 6 2.37 0.72 0.21 -0.43 2.46 0.78 0.05 -0.76 Aggressive Drivinga 9 1.78 0.63 1.12 1.63 1.78 0.69 1.26 2.26 Not Adhering to Traffic Lawsa 7 2.77 0.57 -0.50 1.02 2.71 0.64 -0.25 0.34 Note. Judgements were made on a 5-point scale (1 = almost never, 5 = almost always). Table 4.2 Mean, Standard Deviation, Skewness and Kurtosis Val ues for the Construct: Risky Driving Behavior Scale Name Items Pilot1 (n = 47) Pilot 2 (n = 32) Risk-taking Attitudes M SD Skewness Kurtosis M SD Skewness Kurtosis Riding with an Unsafe Drivera 4 1.71 0.49 0.23 -0.46 1.70 0.65 1.22 1.35 Speedinga 3 2.96 0.67 0.07 -0.79 2.93 0.58 0.62 -0.04 Concern for Othersa 4 1.95 0.47 0.07 0.13 1.90 0.39 -0.36 -0.47 Drinking and Drivinga 5 2.00 0.63 0.24 -0.54 2.03 0.58 0.04 -0.92 Note. Judgements were made on a 4-point scale (1 = strongly agree, 4 = strongly disagree).

PAGE 96

84 Table 4.3 Mean, Standard Deviation, Skewness and Kurtosis Val ues for the Construct: Risk Perceptions Scale Name Items Pilot1 (n = 47) Pilot 2 (n = 32) Risk Perceptions M SD Skewness Kurtosis M SD Skewness Kurtosis Cognition-based Perceptiona 3 1.82 0.57 0.86 0.61 1.67 0.49 1.14 2.05 Concern Perceptiona 3 1.73 0.49 0.46 -0.27 1.77 0.49 0.09 0.02 Emotion-based Perceptiona 3 2.44 0.65 0.32 0.21 2.53 0.63 0.00 -1.05 Note. Judgements were made on a 4-point scale (1 = strongly agree, 4 = strongly disagree). Table 4.4 Mean, Standard Deviation, Skewness and Kurtosis Val ues for the Construct: Knowledge of Road Laws and S igns Scale Name Items Pilot1 (n = 47) Pilot 2 (n = 32) Knowledge of Road Laws and Signs M SD Skewness Kurtosis M SD Skewness Kurtosis Knowledge of Road Laws 9 57.45a 22.02 -0.67 0.41 65.28a 20.50 -0.92 2.00 Knowledge of Road Signs 4 42.45a 25.03 0.24 -0.56 43.75a 27.68 0.23 -0.54 Note. a Means are out of a total of 100%.

PAGE 97

85 Main Study Results Survey Summary. Data were collected for this study from a convenie nce sample of students enrolled at the University of Belize [U B] during the second semester of the academic year 2006-2007. A total of 775 questionnai res were distributed of which 773 were completed. Of the 773 questionnaires collected 532 questionnaires were collected from students within the study’s age range of 18-to -24-years. The final sample of 532 represented 42% of the 18-to-24-year-olds and 22% o f the total number of students enrolled at UB during the second semester of the ac ademic year 2006-2007. The survey was carried out at UB’s main campus in Belmopan Cit y and its satellite campus in Belize City. Univariate Analyses Results Population Demographics. Univariate analyses were conducted on section F of the questionnaire. Section F questions collected de mographic data on the participants who filled out the questionnaire. The age range for data collection for this study was from 18 to 24 years with 78.0% (N=415) of the respondent s being 21 years or younger. Participants in this study were mostly female 58.1% (N=309). The participants’ ethnic backgrounds were described as mainly Mestizo (42.6% N=225) or Creole (33.3%, N= 176). Participants’ main places of residence were t he Cayo (36.5%, N=193) and Belize (34.4%, N=182) districts. Of the total sample, 57% (N=303) were from the UB’s Belmopan Campus. Participants were mostly enrolled as fulltime students 88.2% (N=469). Table 4.5 provides complete demographic c haracteristics of the participants.

PAGE 98

86 Table 4.5 Participants’ Demographic Characteristics Variable Frequency Percent Gender (N=532) Female Male 309 223 58.1 41.9 Age (N=532) 18 19 20 21 22 23 24 121 118 97 79 48 36 33 22.7 22.2 18.2 14.8 9.0 6.3 6.2 Ethnicity (N=528) Mestizo/Spanish Creole Garifuna East Indian Chinese Maya Mennonite Other 225 176 29 20 15 15 5 43 42.6 33.3 5.5 3.8 2.8 2.8 0.9 8.1 District of Residence (N=529) Cayo Belize Orange Walk Corozal Stann Creek Toledo 193 182 72 38 24 20 36.5 34.4 13.6 7.2 4.5 3.8 Campus (N=532) Belmopan City Belize City 303 229 57.0 43.0 Enrolment Status (N=532) Full-time Part-time 469 63 88.2 11.8

PAGE 99

87 Driver and Crash Experience. A total of 72.6% (N=380) of those surveyed described themselves as drivers. Of those who descr ibed themselves as drivers, only 47.9% (N=252) of them had a valid driver’s license. Eighty-two percent of the participants had driving experience of 4 years or l ess. Participants described their driving skills as being either Mostly Good (32.9%, N=147) o r Very Good (34.0%, N=152). In the past 12 months, participants reported that they had neither been in a car crash as a driver (89.5%, N=459) nor as a passenger (86.9%, N=456). T he majority of the participants had not experienced any injuries from car crashes (98.1 %, N=513) nor had they been in a car crash in which someone else was injured (95.6%, N=5 00). Table 4.6 provides more details of the participants’ driving and crash expe rience.

PAGE 100

88 Table 4.6 Participants’ Driving and Crash Characteristics Variable Frequency Percent Drive (N=531) Yes No 380 151 72.6 28.4 Have a valid driver’s license (N=526) Yes No 252 274 47.9 52.1 Rate your driving skills (N=447) Fair Good Very Good Excellent 86 147 152 62 19.2 32.9 34.0 13.9 Years of Driving (N=464) 0 1 2 3 4 5 6 7 8 88 83 82 80 48 35 15 15 18 19.0 17.9 17.7 17.2 10.3 7.5 3.2 3.2 3.9 Experience Car Crash as Driver (N=513) Yes No 54 459 10.5 89.5 Experience Car Crash as Passenger (N=525) Yes No 69 456 10.5 89.5 Been in a Car Crash where experience injury to self occurred (N=523) Yes No 10 513 1.9 98.1 Been in a Car Crash where experience injury to othe rs occurred (N=523) Yes No 23 500 4.4 95.6

PAGE 101

89 Precaution Adoption Process Model Staging Variables Just 3.0% (N=16) of those surveyed did not travel by car in the past 30 days with the rest either riding in a car or driving one. Participants had heard about a youn g person being hurt in a motor vehicle crash [MVC] (87.7%, N=465). In the past 12 months, participants reported that they had heard a message on the importance of using seatbelt s to prevent injury as a result of an MVC (89.5%, N=459). Participants reported that they seldom wore a seatbelt when riding in car driven by someone (50.6%, N=268). Pa rticipants reported that they planned to wear a seatbelt the next time they drove a car ( 57.3%, N=297). Table 4.7 provides more details of the participants’ responses to the questions used in the survey section that comprises the Precaution Adoption Process Model Sta ging Variables. Table 4.7 PAPM Staging Questions Frequency Distribution Value s Variable Frequency Percent Travel by Car in Past 30 Days (N=530) Yes No 514 16 97.0 3.0 Have heard about a young person hurt in MVC (N=530) Yes No 465 65 87.7 12.2 Have heard message of seatbelt importance (N=447) Yes No 461 69 87.0 13.0 Reported Seatbelt Use (N=529) Never Seldom Always 87 268 174 16.4 50.6 32.9 Plan to use Seatbelt (N=518) Yes No Don’t Know 297 73 148 57.3 14.1 28.6

PAGE 102

90 Normality. This section provides the statistics used to asses s normality of the data collected on the constructs of interest. Means, sta ndard deviation, skewness and kurtosis values were used to assess distribution of data. The means for the construct Risky Driving Behavior ranged from 1.70 to 2.56 (SD 0.55 to 0.81); for the construct Risk-Taking Attitude from 1.76 to 3.00 (SD 0.41 to 0.59); for the construct Risk Perception from 1.80 to 2.48 (SD 0.52 to 0.56); and for the construct Knowledge of Road Laws and Signs from 56.68 to 66.08 (SD 21.32 to 26.79). The Skewness values, for the construct Risky Driving Behavior ranged from -0.47 to 0.80; for the construct Risk-Taking Attitude from -0.27 to 0.41; for the construct Risk Perception from 0.09 to 0.86; and for the construct Knowledge of Road Laws and Signs from -0.74 to -0.08. The Kurtosis values, for the c onstruct Risky Driving Behavior ranged from -0.43 to 1.11; for the construct Risk-Taking Attitude from -0.64 to 0.77; for the construct Risk Perception from 0.47 to 1.09; and for the construct Knowledge of Road Laws and Signs from -0.73 to 0.34. Table 4.8 provides more detai ls on the normality values of the Driving Behavior Survey sca les.

PAGE 103

91 Table 4.8 Normality Values of Driving Behavior Survey Scales Scale Name Variables Number of Items M SD Skewness Kurtosis Risky Driving Behaviors Speedinga 5 2.20 0.81 0.26 -0.33 Distracted Drivinga 6 2.36 0.78 0.16 -0.43 Aggressive Drivinga 9 1.70 0.55 0.80 0.24 Not adhering to traffic lawsa 7 2.56 0.58 -0.47 1.11 Risk-taking Attitudes Riding with an Unsafe Driverb 4 1.76 0.53 0.41 -0.40 Speedingb 3 3.00 0.59 -0.27 -0.39 Concern for Othersb 4 1.91 0.41 0.27 0.77 Drinking and Drivingb 5 1.88 0.54 0.10 -0.64 Risk Perception Cognitionbasedb 3 1.84 0.52 0.86 1.09 Concernbasedb 3 1.80 0.53 0.50 0.53 Emotionbasedb 3 2.48 0.56 0.09 0.47 Knowledge of Road Laws and Signs Knowledge of Road Laws 9 66.08 c 21.32 -0.74 0.34 Knowledge of Road Signs 4 56.68 c 26.79 -0.08 -0.73 Note. a Judgements were made on a 5-point scale (1 = almost never, 5 = almost always). b Judgements were made on a 4-point scale (1 = strongly agree, 4 = strongly disagree). c Means are out of a total of 100%.

PAGE 104

92 Bivariate Analyses Results Gender and Age Differences Bivariate analyses were conducted on the data determine the relationship between gender and the c onstructs under study and the relationship between age and the constructs under s tudy. One of the bivariate analyses assessed the effect size of the comparison between means of the responses of males and females to questions that form the constructs in th e Driving Behavior Survey. Effect size was determined by calculating Cohen’s d values. The following formula was used to calculate the Cohen’s d values: The construct Risky Driving Behavior comprising grouped variables measuring Speeding, Distracted Driving, Aggressive Driving an d Not adhering to Traffic Laws behaviors, had Cohen’s d values that ranged from 0.01 to 0.65. The construc t Risk-Taking Attitude consisted of group variables measuring attitudes t oward Riding with an Unsafe Driver, Speeding, Concern for Others and Drinking a nd Driving. The effects size for the Risk-Taking Attitude construct ranged from 0.01 to 0.54. The construct Risk Perception had three grouped variables measuring Cognition-bas ed, Concern and Emotion-based perceptions. The construct Risk Perception had Cohen’s d values ranging from 0.00 to 0.31. The construct Knowledge of Road Laws and Signs had two grouped variables, namely Knowledge of Road Laws and Knowledge of Road Signs with Cohen’s d values of 0.48 and 0.33, respectively. Six subscales produ ced medium to large effect sizes (0.50 to 0.80), Speeding Behavior (0.65), Knowledge of Ro ad Signs (0.60), Knowledge of Road Law (0.56), Riding with and Unsafe Driver Atti tude (0.54), Speeding Attitude M male – M female Pooled SD

PAGE 105

93 (0.52), and Drinking and Driving Attitude (0.52). F our subscales produced small to medium effect sizes (0.20 to 0.50), Aggressive Driv ing (0.37), Concern for Others attitude (0.35), Concern perception (0.31), and Agg ressive Driving behavior (0.26). The three remaining subscales had small effect sizes of less than two. Table 4.9 presents the adjusted means and pooled SD of the subscales, as well as, the Cohen’s d values for the constructs measuring Risky Driving Behavior Risk-Taking Attitude Risk Perception and Knowledge of Road Laws and Signs

PAGE 106

94 Table 4.9 Effect Size Values of Driving Behavior Survey Scale s Scale Name Subscales # of Items M (Males) SD M (Females) SD Cohen’sd Risky Driving Behaviors Speeding a 5 2.48 0.79 1.99 0.75 0.65 Distracted Driving a 6 2.37 0.72 2.36 0.82 0.01 Aggressive Driving a 9 1.74 0.55 1.55 0.47 0.37 Not adhering to Traffic Laws a 7 2.64 0.54 2.49 0.61 0.26 Risk-taking Attitudes Riding with an Unsafe Driver b 4 1.93 0.52 1.65 0.51 0.54 Speeding b 3 2.17 0.62 1.87 0.54 0.52 Concern for Others b 4 1.98 0.41 1.84 0.39 0.35 Drinking and Driving b 5 2.05 0.52 1.78 0.51 0.52 Risk Perception Cognitionbased b 3 2.42 0.57 2.42 0.57 0.01 Concernbased b 3 2.19 0.53 2.03 0.51 0.31 Emotionbased b 3 2.08 0.61 2.04 0.54 0.07 Knowledge of Road Laws and Signs Knowledge of Road Laws c 9 72.88 17.83 61.46 22.17 0.56 Knowledge of Road Signs c 4 65.68 23.74 50.17 27.07 0.60 Note. a Judgements were made on a 5-point scale (1 = almost never, 5 = almost always). b Judgements were made on a 4-point scale (1 = strongly agree, 4 = strongly disagree). c Means are out of a total of 100%.

PAGE 107

95 The second bivariate analyses assessed the relatio nship between age and the responses to scales that form the constructs in the Driving Behavior Survey. The statistical procedure used to examine the strength of this relationship was conducted. The Pearson correlation coefficient ( r ) served to measure the strength of the relationshi p between age and the scales that formed the construc ts in the Driving Behavior Survey. Both age and the scales of the Driving Behavior Sur vey were measured on either an intervalor ratio level of measurement (O’Rourke, Hatcher & Stepanski, 2005). Pearson correlation values can range from a -1.00 through 0 .00 to +1.00. Whereas a Pearson correlation value closer to -1.00 or +1.00 is indic ative of a strong relationship values closer to 0.00 are indicative of weaker relationshi ps. A value of 0.00 is indicative of no relationship between the variables. The scales measuring the construct Risky Driving Behavior had r values that ranged from -.006 (Speeding) to .016 (Aggressive Dr iving). None of these values were statistically significant. The scales measuring the construct Risk-Taking Attitude had r values that ranged from -.136 (Speeding) to .030 (R iding with an Unsafe Driver). Only the values for the Speeding attitude scale was stat istically significant at p < 0.01 (2tailed). The construct Risk Perception had r values ranging from -.065 to .020 with none of these values being statistically significant. Th e construct Knowledge of Road Laws and Signs had two grouped variables, namely Knowledge of Roa d Laws and Knowledge of Road Signs with r values ranging from .085 to .090 with the only val ue for Knowledge of Road Laws being statistically significant at p < 0.05 (2-tailed). Tables 4.10 to 4.13 provide more information on the correlation analyse s for the four constructs of interest and Age.

PAGE 108

96 Table 4.10 Correlation Matrix for Risky Driving Behavior Scale s and Age F1 Age Speeding Pearson Correlation Sig. (2-tailed) N -.006 .899 465 Distracted Driving .007 .881 465 Aggressive Driving .016 .731 464 Not Adhering to Laws .013 .782 463 Table 4.11 Correlation Matrix for Risk Taking Attitude Scales and Age F1 Age Riding with Unsafe Driver Pearson Correlation Sig. (2-tailed) N .030 .498 529 Speeding -.136** .002 503 Concern for Others -.083 .063 507 Drinking and Driving -.018 .676 524 ** Correlation is significant at the 0.01 level (2tailed)

PAGE 109

97 Table 4.12 Correlation Matrix for Risk Perception Scales and A ge F1 Age Cognition-based perception Pearson Correlation Sig. (2-tailed) N .020 .656 481 Concern-based perception -.065 .157 479 Emotion-based perception -.029 .506 511 Table 4.13 Correlation Matrix for Knowledge of Road Laws and S igns Scales and Age F1 Age Knowledge of Road Laws Pearson Correlation Sig. (2-tailed) N .090* .037 532 Knowledge of Signs .085 .050 532 Correlation is significant at the 0.05 level (2-t ailed)

PAGE 110

98 Research Question Results Analyses were conducted on the main study’s survey results to address the five research questions: (1) To what extent are the spec ific stages of the Precaution Adoption Process Model (PAPM) observed in the study populati on?, (2) What is the relationship between selected demographic factors and risky driv ing behavior?, (3) What is the relationship between young adults’ risk-taking atti tudes and risky driving?, (4) What is the relationship between young adults’ knowledge of road laws and signs and risky driving behaviors?, and (5) What is the relationshi p between young adults’ risk perceptions and risky driving behaviors? Frequency distributions for the items of the PAPM s taging algorithm were used to answer the first research question. The staging alg orithm placed the responses into the various PAPM stages to illustrate the distribution among the participants. For questions two through four, multivariate analyses was conduct ed. Research Question 1: To what extent are the specifi c stages of the Precaution Adoption Process Model (PAPM) observed in the study population?. To determine the proportion of young adults that were in each of the stages of the PAPM, an algorithm to place the responses into the PAPM stages was used. The algorithm used to assign the participants in the various PAPM stages can be foun d in Chapter 3. Frequency distributions demonstrated the proportion of respon dents in each of the PAPM stages as seen in Figure 4.

PAGE 111

99 Table 4.14 Proportions of Young Adults in Each Stage of Precau tion Adoption Process Model [PAPM] PAPM Stage (N=471) Frequency Percent Stage 1 Unaware of issue 65 13.8 Stage 2 Unengaged by issue 1 0.2 Stage 3 Undecided 5 1.1 Stage 4 Decided not to act 6 1.3 Stage 5 Decided to act 2 0.4 Stage 6 Acting 321 68.2 Stage 7 Maintenance 71 15.1

PAGE 112

100

PAGE 113

101 Multivariate Analyses: Structural Equation Modeling (SEM) Results The statistical procedure used was structural equat ion modeling (SEM). SEM was conducted to obtain results that would provide answ ers to these four questions. The SEM analytic method is appropriate when the variables u nder study are latent variables that are measured indirectly through two or more indicators. SEM was used to test the relationship between and among variables. The relat ionship among the independent variables ( Risk-Taking Attitude Risk Perception and Knowledge of Road Laws and Signs ) and the dependent variable ( Risky Driving Behavior ), was examined to help explain the variance of the variables as presented by the model under study. SEM was conducted using a two-step process (Hatcher, 1994; Maruyama, 1998). Step one used the confirmatory factor analysis (CFA) to develop the m easurement model (Hatcher, 1994; Maruyama, 1998). The first step investigated whethe r the factors for the constructs provided a measurement model with acceptable fit to the data collected. The results of the CFA have been provided in Chapter 3. The second ste p was to conduct the path analysis for the latent variables. This second step specifie s the relationship between and among the latent variables. The results of these analyses are presented in the following sections. To assess whether the structural model was acceptab le for further analysis, SEM testing of the full model with the data was conduct ed to obtain measures of overall model fit and to provide the structural relationships amo ng the latent constructs of Risky Driving Behavior, Risk-Taking Attitude Risk Perception and Knowledge of Road Laws and Signs The indices used to test model fit include the C hi-Square test of model fit (2), Comparative Fit Index (CFI), Tucker-Lewis Index (TL I), Root Mean Square Error of Approximation (RMSEA), Standardized Root Mean Squar e Residual (SRMR). The

PAGE 114

102 recommended cutoff value for the indices are CFI an d TLI 0.90, RMSEA 0.06 and SRMR 0.07 (Buhi, Goodson, & Neilands, 2007; Hu & Bentle r, 1999; Yu, 2002). The Chi-Square test of overall model fit provided value s of 2 (95, N=532) = 309.87, p<0.000 / 2/df = 3.26. The 2/df ratio falls within the 2 and 5 values commonly observed in health behavior (Buhi, Goodson, & Neilands, 2007). CFI for the structural model was 0.87; the TLI was 0.82; and the RMSEA was 0.07 p<0. 001 (90% confidence interval = 0.06–0.07) (Boomsma, 2000). Table 4.15 Structural Model Fit Indices provides the fit indices obtained from the SEM analysis. Figure 5 pr ovides the standardized path coefficients demonstrating direction and magnitude of the relationship among the latent constructs that were obtained from the final struct ural model. The entire Mplus output of the final structural model SEM analysis can be foun d on Appendix K. Globally, the CFI and TLI indices do not provide a good fit for interpretation of the model and are less than acceptable. However, fo r this study the model CFI and TLI indices obtained from the SEM analysis still provid e reasonable results for further interpretation. The indices suggest that the discre pancy between the theoretical model and the observed relations are acceptable and the model fits the data. As such, the model fit values suggest that the variables are reliably asso ciated in the context of the model and can be use to explain risky driving behaviors. Table 4.15 Structural Model Fit Indices Model 2 df p-value CFI TLI RMSEA SRMR Structural 309.87 95 0.000 0.87 0.82 0.07 0.06

PAGE 115

103

PAGE 116

104 Research Question 2: What is the relationship betwe en selected demographic factors and risky driving behavior?. To determine the relationship between demographic factors (being male, driver, crash experience, and age) and risky driving behaviors, SEM analysis was conducted on the data. As shown in Fig ure 5, a statistically significant direct effect between being a Driver and Crash Experience on Risky Driving Behaviors is observed ( = .177, p < .05 and .228, p < .01, respectively). The beta values for male and age on risky driving behaviors did not produce stat istically significant direct effects (.005 and -.020, respectively). The low beta values for male and age suggest a spurious relationship with Risky Driving Behaviors. The demo graphic variables also had direct effects on the constructs Risk-Taking Attitudes (RT A), Risk Perception (RP), and Knowledge of Road Laws and Signs (KLS). These addit ional values that were obtained through the SEM analysis (See Figure 5) are provide d on Tables 4.16 and 4.17. The beta value, for the relationship between select ed demographics factors and RDB, suggest that being a Driver and having Crash E xperience are associated with Risky Driving Behaviors. The magnitude of the beta weight for persons with Crash Experience suggests that they will most likely exhibit Risky D riving Behaviors. As expected, the beta weight for Drivers also indicates that Drivers will most likely exhibit Risky Driving Behaviors but not as much as those with Crash Exper ience.

PAGE 117

105 Table 4.16 Direct Effects of Demographic Factors on Risk-Takin g Attitudes [RTA], Risk Perception [RP], and Knowledge of Road Laws and Signs [KLS] Relationship value MaleRP .182* MaleRTA .372*** MaleKLS .269*** DriverRP .013 DriverRTA .150** DriverKLS .565*** Crash ExperienceRP -.052 Crash ExperienceRTA -.010 Crash ExperienceKLS -.071 AgeRP -.062 AgeRTA -.062 AgeKLS .141** P < .05 ** P < .01 *** P < .000

PAGE 118

106 Table 4.17 Indirect Effects of Demographic Factors on Risk-Tak ing Attitudes [RTA], Risk Perception [RP], Knowledge of Road Laws and Signs [ KLS], and Risky Driving Behavior [RDB] Relationship – value MaleRP RPRDB -.033 MaleRTA RTARDB .172 MaleKLS KLSRDB .026 DriverRP RPRDB -.002 DriverRTA RTARDB .069 DriverKLS KLSRDB .055 Crash ExperienceRP RPRDB .010 Crash ExperienceRTA RTARDB .005 Crash ExperienceKLS KLSRDB .007 AgeRP RPRDB .011 AgeRTA RTARDB -.029 AgeKLS KLSRDB .014

PAGE 119

107 Research Question 3: What is the relationship betwe en young adults’ risk-taking attitudes and risky driving behaviors?. To determine the relationship between young adults’ Risk-Taking Attitudes and Risky Driving Beh aviors, SEM analysis was conducted on the data. As shown in Figure 5, a stat istically significant direct effect between Risk-Taking Attitudes (RTA) and Risky Drivi ng Behaviors (RDB) is observed with a beta value of .463, p < .000. The factors for RTA are Riding with an Uns afe Driver (A1), Speeding (A2), Concern for Others (A3), and D rinking and Driving (A4). The factors for RDB are Speeding (B1), Distracted Drivi ng (B2), Aggressive Driving (B3), and Not Adhering to Traffic Laws (B4). The factor loadings and residual errors of the fact ors for the RTA construct, .323 to .648 and .580 to .896, respectively, are provide d on Figure 5. The factor loadings and residual errors of the factors for the RDB construc t are also provided on Figure 5 and ranged from .716 to .767 and .411 to .487. The fact or loadings were acceptable and explained a significant portion of the variance for RTA and RDB. For RTA, the pooled items for Drinking and Driving provided the highest factor loadings value followed by Riding with an Unsafe Driver, Speeding, and Concern for Others. For RDB, the pooled items for Aggressive Driving provided the highest f actor loadings value followed by Not Adhering to Traffic Laws, Distracted Driving, and S peeding. The beta value, for the relationship between RTA an d RDB, shows a strong relationship between RTA and RDB. The magnitude of the beta weight for RTA suggests that persons exhibiting high propensity for RTA wil l most likely exhibit Risky Driving Behaviors.

PAGE 120

108 Research Question 4: What is the relationship betwe en young adults’ knowledge of road laws and signs and risky driving behaviors?. To determine the relationship between young adults’ Knowledge of Road Laws and Si gns, and Risky Driving Behaviors, SEM analysis was conducted on the data. As shown in Figure 5, the direct effect between Knowledge of Road Laws and Signs (KL S) and RDB was not statistically significant with an observed beta value of .098. Th e factors for RTA are Knowledge of Road Laws (K1), and Knowledge of Road Signs (K2). T he factors for RDB are Speeding (B1), Distracted Driving (B2), Aggressive Driving ( B3), and Not adhering to Traffic Laws (B4). The factor loadings and residual errors of the fact ors for the KLS construct, .603 to .691 and .637 to .805, respectively, are provide d on Figure 5. The factor loadings and residual errors of the factors for the RDB construc t are also provided on Figure 7 and ranged from .716 to .767 and .411 to .487. The factor loadings were acceptable and explained a significant portion of the variance for KLS. For KLS, the pooled items for Kno wledge of Road Signs had a higher factor loading value than Knowledge of Road Laws. The beta value, for the relationship between KLS an d RDB, does not show a strong relationship between KLS and RDB. The magnit ude of the beta weight for KLS suggests that a high score on KLS did not influence the expression of Risky Driving Behaviors.

PAGE 121

109 Research Question 5: What is the relationship betwe en young adults’ risk perceptions and risky driving behaviors?. To determine the relationship between young adults’ Risk Perceptions and Risky Driving Behavior s, SEM analysis was conducted on the data. As shown in Figure 5, the direct effect b etween Risk Perceptions (RP) and RDB was statistically significant with an observed beta value of .183, p < .05. The factors for RP are Cognition-based perceptions (P1), Concern-ba sed perceptions (P2), and Emotionbased perceptions (P3). The factors for RDB are Spe eding (B1), Distracted Driving (B2), Aggressive Driving (B3), and Not Adhering to Traffi c Laws (B4). The factor loadings and residual errors of the fact ors for the RP construct were, .259 to .759 and .423 to .933, respectively, are pr ovided on Figure 5. The factor loadings and residual errors of the factors for the RDB cons truct are also provided on Figure 8 and ranged from .716 to .767 and .411 to .487. The factor loadings were acceptable and explained a significant portion of the variance for RP and RDB. For RP, the pooled items f or Concern-based perceptions provided the highest factor loadings value followed by Emotion-based perceptions, and Cognition-based perceptions. The beta value, for the relationship between RP and RDB, shows a strong negative relationship between RP and RDB. The magni tude of the beta weight for RP suggests that persons exhibiting low RP will most l ikely exhibit Risky Driving Behaviors.

PAGE 122

110 Additional SEM Results. Risk-Taking Attitudes [RTA] ( = .463 p < .000), Knowledge of Road Laws and Signs [KLS] ( = .098), Driver ( = .177 p < .05) and Crash Experience ( = .228, p < .000) each had a positive direct effect on Risky Driving Behavior [RDB]. Risk Perception [RP] ( = -.183 p < .05), being a Male ( = -.005) and a Driver ( = -.020) each had a negative direct effect on Risk y Driving Behavior. According to the results in Figure 5, Risk-Taking A ttitudes, Knowledge of Road Laws and Signs, Risk Perception, and the combined demogr aphic factors (male, age, driver, and crash experience) account for 35.2% of the vari ance of Risky Driving Behavior. The beta values, for the relationship among selecte d demographics factors, RTA, KLS, RP, and RDB, suggest that these variables are associated with and explain a significant portion of the variance of RDB. The mag nitude of the beta weight for RTA shows the highest influence on RDB followed by Cras h Experience, RP, Driver, and KLS. Consequently, persons, who exhibited high propensit y for RTA, were Drivers, and had Crash Experience, will most likely to exhib it Risky Driving Behaviors. Conversely, persons exhibiting low RP will most lik ely exhibit Risky Driving Behaviors.

PAGE 123

111 Chapter 5: Discussion Introduction This chapter provides a summary of the research and conclusions drawn from the results obtained from the main study’s survey. This chapter comprises seven sections: (a) Research Summary, (b) Discussion of Results, (c) Co nclusions, (d) Strengths and Weaknesses of Study, (e) Data Collection Lessons, ( f) Implications for Public Health, and (g) Implications for Future Research. Research Summary Research has shown that young adults’ risk-taking a ttitudes [RTA], risk perception [RP], and knowledge of road laws and sig ns [KLS] are related to their driving behaviors. The adoption of risky driving behaviors increases young adults’ risk of motor vehicle crashes [MVCs]. This increase in young adul ts’ risk of MVCs helps explain the increased mortality and morbidity experienced by yo ung adults. This research used a conceptual model (see Figure 1) that incorporated t he four constructs (RTA, RP, KLS and risky driving behaviors [RDB]) to analyze the r elationships among them. The purpose of this study was to improve understanding of the factors that lead to increased risks of MVC-related mortality and morbidity for yo ung adults in Belize, to provide support for the development of evidence-based progr ams, and, more importantly, to investigate the relationships involving young adult s’ RTA, RP, KLS and RDB. This study obtained data to investigate the relationship among the fours constructs (RTA, RP,

PAGE 124

112 KLS and RDB) to address five research questions of interest (a) To what extent are the specific stages of the Precaution Adoption Process Model [PAPM] observed in the study population?, (b) What is the relationship between s elected demographic factors and risky driving behavior and the stages of PAPM?, (c) What is the relationship between young adults’ risk-taking attitudes and risky driving beh aviors and the stages of PAPM?, (d) What is the relationship between young adults’ know ledge of road laws and signs and risky driving behaviors and the stages of PAPM?, an d (e) What is the relationship between young adults’ risk perceptions and risky dr iving behaviors and the stages of PAPM? This study used a nonexperimental, cross-sectional research design to illustrate the relationships among RTA, RP, KLS and RDB of you ng adults at the University of Belize [UB]. The study design permitted the examina tion of the strength and direction of relationships among these variables and provided a better understanding of the relationships among the variables that may serve as the basis for future studies, the development of evidence-based intervention programs policy development and health education programs. A convenience sample of 532 stu dents, enrolled at UB during the second semester of the academic year 2006-2007, com pleted the questionnaires. Frequency distributions were used to investigate th e presence of the various stages of PAPM. To investigate the relationships of young adu lts’ RTA, RP, KLS and RDB, structural equation modeling [SEM] was used.

PAGE 125

113 Discussion of Results This section provides a summary of the results used to address the five research questions of this study. The results are summarized after each of the listed research question below. Research Question 1: To what extent are the specifi c stages of the Precaution Adoption Process Model (PAPM) observed in the study population? The enforcement of road laws in Belize does not appear to be a priorit y. As a result, infractions such as speeding, drinking and driving and lack of seatbelt use are frequently committed. It was expected that stages three, four, and five within t he PAPM would be the most prominent stages observed in the population. The results reve aled that the majority of the participants were past the anticipated stages and p articipants were predominantly in stage 6 followed by stage 7. Several factors can help explain the results relate d to question 1. First, the PAPM staging algorithm assigns participants into stages by their responses to questions selected from the survey on seat belt usage. The limitation of this assignment is that individual responses are guided by broad variations of seat be lt usage. The broad variations artificially increased the distributions of individ uals assigned to stages 6 and 7. Second, persons were assigned to stages that were realistic ally inappropriate. For example, some participants reported that they were not aware of t he risks of MVCs and yet they reported that they always used seatbelts. Therefore, a perso n who was assigned to stage 1 may have reported behavior consistent with stage 7. Thi rd, persons may be using seatbelts independent of the recognition or awareness of the risks of MVCs which may help to explain the apparent contradiction of being assigne d to two separate stages. Social

PAGE 126

114 pressures may also encourage persons to adopt seatb elt use in the absence of any acknowledgement of the risks of MVCs. Other factors that occurred prior to the survey may have also contributed to these results. Prominent members of the student body were involved in separate MVCs resulting in major injuries prior to the survey. Th is may have heightened the awareness of the dangers of MVCs. Additionally, the Department o f Transport held its annual “Traffic Safety Week” in mid-November of 2006. This may have had a lasting effect on seatbelt usage. Finally, social desirability response bias m ay have influenced the responses to the survey items for PAPM algorithm. Based on the results, the PAPM staging algorithm wa s not an efficient tool and may be fallible in the identification of the variou s PAPM stages present in this young adult population. Previous research has shown that similarly constructed algorithms have led to the grouping of several PAPM stages (Trifile tti, 2003). The collapsing of these stages reflects the inherent disadvantage of using the algorithm to identify the various PAPM stages. Research Question 2: What is the relationship betwe en selected demographic factors and risky driving behavior?. Of the four demographic factors (being male, being a driver, crash experience, and age) investigated, dr iver and crash experience had a statistically significant positive direct effect on risky driving behaviors. These direct effects on risky driving behaviors can be categoriz ed as small to medium (Kline, 1998). These direct effects suggest that persons with Cras h Experience will most likely exhibit Risky Driving Behaviors and persons who were Driver s would most likely exhibit Risky Driving Behaviors but not as much as those with Cra sh Experience.

PAGE 127

115 Belizean drivers can apply and obtain a driver’s li cense once they meet the prerequisite criteria set by the Transport Departme nt. A written exam is included in this criteria and is based on a 46-item handout sheet c ontaining a minimum number of topics related to the Motor Vehicle and Road Traffic Act ( Attorney General’s Ministry [AGM], 2003).The licensing process is not an intensive pro cess and can be applied at the discretion of the transport officer (AGM, 2003). Th erefore, drivers who pass the written exam would have the bare minimum knowledge of the r oad laws and signs governing motor vehicles in Belize. The items for the survey used in this study were developed from the 46-item handout sheet provided by the Transport Department. Hence, the result showing that drivers would most likely exhibit Risk y Driving Behaviors was not unexpected. The result showing that persons with Crash Experien ce will most likely exhibit Risky Driving Behaviors suggests that these persons have not adopted safe driving practices. However, an overt relationship exists be tween being a driver and demonstrating risky driving behaviors, as the latte r cannot be present without the former. Therefore, the result of interest would be the sign ificant relationship between crash experience and risky driving behaviors. Interesting ly, crash experience did not lower risky driving behaviors; rather, it positively infl uenced the practice of risky driving behaviors. This phenomenon may be largely explained by optimism bias, which may lead to a lowered interpretation of risks of MVC. Optimi sm bias may lead young adults to believe that they have perceived control over these risks as well as a perception of low probability of these risks (Brnstrm et al., 2005; Brown, 2005; Chambers & Windschitl, 2004; Deery, 1999; Dejoy, 1989; Harre, Foster, & O' Neill, 2005; Weinstein, 1987; 1989;

PAGE 128

116 1998; 2003). Even though the risks of MVCs are reco gnized by these drivers, their interpretation of these risks may lead them to beli eve that they have perceived control over these risks as well as a perception of low pro bability of these risks. This finding is in line with previous studies focusing on the interpre tation of risks by persons in 18-to-24 year age group (Brnstrm et al., 2005; Brown, 2005 ; Chambers & Windschitl, 2004; Deery, 1999; Dejoy, 1989; Harre, Foster, & O'Neill, 2005; Jonah, 1986; Weinstein, 1980; 1987; 1989; 1998; 2003). Research Question 3: What is the relationship betwe en young adults’ risk-taking attitudes and risky driving behaviors?. Risk-Taking Attitudes [RTA] had a statistically significant positive direct effect on Risky Driving Behaviors [RDB]. The size of this RTAs’ direct effect on RDB can be categorized as medium t o large (Kline, 1998). The factor loadings were acceptable and explained a significant portion of the variance for RTA. For RTA, the pooled items for Dri nking and Driving provided the highest factor loadings value followed by Riding wi th an Unsafe Driver, Speeding, and Concern for Others. The beta value for the relation ship between RTA and RDB shows a strong relationship between RTA and RDB. The magnit ude of the beta weight for RTA suggests that persons exhibiting high propensity fo r RTA will most likely exhibit Risky Driving Behaviors. The strong relationship between RTA and RDB coincid es with previous research establishing the correlation between attitude and b ehavior and shows that the appropriate corresponding measures for each RTA concept were ut ilized (Kraus, 1995). The results showing a strong relationship between RTA and RDB s upport evidence identifying the specific attitude-behavior correlations (Ajzen, 198 8; Ajzen & Fishbein, 1977; Assum,

PAGE 129

117 1997; Fishbein & Ajzen, 1972; Kraus, 1995; Parker, 2002; Ulleberg & Rundmo, 2002; Ulleberg & Rundmo, 2003; Whissell & Bigelow, 2003). Research Question 4: What is the relationship betwe en young adults’ knowledge of road laws and signs and risky driving behaviors? Knowledge of road laws and signs [KLS] had a positive direct effect on risky driving behaviors that was not statistically significant. As a result of failing to find a stati stically significant result, research question four could not be answered. The factor loadings were acceptable and explained a significant portion of the variance for KLS. For KLS, the pooled items for Kno wledge of Road Signs had a higher factor loading value than Knowledge of Road Laws. T he beta value, for the relationship between KLS and RDB, does not show a strong relatio nship between KLS and RDB. The magnitude of the beta weight for KLS suggests that a high score on KLS was not related to the expression of Risky Driving Behaviors. As mentioned previously, the process of obtaining a driver’s license in Belize is not an intensive one and requirements can be applie d at the discretion of the transport officer (AGM, 2003). Drivers who pass the written e xam would have the bare minimum knowledge of the road laws and signs governing moto r vehicles in Belize. In addition, the SEM analysis of the survey’s knowledge may have bee n influenced by missing data. Although the link between knowledge and behavior ha d been establish, previous research indicated that possessing knowledge does not necess arily translate into and adoption of safe behaviors. The lack of a comprehensive knowled ge of road laws and signs and the effect of missing data would help explain the lack of a statistically significant relation between KLS and RDB.

PAGE 130

118 Research Question 5: What is the relationship betwe en young adults’ risk perceptions and risky driving behaviors?. Of the three constructs of interest (Risk-Taking Attitudes, Risk Perception and Knowledge of Road La ws and Signs) investigated, Risk Perception [RP] had a statistically significant neg ative direct effect on risky driving behaviors. The size of this risk perceptions’ direct effect on risky driving behaviors can be categorized as small to medium (Kline, 1998). The beta value, for the relationship between RP and RDB, shows a negative relationship between RP and RDB and suggests that p ersons exhibiting low RP will most likely exhibit Risky Driving Behaviors. The finding of lowered risk perception is consistent with previous research showing that youn g drivers were more likely to adopt risky driving behaviors due to their low perceived risks of being involved in a crash (Deery, 1999; Finn & Bragg, 1986; Mathews & Moran, 1986; Williams 2003). Conclusions Overall this study found that a significant relatio nship exists among the three attitudinal factors of interest (young adults’ risk -taking attitudes [RTA], risk perception [RP], and knowledge of road laws and signs [KLS]) a nd risky driving behaviors [RDB]. These three factors in the model accounted for 35.2 % of the variance and helped to explain risky driving behaviors. Of the three attit udinal factors studied, young adults’ RTA was the main predictor for risky driving behavi ors. The influence of RTA is supported by previous research investigating the re lationship between RTA and RDB (Iversen, 2004; Malfetti, Rose, DeKorp & Basch, 198 9; Parker, 2002; Ulleberg & Rundmo, 2002; Ulleberg & Rundmo, 2003; West & Hall, 1997). Previous research has shown a strong correlation between RTA and RDB (Ive rsen, 2004; Malfetti, Rose,

PAGE 131

119 DeKorp & Basch, 1989; Parker, 2002; Ulleberg & Rund mo, 2002; Ulleberg & Rundmo, 2003; West & Hall, 1997). The second factor related to RDB was the factor RP. The relationship between RP and RDB was different than the RTA-RDB relationship in that RP was negatively related to RDB. This negative rel ationship means that as RP is lowered more RDBs are manifested. The finding is su pported by previous research (Deery, 1999; Finn & Bragg, 1986; Frick, Rehm, Knol l, Reifinger, & Hasford, 2000; Sagberg & Bjrnskau, 2006; Williams 2003). The weak est relationship was observed between KLS and RDB. Previous research has highligh ted the importance of utilizing educational strategies as an integral component of effective interventions to address MVCs (Al-Madani, 2000; Al-Madani, & Al-Janahi, 2002 a, 2002b; Carstensen, 2002; Charlton, 2004, 2005; Dinh-Zarr et al., 2001; Hatak ka, Keskinen, Gregersen, Glad, & Hernetkoski, 2002; Hedlund & Comptom, 2005; Mayhew & Simpson, 2002). Although this study did not demonstrate a significant relati onship between KLS and RDB, the importance of educational strategies to increase aw areness of and adherence to motor vehicle laws, safety measures and risks of MVCs is supported by previous research (Dinh-Zarr, et al., 2001; Task Force on Community P reventive Services, 2001; WHO, 2004b). In conclusion, this study showed that young adults had elevated risk-taking attitudes and low risk perceptions, all of which in creased the manifestation of risky driving behaviors. This study did not identify a re lationship between knowledge of road laws and road signs and risky driving behaviors tha t was statistically significant. Overall, the findings suggest that interventions should focu s on reducing risk-taking attitudes and

PAGE 132

120 on increasing risk perceptions. Such interventions may reduce the manifestation of risky driving behaviors and in turn, lower the incidence of MVC-related injuries and deaths. Strengths and Limitations of Study Strengths. MVCs have been one of the 10 leading causes of dea th in Belize. However, research into the behavioral factors that contribute to MVCs has not been carried out. This study is the first one of its kin d that investigates the factors that contribute to MVCs in Belize. Apart from national a nd Non-Governmental Organizations [NGO] reports, only one non-peer reviewed journal a rticle about MVCs was discovered through the literature review. The national and NGO ’s reports dealt only with univariate analyses of MVCs and have provided sparse details o n epidemiological data related to MVC-related injuries and deaths without delving int o the underlying causes of MVCs. Hence, this study initiates a formal attempt to rec ord and understand the factors related to behaviors that may increase the risk of MVCs. There fore provides solid data that could be used to develop interventions seeking to mitigat e the effects of MVCs. The participants in this study were young adults 18 to 24 years of age who were enrolled at UB during the 2006-2007 academic year. The results of this study may be generalizable to young adults enrolled at the Unive rsity of Belize. The results may also be generalizable to students enrolled in other tert iary level institutions in Belize, because the university student population is derived from t hese feeder institutions Apart from being the first study to investigate th e factors related to risky driving behavior, this study also places the focus on attit udinal factors that may influence the risk of being involved in MVCs, and by extension, be at risk of injury or death due to MVCs. The two previous studies focused on basic univariat e analysis on the number of crashes

PAGE 133

121 and legal charges that were levied. Hence, this stu dy brings a more in-depth analysis of factors that would provide a sound basis for interv ention development. The structural equation modeling [SEM] procedure pr ovided a robust simultaneous analysis of the multiple factors of in terest rather than conducting multiple individual univariate or bivariate analyses of the same factors. SEM also provided a platform to analyze latent variables. Consequently, constructs that can not be measured directly are measured by using multiple indicators that provide a conceptual basis for a particular construct. As a result, SEM makes it pos sible to measure constructs that were used in this study to identify their relationships. Limitations This study has several limitations or methodologi cal weaknesses that necessitate caution in interpreting or generalizing the study’s results (Heppner & Heppner, 2004; Pyrczak & Bruce, 2000). Participation in this study was voluntary and based on self-reporting from the participants and may be sensitive to social desirab ility bias. The self-reporting may increase the possibility of social desirability bia s that has been found in studies utilizing questionnaires and interviews. Over-reporting of se atbelt use is another phenomenon that would influence seatbelt use reports. Nelson (1996) found over-reporting of seatbelt use rates ranging from 2% to 4%. The participants in this study were young adults 18 to 24 years of age who were enrolled at UB. These participants may differ from young adults in the general population including those who are enrolled in other junior co lleges. This study’s results may only be applicable to young adults enrolled at UB.

PAGE 134

122 The fit indices for this study would globally not b e acceptable but still provided a reasonable basis for further interpretation. Howeve r, good or perfect fitting models can also present unique problems for interpretation (To marken & Waller, 2003) and the suggestion is that focus should center on a model t hat fits closely rather than one that fits exactly. The less than acceptable fit indices could be due to a number of reasons not limited to the correlation of error variances which have been observed in research utilizing survey instruments as well as to the effe cts of sample size. Error variances are expected to be uncorrelated and adjustments to any correlation would require analyses that have not provided reliable statistical adjustm ents. SEM analysis has been referred to as causal models with unmeasured variables and has been used to establish causality. However, the cross-sectional design of this study limits the conclusions that can be drawn and the re sults are not appropriate for the establishment of cause and effects of the variables in this study. Thus, this cross-sectional study is correlational in nature and any causal rel ationship cannot be inferred. Whereas the argument can be made that this study may meet t wo of the three conditions needed to establish causality, this study does not meet the e xperimental rigor criterion to establish causality. Data Collection Lessons Questionnaire Reception. Students were receptive to the idea of completing the questionnaire even though the questionnaire seemed lengthy. Only two students declined to participate in this survey. I believe this high participation or low attrition rate may be partly due to the novelty of participation in surve ys at the university. Students who completed the survey expressed interest in the time liness of the study. On various

PAGE 135

123 occasions after the completion of the survey, stude nts asked questions pertaining to the importance of addressing the perceived high numbers of MVCs and the increased risk of injury due to MVCs. In addition, lecturers were sup portive of the research by generously providing class time for the distribution, completi on and collection of the questionnaire. Logistics. Initially, the survey was scheduled for the first semester of the 2006 to 2007 academic year. Final IRB approval was not rece ived until late into that semester. Preparations were made to continue with the survey, however, after consultation with faculty members the plans for data collection were postponed to the following semester. Logistically, this proved to be a fortuitous decisi on. The commencement of a new semester provides an opportune time when lecturers have an inclination to accommodate surveys during class time. Permission and support w ere received from the deans of the various faculties to contact the individual lecture rs of the selected courses. This individualized contact more than likely served to c onvince lecturers to support the survey by allotting class time. Implications for Public Health WHO (2004b) has recognized the impact of MVCs world wide and declared the 2004 World Health Day to promote awareness, encoura ge discussion and mobilize action to address MVCs. Belize has also recognized the imp act of MVCs, as well as, the urgency of developing research-based intervention p rograms to address the enormous challenge in maintaining a healthy young population The collection of data relating to MVC injuries in Belize, as an integral part of rese arch-based intervention programs, is essential. Currently, a systematized approach for a ddressing injuries, especially those related to MVCs, in Belize is practically nonexiste nt. The interventions applied in Belize

PAGE 136

124 do not appear to be based on studies providing nece ssary data or a theoretical basis for their application. The approach of implementing int erventions without in-depth investigation as to whether they are relevant to th e Belizean population may not appropriately incorporate the factors affecting or impacting drivers’ behavior and other factors contributing to MVCs in Belize. However, th is trend is about to change. The government of Belize has recently established an in ter-ministerial task force to provide a more cohesive and comprehensive approach to address MVC-related injury and deaths. This study’s result can be used to formulate interv entions to decrease young adults’ risky driving behaviors. The results can be used to provide direction to current health education strategies. Since risk-taking atti tudes seems to have a large impact on risky driving behavior, interventions that are deve loped should focus on improving young adults’ risk-taking attitudes to help them lower th eir risky driving behaviors. These interventions would target the attitudes that lead to speeding, aggressive driving, distracted and not adhering to traffic laws. Second ly, interventions should also target young adults’ low risk perceptions and assist them in recognizing dangers associated with risky driving behaviors. The study’s result showed that the participants had lowered perceptions of risks related to risky driving behav iors. Therefore, interventions should highlight the risks associated with risky driving b ehaviors and enable young adults to identify risky driving conditions that may lead to motor vehicle crashes. The analyses did not show any significant relations hip between knowledge of road laws and signs with risky driving behaviors. H owever, the information pamphlet and written exam, that applicants receive, does not ade quately provide sufficient information on road laws and signs. The information for the wri tten test consists of 48 statements that

PAGE 137

125 cover topics related to road laws and signs as well as other miscellaneous topics. These miscellaneous topics cover rules on roadways that n o longer exist or are applicable to only one city in Belize. Therefore, a recommendatio n to the Department of Transport would be to conduct a complete revision of the exam ination process. The revision would ensure that updated information is included in the driver education package and that obsolete and irrelevant information are removed. Th e revised knowledge section of the license process would allow the driver’s license ap plicant to become familiar with Belize’s road laws and signs. Because a lack of adequate epidemiological and soci oeconomic data on MVCs at the national level impedes effective national and i nternational response (WHO, 2004d), this study adds significantly to the body of knowle dge on MVCs in Belize. This study provides support for a systematized approach in the development and implementation of intervention programs addressing the effects of MVC s. Although these findings may be applicable to a limited sample, they coincide with previous research that supports the relationship among risk-taking attitudes, risk perc eption, and risky driving behaviors. As such, these findings add to the body of knowledge f ocusing on behavioral factors related to MVCs, especially in Belize where limited researc h on this topic has been conducted. Implication for Future Research Research findings from these studies may not be app licable to the larger population of young adults in Belize. Therefore, fu rther research must be conducted to investigate whether similar findings will be obtain ed if the survey was conducted in the other 11 junior colleges in Belize. A comparison ca n then be made to determine whether

PAGE 138

126 the findings from this study differ from findings o f research that includes a selected number of junior colleges in Belize. Apart from surveying the other junior colleges, fur ther research must include the young adult population who are not enrolled in trad itional junior colleges (e.g. Institute for Technical Vocational Education and Training) or those who are not enrolled in any educational institution. Research that includes a m ore diverse sample of the young adult population may validate the findings of this study or may provide results that differ from the ones of this study. Such findings are of key im portance to any intervention strategies that are developed to address the risk of injury or death due to MVCs. In addition, a thorough examination of the process of obtaining a driver’s license must be carried out with the purpose of strengtheni ng the criteria required to obtain a driver’s license. Currently, the criteria for obtai ning a driver’s license are not stringent and may be discretionary in application. A written exam is included in the criteria and is based on a 46-item handout sheet containing a minim um number of topics related to the Motor Vehicle and Road Traffic Act (AGM, 2003). The refore, drivers who pass the written exam would have the bare minimum knowledge of the road laws and signs governing motor vehicles in Belize. A revision of t he test material is warranted. This revision must be in tandem with a comprehensive ove rhaul of the 46-item informational handout sheet provided by the Transport Department. Furthermore, research into whether changes in attit udinal factors occur over time must be explored and conducted. These longitudinal studies would capture any changes or the stability of the attitudinal factors related to risky driving behaviors. Longitudinal studies would also capture any changes that may be attributed to research-based

PAGE 139

127 intervention programs and serve as a means of evalu ating the effectiveness of these programs. Although these findings are derived from attitudina l factors related to driving behaviors, the key concepts of this study have been found to be applicable and relevant to other health behaviors of young adults (Brnstrm e t al., 2005; Brown, 2005; Chambers & Windschitl, 2004; Deery, 1999; Dejoy, 1989; Harre Foster, & O'Neill, 2005; Weinstein, 2003; Williams 2003). Therefore, future research could investigate whether the finding of this study can be applied to other r isky behaviors such as substance abuse, intimate partner violence, and youth violence that pose a risk to the wellbeing of young adults in Belize. One of the results was that the majority of the pa rticipants reported using seatbelts either occasionally or always. The freque ncy of reported seatbelt use seemed high considering that enforcement of seatbelt use i s not a priority in Belize. The reported high seatbelt usage rate should be verified by furt her research to determine whether it represent an accurate representation of adoption of precautions or just a figure influenced by social desirability response bias. If the former is supported, then an opportunity exists for the development of interventions that highlight the adoption of seatbelt use and other precautions. Although this study only investigated human factors related to MVCs, other aspects that are related to MVCs also need to be in vestigated. Further research would focus on road engineering, enforcement, legislative and educational factors that are related to MVCs in Belize. As an example, research could be conducted to determine

PAGE 140

128 whether the road and highways are built with engine ering designs to reduce the risk of MVCs. Students expressed an interest in the study’s topic During both the pilot testing of the survey and the main study, students asked quest ions pertaining to the importance of addressing the perceived high numbers of MVCs and t he increased risk of MVCs related injury and death. The students’ expressed interest may provide an opportunity to conduct qualitative research on their interpretation and pe rceptions of factors related to risky driving behaviors. Qualitative research could be us ed to develop evidenced-based interventions that focus on risk-taking attitudes a nd risk perception of selected driving behaviors.

PAGE 141

129 References Afukaar, F. K. (2003). Speed control in developing countries: Issues, challenges and opportunities in reducing road traffic injuries. Injury Control and Safety Promotion, 10 (1-2), 77-81. Afukaar, F. K., Antwi, P., & Ofosu-Amaah, S. (2003) Pattern of road traffic injuries in Ghana: Implications for control. Injury Control and Safety Promotion, 10 (1-2), 69-76. Ajzen, I. (1988). Attitudes, personality, and behavior Chicago: The Dorsey Press. Ajzen, I., & Fishbein, M. (1977). Attitude-behavior relations: A theoretical analysis and review of empirical research. Psychological Bulletin, 84 888-918. Al-Madani, H. (2000). Influence of drivers' compreh ension of posted signs on their safety related characteristics. Accident Analysis and Prevention, 32 (4), 575-581. Al-Madani, H., & Al-Janahi, A. R. (2002a). Role of drivers' personal characteristics in understanding traffic sign symbols. Accident Analysis and Prevention, 34 (2), 185196. Al-Madani, H., & Al-Janahi, A. R. (2002b). Assessme nt of drivers' comprehension of posted signs on their safety related characteristic s. Transportation Research Part F: Traffic Psychology and Behaviour, 5 (1), 187-200. Arnett, J. J. (2000). Emerging adulthood. A theory of development from the late teens through the twenties. The American Psychologist, 55 (5), 469-480.

PAGE 142

130 Ashley, D., & Holder, Y. (2002). The Jamaican injur y surveillance system: Lessons learnt. Injury Control and Safety Promotion, 9 (4), 263-264. Asiamah, G., Mock, C., & Blantari, J. (2002). Under standing the knowledge and attitudes of commercial drivers in Ghana regarding alcohol im paired driving. Injury Prevention: Journal of the International Society fo r Child and Adolescent Injury Prevention, 8 (1), 53-56. Assum, T. (1997). Attitudes and road accident risk. Accident Analysis and Prevention, 29 (2), 153-159. Attorney General’s Ministry, (2003 Ed.). The Substa ntive Laws of Belize, Volume V Title XXI Chapter 230: Motor Vehicles and Road Traf fic. Government of Belize: The Government Printer. Berk, L.E. (2004). Development through the lifespan (3rd Ed.). Boston: Allyne and Bacon. Blalock, S. J., DeVellis, R. F., Giorgino, K. B., D eVellis, B. M., Gold, D. T., Dooley, M. A., Anderson, J. J., & Smith, S. L. (1996). Osteopo rosis prevention in premenopausal women: using a stage model approach t o examine the predictors of behavior. Health Psychology, 15 (2), 84-93. Boomsma, A. (2000). Reporting analyses of covarianc e structures. Structural Equation Modeling 7 (3), 461-483. Brnstrm, R., Kristjansson, S., & Ulln, H. (2005) Risk perception, optimistic bias, and readiness to change sun related behaviour. European Journal of Public Health. Retrieved January 10, 2006, from http://eurpub.oxfordjournals.org/cgi/rapidpdf/cki19 3v1

PAGE 143

131 Brown, S. L. (2005). Relationship between risk-taki ng behavior and subsequent risk perceptions. British Journal of Psychology, 96 (Pt. 2), 155-164. Buhi, E. R., Goodson, P., & Neilands, T. B. (2007). Structural equation modeling: A prime for Health Behavior Researchers. American Journal of Health Behavior, 31 (1), 74-85. Cameron, N. (2001). Human growth and development San Diego, California: Academic Press. Carstensen, G. (2002). The effect on accident risk of a change in driver education in Denmark. Accident Analysis and Prevention, 34 (1), 111-121. Centers for Disease Control and Prevention, (2005). Eliminate disparities in infant mortality. Retrieved March 10, 2005, from http://www.cdc.gov/brfss/ Central Statistical Office. (2001). Abstract of Statistics: Belize 2001. Ministry of Finance, Government of Belize. Central Statistical Office. (2005). Abstract of Statistics: Belize 2004 Ministry of Finance, Government of Belize. Charlton, S. G. (2004). Perceptual and attentional effects on drivers' speed selection at curves. Accident Analysis and Prevention, 36 (5), 877-884. Charlton, S. G. (2005). Conspicuity, memorability, comprehension, and priming in road hazard warning signs. Accident Analysis and Prevention, Article in Press. Retrieved February 1, 2005, from http://www.sciencedirect.com.proxy.usf.edu/ Chambers, J. R., & Windschitl, P. D. (2004). Biases in social comparative judgments: The role of nonmotivated factors in above-average a nd comparative-optimism effects. Psychological Bulletin, 130 (5), 813-838.

PAGE 144

132 Cottrell, R. R., Girvan, J. T., & McKenzie, J. F. ( 2nd Ed.) (2002). Principle & foundations of health promotion and education. San Francisco: Benjamin Cummings. Crandon, I. W., Branday, J. M., Simeon, D. T., Rhod en, A., Thompson, H., & Carpenter, R. (1996). The prevalence of seatbelt use in Jamaic a: An observational study. West Indian Medical Journal, 45 (1), 31-33. Crandon, I. W., Carpenter, R., & McDonald, A. (1994 ). Admissions for trauma at the University Hospital of the West Indies. West Indian Medical Journal, 45 (4), 117120. Cronbach, L. J. (1951). Coefficient alpha and the i nternal structure of tests. Pyschomertrika, 16 (3), 297-334. Crundall, D., & Underwood, G. (2001). The priming f unction of road signs. Transportation Research Part F: Traffic Psychology and Behaviour, 4 (3), 187200. Deery, H.A. (1999). Hazard and risk perception amon g young novice drivers. Journal of Safety Research 30 (4), 225-236. DeJoy, D. M. (1989). The optimism bias and traffic accident risk perception. Accident Analysis and Prevention, 21 (4), 333-340. DeYoung, D. (1999). An evaluation of the specific d eterrent effects of vehicle impoundment on suspended, revoked, and unlicensed d rivers in California. Accident Analysis and Prevention, 31 45-53.

PAGE 145

133 DeYoung, D. (2000). An evaluation of the general de terrent effect of vehicle impoundment on suspended and revoked drivers in Cal ifornia. Journal of Safety Research, 31 (2), 51-59. Department of Transport [DOT]. (2004). Information for the written test. Ministry of Works, Transport and Communications, Government of Belize. DiClemente, R.J. Crosby, R.A., & Kegler, M.C. (2002 ). Emerging theories in health promotion practice and research: Strategies for imp roving public health San Francisco: Jossey-Bass. Dinh-Zarr, T. B., Sleet, D. A., Shults, R. A., Zaza S., Elder, R. W., & Nichols, J. L. et al. (2001). Reviews of evidence regarding interventions to increase the use of safety belts. American Journal of Preventive Medicine, 21 (4 Suppl), 48-65. Eby, D. W., Molnar, L. J., Shope, J. T., Vivoda, J. M., & Fordyce, T. A. (2003). Improving older driver knowledge and self-awareness through self-assessment: The driving decisions workbook. Journal of Safety Research, 34 (4), 371-381. Evans, L. (2004). Traffic Safety. Bloomfield, Michigan: Science Serving Society. Evans, L. (2003). A new traffic safety vision for t he United States. American Journal of Public Health, 93 (9), 1384-1386. Evans, C. A. Jr, Fielding, J. E., Brownson, R. C., et al. (2001). Motor-vehicle occupant injury: strategies for increasing use of child safe ty seats, increasing use of safety belts, and reducing alcohol-impaired driving. Morbidity and Mortality Weekly Report, 50 (RR07), 1-13.

PAGE 146

134 Ferguson, S. A. (2003). Other high-risk factors for young drivers--how graduated licensing does, doesn't, or could address them. Journal of Safety Research, 34 (1), 71-77. Ferguson, S. A., & Williams, A. F. (2002). Awarenes s of zero tolerance laws in three states. Journal of Safety Research, 33 (3), 293-299. Figuerido, L. F., Rasslan, S., Bruscagin, V., Cruz Jr., R., & Rocha, M., (2001). Increases in fines and driver licence withdrawal have effecti vely reduced immediate deaths from trauma on Brazilian roads: First-year report o n the new traffic code. Injury: International Journal of the Care of the Injured 32 (2), 91-94. Finn, P., & Bragg, W.E. (1986). Perception of the r isk of an accident by young and older drivers. Analysis and Prevention, 18 (4), 289-298. Fishbein, M., & Ajzen, I. (1972). Attitudes and opi nions. Annual review of Psychology, 23 487-544. Fisher, J. (1992). Testing the effect of road signs ’ informational value on driver behavior. Human Factors, 34 (2) 231-237. Flores, G. (1999). 1998 road traffic accidents in t he Belize District, Belize Central America. Draft of Diploma Thesis, Faculty of Medica l Sciences, University of the West Indies, Mona, Jamaica. Forjuoh, S.N. (2003). Traffic-related injury preven tion interventions for low-income countries. Injury Control and Safety Promotion, 10 (1-2), 109-118. Fraser, B. (2005). Traffic accidents scar Latin Ame rica’s roads. Lancet, 366 (9487), 703704.

PAGE 147

135 Frick, U., Rehm, J., Knoll, A., Reifinger, M., & Ha sford, J. (2000). Perception of traffic accident risk and decision to drive under light alc ohol consumption--a doubleblind experimental study. Journal of Substance Abuse, 11 (3), 241-251. Glanz, K., Rimer, B.K., & Lewis, M.L. (Eds.) (2002) Health behavior and health education: Theory, research, and practice. Jossey-Bass: John Wiley and Sons, Inc. Goldscheider, F., & Goldscheider, C. (1999). The changing transition to adulthood: Leaving and returning home Thousand Oaks, California: Sage Publications, Inc Greening, L., & Stoppelbein, L. (2000). Young drive rs' health attitudes and intentions to drink and drive. Journal of Adolescent Health, 27 (2), 94-101. Grossman, D. C., & Garcia, C.C. (1999). Effectivene ss of health promotion programs to increase motor vehicle occupant restraint use among young children. American Journal of Preventive Medicine 16 (1S), 12-22. Harre, N., Foster, S., & O'Neill, M. (2005). Self-e nhancement, crash-risk optimism and the impact of safety advertisements on young driver s. The British Journal of Psychology, 96 (Pt 2), 215-230. Hatakka, M., Keskinen, E., Gregersen, N. P., Glad, A., & Hernetkoski, K. (2002). From control of the vehicle to personal self-control; br oadening the perspectives to driver education. Transportation Research Part F: Traffic Psychology and Behaviour, 5 (3), 201-215. Hatcher, L. (1994). A step-by-step approach to using SAS for Factor Analysis and Structural Equation Modeling. Cary, NC: SAS Institute Inc.

PAGE 148

136 Hedlund, J., & Compton, R. (2004). Graduated driver licensing research in 2003 and beyond. Journal of Safety Research, 35 (1), 5-11. Hedlund, J., & Compton, R. (2005). Graduated driver licensing research in 2004 and 2005. Journal of Safety Research, 36 (2), 109-119. Hedlund, J., Shults, R. A., & Compton, R. (2003). W hat we know, what we don't know, and what we need to know about graduated driver lic ensing. Journal of Safety Research, 34 (1), 107-115. Heppner, P. P., & Heppner, M.J. (2004). Writing and publishing your Thesis, Dissertation & Research: A guide for students in th e helping professions. Belmont, CA: Thomson-Brooks/Cole Press. Hjar, M., (2004). Road traffic injuries research i n the Latin American and the Caribbean region. Unpublished document Retrieved June 10, 2005, from http://www.globalforumhealth.org/Forum8/Forum8CDROM/OralPresentations/Hijar%20M%20F8-647.doc Hjar-Medina, M. C., Carillo-Ordaz, C. E., Flores-A ldana, M. E., Anaya, R., & LpezLpez, M. V. (1999). Factores de riesgo de lesin p or accidentes de trfico y el impacto de una intervencin sobre la carretera. Revista de Sade Pblica, 33 (5), 505-12. Retrieved June 10, 2005, from http://www.fsp.usp.br/rsp/ Hjar, M. C., Kraus, J. F., Tovar, V., & Carillo, C (2001). Analysis of fatal pedestrian injuries in Mexico City, 1994-1997. Injury: International Journal of the Care of the Injured 32 (4), 279-284.

PAGE 149

137 Hjar, M., Vazquez-Vela, E., & Arreola-Risa, C. (20 03). Pedestrian traffic injuries in Mexico: A country update. Injury Control and Safety Promotion, 10 (1-2), 37-43. Hjar, M. Arredondo, A., Carillo. C., & Solrzano, L. (2004). Road Traffic injuries in an urban area in Mexico: An epidemiological and cost a nalysis. Accident Analysis and Prevention, 36 (1), 37-42. Holder, Y. (2002). Injuries in Jamaica. Injury Control and Safety Promotion, 9 (4), 217218. Hoyle, R. H., M.J. (Ed.) (1995). Structural Equation Modeling: Concepts, Issues and Application. Thousand Oaks, London: Sage Publications. Hu, L., & Bentler, P.M. (1999). Cutoff criterion fo r fit indexes in covariance structure analysis: conventional criteria versus new alternat ives. Structural Equation Modeling, 6 (1), 1-55. Huebner, A. (2000). Adolescent Growth & Developmen t. Family and Child Development Publication #350-850, Retrieved June 10, 2005, fr om Virginia Cooperative Extension Web site: http://www.ext.vt.edu/resources/fcspublications.htm l Iversen, H. (2004). Risk-taking attitudes and risky driving behavior. Transportation Research Part F: Traffic Psychology and Behaviour, 7 (3), 135-150. Iversen, H. & Rundmo, T. (2004). Attitudes towards traffic safety, driving behavior and accident involvement among the Norwegian public. Ergonomics, 47 (5), 555-572. Iversen, H., Rundmo, T, & Klempe, H. (2005). Risk a ttitudes and behavior among Norwegian adolescents. European Psychologist, 10 (1), 25-38.

PAGE 150

138 Jacobs, G., Aeron-Thomas, A., & Astrop, A. (2000). Estimating global road fatalities. TRL Report 445. London, England: Transport Research Laboratory. Joint Intelligence Coordinating Center of the Polic e Department [JICCPD] (2005). Road Traffic Accident for the years 2003 and 2004. Unpub lished data February, 2005. Johansson, G., & Backlund, F. (1970). Drivers and r oad signs. Ergonomics, 13, 749-756. Jonah, B. A. (1986). Accident risk and risk-taking behavior among young drivers. Accident Analysis and Prevention, 18 (4), 255-271. Jorgensen, F., & Wentzel-Larsen, T. (1999). Optimal use of warning signs in traffic. Accident Analysis and Prevention, 31 (6), 729-738. King, P. M. (1978). William Perry’s theory of intel lectual and ethical development. In L. Knefelkamp, C. Widick, & C. A. Parker. (Eds.) Applying new developmental findings. (pp. 35-50) New Directions for Student Services, No. 4. San Fr ancisco: Jossey-Bass. Kline, R.B. (1998). Principles and practice of structural equation mode ling New York: Guilford Press. Kopits, E., & Cropper, M. (2003). Traffic fatalitie s and economic growth. World Bank Policy Research Working Paper 3035. World Bank. Kopits, E., & Cropper, M. (2005). Traffic fatalitie s and economic growth. Accident Analysis and Prevention, 37 169-178. Kraus, S. J. (1995). Attitudes and the prediction o f behavior: A meta-analysis of the empirical literature. Personality and Social Psychology Bulletin 21(1), 58-75.

PAGE 151

139 Labouvie-Vief, G. (1996). Emotion, thought, and Gen der. In C. Magai & S. H. McFadden (Eds.), Handbook of emotion, adult development, and aging (pp. 101117). San Diego: Academic Press. Labouvie-Vief, G. (1999). Emotions in Adulthood. I n V. L. Bengtson & K. W. Schaie (Eds.), Handbook of theories of aging (pp. 253-267) New York: Springer Pub. Co. Laapotti, S., & Keskinen, E. (2004). Has the differ ence in accident patterns between male and female drivers changed between 1984 and 2000? Accident Analysis and Prevention, 36 (4), 577-584. Le Franc, E. & Alleyne, D. (2004). Morbidity and m ortality from external causes as obstacles to development. Retrieved September 10, 2 005, from http://www.cpc.paho.org/%5CFiles%5CDocFiles%5C60_13 0.pdf Leung, S. & Starmer, G. (2005). Gap acceptance and risk-taking by young and mature drivers, both sober and alcohol-intoxicated, in a s imulated driving task. Accident Analysis and Prevention, 37 1056-1065. Retrieved December 15, 2005, from http://www.sciencedirect.com.proxy.usf.edu/ Litwin, M. S. (1995). How to measure survey reliability and validity Thousand Oaks, California: Sage Publications, Inc. Love, P. G., & Guthrie, V. L. (1999, Winter). Perry ’s intellectual scheme. In Understanding and applying cognitive development th eory. New Directions for Student Services, No. 88. San Francisco: Jossey-Bas s. Malfetti, J.L., Rose, P.R., DeKorp, N.A., & Basch, C.E. (1989). Young Driver Attitude Scale: The development & field testing of an instr ument to measure young driver

PAGE 152

140 risk-taking attitudes & some exploratory interventi ons. New York: New York Teachers College, Columbia. Malina, R. M., Bouchard, C. & Bar-Or, O. (2004). Growth, maturation, and physical activity. Champaign, Illinois: Human Kinetics. Maruyama, G. M. (1998). Basics of Structural Equation modeling. SAGE Publications Inc., Thousand Oaks, California. Masten, S. V., & Chapman, E. A. (2004). The effecti veness of home-study driver education compared to classroom instruction: The im pact on student knowledge and attitudes. Traffic Injury Prevention, 5 (2), 117-121. Masten, S. V., & Hagge, R. A. (2004). Evaluation of California's graduated driver licensing program. Journal of Safety Research, 35 (5), 523-535. Mathews, M. L., & Moran, A. R. (1986). Accident ris k and risk-taking behavior among young drivers. Accident Analysis and Prevention, 18 (4), 299-313. Mayhew, D. R., & Simpson, H. M. (2002). The safety value of driver education and training. Injury Prevention: Journal of the International So ciety for Child and Adolescent Injury Prevention, 8 (Suppl 2); ii7-8. McArthur, D.L., & Kraus, J.F. (1999). The specific deterrence of administrative per se laws in reducing drunk driving recidivism. American Journal of Preventive Medicine, 16 (1S), 68-75. McDonald, A. H. (2002). Trauma care in Jamaica. West Indian Medical Journal, 51 (3), 167-170.

PAGE 153

141 McDermott, R.J., & Sarvela, P.D. (1999). Health Education Program Evaluation and Measurement: A Practitioner's Perspective 2nd edition. Madison, WI: Brown & Benchmark. McKnight, A. J., & Peck, R. C. (2003). Graduated dr iver licensing and safer driving. Journal of Safety Research, 34 (1), 85-89. Murray, C. J., & Lopez, A.D. (1997a). Global mortal ity, disability, and the contribution of risk factors: Global Burden of Disease Study. Lancet, 349 (9061), 1269-1276. Murray, C. J., & Lopez, A.D. (1997b). Global mortal ity, disability, and the contribution of risk factors: Global Burden of Disease Study. Lancet, 349 (9063), 1436-1442. Murray, C. J., & Lopez, A.D. (1997c). Global mortal ity, disability, and the contribution of risk factors: Global Burden of Disease Study. Lancet, 349 (9064), 1498-1504. Nantulya, V. M., & Reich, M. R. (2002). The neglect ed epidemic: road traffic injuries in developing countries. British Medical Journal, 324 1139-1141. Nantulya, V. M., & Reich, M. R. (2003). Equity dime nsions of road traffic injuries in lowand middle-income countries. Injury Control and Safety Promotion, 10 (12), 13-20. National Committee for Injury Prevention and Contro l, (2005). Teen drivers: Fact sheet. Retrieved March 10, 2005, from http://www.cdc.gov/ncipc/factsheets/teenmvh.htm National Health Information and Surveillance Unit [ NHISU] (2003). Summary of annual mortality and morbidity statistics. Unpublished dat a. Ministry of Health, Government of Belize. National Research Council and Institute of Medicine [NRCIM], 2005. Growing up

PAGE 154

142 global: The changing transitions to adulthood in de veloping countries. Panel on Transitions to adulthood in developing countries. C B. Lloyd (Ed.). Committee on population and board on children, youth and fami lies. Division of behavioral and social sciences and education. Washington, D.C. : The National Academies Press. Nelson, D.E., (1996). Validity of self-reported dat a on injury prevention behavior: Lessons from observational and self-reported survey s of safety belt use in the US. Injury Prevention 2 67-69. Nelson, L. J., & Barry, C. M. (2005). Distinguishin g features of emerging adulthood: The role of self-classification as an adult. Journal of Adolescent Research, 20 (2), 242262. Noar, S. M., & Zimmerman, R. S. (2005). Health beha vior theory and cumulative knowledge regarding health behaviors: are we moving in the right direction? Health Education Research, 20 (3), 275-290. Odero, W., Khayesi, M., & Heda, P. M. (2003). Road traffic injuries in Kenya: Magnitude, causes and status of intervention. Injury Control and Safety Promotion, 10 (1-2), 53-61. Odero, W., Garner, P., & Zwi, A. (1997). Road traff ic injuries in developing countries: A comprehensive review of epidemiological studies. Tropical Medicine and International Health, 2 (5), 445-460. O’Niell, B., & Mohan, D. (2002). Reducing motor veh icle crash deaths and injuries in newly motorizing countries. British Medical Journal, 324 1142-1145. O’Rourke, N., Hatcher, L., & Stepanski, E.J. (2005) A step-by-step approach to using

PAGE 155

143 SAS for Factor Analysis and Structural Equation Modeli ng. Cary, NC: SAS Institute Inc. Pan American Health Organization, (1998). Health co nditions in the Americas Vol. II. Belize., Washington, D.C. Retrieved June 16, 2005, from http://www.blz.paho.org/ Pan American Health Organization, (2002). Health co nditions in the Americas Vol. II. Belize., Washington, D.C. Retrieved June 16, 2005, from http://www.blz.paho.org/ Pan American Health Organization, (2004a). Road Safety is no accident. Retrieved June 20, 2005 from http://www.paho.org/ Pan American Health Organization, (2004b). Minister Hyde coordinates with PAHO/WHO to Promote Road Safety. Retrieved June 16, 2004, from http://www.blz.paho.org/ Parker, D (2002). Changing drivers’ attitudes to sp eeding: Using the theory of planned behavior. In D. Rutter & L. Quine (Eds.). Changing health behavior: Intervention and research with social cognition models. (pp. 138152). Open University Press. Peck, R., & Voas, R.B. (2002). Forfeiture programs in California: Why so few? Journal of Safety Research 33 (2), 245-258. Peden, M., McGee, K., & Sharma, G., (2002). The inj ury chart book: a graphical overview of the global burden of injuries. Geneva, World Health Organization Retrieved September 15, 2004, from http://whqlibdoc.who.int/publications/924156220X.pd f

PAGE 156

144 Pinksky, I., Labouvie, E., Pandina, R., & Laranjeir a, R. (2001). Drinking and driving: pre-driving attitudes and perceptions among Brazili an youth. Drug and Alcohol Dependence, 62 (3), 231-237. Posada, J., Ben-Michael, E., Herman, A., Kahan, E., & Richter, E., (2000). Death and injury from motor vehicle crashes in Colombia. Pan American Journal of Public Health, 7 (2), 88-91. Pyrczak, F., & Bruce, R.R. (2000). Writing empirical research reports (3rd ed.). Los Angeles, CA: Pyrczak Publishing. Retting, R. A., Ferguson, S. A., & McCartt, A. T. ( 2003). A review of evidence-based traffic engineering measures designed to reduce ped estrian-motor vehicle crashes. American Journal of Public Health, 93 (9), 1456-1463. Rice, T. M., Peek-Asa, C., & Kraus, J. F. (2004). E ffects of the California graduated driver licensing program. Journal of Safety Research, 35 (4), 375-381. Rivara, F. P., Thompson, D. C., & Cummings, P. (199 9). Effectiveness of primary and secondary enforced seat belt laws. American Journal of Preventive Medicine, 16 (1S), 30-39. Rodrguez, D. Y., Fernndez, F. J., & Velsquez, H. A. (2003). Road traffic injuries in Colombia. Injury Control and Safety Promotion, 10 (1-2), 29-35. Rundmo, T., & Iversen, H. (2004). Risk perception a nd driving behavior among adolescents in two Norwegian counties before and af ter a traffic safety campaign. Safety Science, 42 (1), 1-21.

PAGE 157

145 Rutter, D. & Quine, L. (2002). Social cognition mod els and changing health behaviors. In D. Rutter & L. Quine (Eds.). Changing health behavior: Intervention and research with social cognition models. (pp. 1-27). Open University Press. Sagberg, F., & Bjrnskau, T. (2006). Hazard percept ion and driving experience among novice drivers. Accident Analysis and Prevention, 38 (2), 407-414. Schaie, K. W. (1983). The Seattle Longitudinal stud y: A 21-year exploration of psychometric intelligence in Adulthood. In K. W. Sc haie (Ed.), Longitudinal studies of adult psychological development (pp. 64-135). New York: Guilford Press. Schwarzer, R. (1999). Self-regulatory processes in the adoption and maintenance of health behavior: Role of optimism, goals, and threa ts. Health Psychology, 4 (2), 129-148. Shanahan, M. J., Porfeli, E. J., Mortimer, J. T., & Erickson, L. D. (2005). Subjective age identity and the transition to adulthood: When do a dolescents become adults? In R. A. Sattersen Jr., F.F. Furstenberg Jr., & R. G. Rumbaut (Eds.), On the frontier of adulthood: Theory, research, and public policy. (pp. 225255). Chicago & London: The University of Chicago Press. Shepherd, J. P. (2001). Criminal deterrence as a pu blic health strategy. Lancet 358 (9294), 1717-1722. Shults, R. A., Elder, R. W., Sleet, D. A., et al. ( 2001). Reviews of evidence regarding interventions to reduce alcohol-impaired driving. American Journal of Preventive Medicine, 21 (4S), 66-88.

PAGE 158

146 Silvi, J. (2004). Deaths from motor vehicle traffic accidents in selected countries of the Americas Epidemiological Bulletin of the Pan American Health Organization, 25 (1), 2-5. Retrieved September 15, 2004, from http://www.paho.org/english/dd/ais/EB_v25n1.pdf Sivak, M., Soler, J., Trnkle, U., & Spagnhol, J. M (1986). Cross-cultural differences in driver risk-perception. Accident Analysis and Prevention, 18 (4), 355-362. Sivak, M., Soler, J., & Trnkle, U. (1989). Cross-c ultural differences in driver risktaking. Accident Analysis and Prevention, 21 (4), 363-369. Smith, K. (1993). Figuring out the accidents. Surveyor, 179 (3), 14-15. Soderlund, N., & Zwi, A. B. (1995). Traffic-related mortality in industrialized and less developed countries. Bulletin of the World Health Organization, 73 (2), 175-182. St. Bernard, G., & Matthews, W. (2003). A contempor ary analysis of road traffic, crashes, fatalities, and injuries in Trinidad and T obago. Injury Control and Safety Promotion, 10 (1-2), 21-27. Svenson, O. (1978). Risks of road transportation in a psychological Perspective. Accident Analysis and Prevention, 10 267-280. Svenson, O., Fischhoff, B., & MacGregor, D. (1985). Perceived driving safety and seatbelt usage. Accident Analysis and Prevention, 12 (2), 119-133. Task Force on Community Preventive Services. (2001) Recommendations to reduce injuries to motor vehicle occupants: Increasing chi ld safety seat use, increasing safety belt use, and reducing alcohol-impaired driv ing. American Journal of Preventive Medicine, 21 (4 Suppl), 16-22.

PAGE 159

147 Tomarken, A. J., & Waller, N. G. (2003). Potential problems with “well fitting” models. Journal of Abnormal Psychology. 112, 578-598. Trnkle, U., Gelau, C., & Metker, T. (1990). Risk p erception and age-specific accidents of young drivers. Accident Analysis and Prevention, 22 (2), 119-125. Trifiletti, L.B. (2003). The utility of the precaution adoption process mode l for understanding safety practices of families who use pediatric emergency department services. Doctoral dissertation, The Johns Hopkins Bloomberg School of Public Health. Trifiletti, L.B., Gielen, D.A., Sleet, D.A. & Hopki ns, K. (2005). Behavioral and social sciences theories and models: Are they used in unin tentional injury prevention research? Health Education Research, 20 (3), 298-307. Ulleberg, P. (2001). Personality subtypes of young drivers. Relationship to risk-taking preferences, accident involvement, and response to a traffic safety campaign. Transportation Research Part F: Traffic Psychology and Behaviour, 4 (4), 279297. Ulleberg, P. & Rundmo, T. (2002). Risk-taking attit udes among young drivers: The psychometric qualities and dimensionality of an ins trument to measure young drivers' risk-taking attitudes. Scandinavian Journal of Psychology, 43 (3), 227237. Ulleberg, P., & Rundmo, T. (2003). Personality, att itudes and risk perception as predictors of risky driving behavior among young dr ivers. Safety Science, 41 427-443.

PAGE 160

148 Van Houten, R., & Retting, R. A. (2001). Increasing motorist compliance and caution at stop signs. Journal of Applied Behavior Analysis, 34 (2), 185-193. Vanlaar, W., & Yannis, G. (2005). Perception of roa d accidents causes. Accident Analysis and Prevention. Article in Press. Retrieved October 10, 2005, from http://www.sciencedirect.com.proxy.usf.edu/ Weinstein, N. D. (1980). Unrealistic optimism about future life events. Journal of Personality and Social Psychology, 39 (5), 806-820. Weinstein, N. D. (1987). Unrealistic optimism about susceptibility to health problems: Conclusions from a community-wide sample. Journal of Behavioral Medicine, 10 (5), 481-500. Weinstein, N. D. (1988). The precaution adoption pr ocess. Health Psychology, 7 (4), 355-386. Weinstein, N. D. (1989). Effects of personal experi ence on self-protective behavior. Psychological Bulletin, 105 (1), 31-50. Weinstein, N. D. (1998). Accuracy of smokers’ risk perceptions. Annals of Behavioral Medicine: a publication of the Society of Behaviora l Medicine, 20 (2), 135-140. Weinstein, N.D. (2003). Exploring the links between risk perceptions and preventive health behavior. In J. Suls & K. Wallston (Eds.), Social psychological foundations of health and illness (pp. 22 53). Malden, MA: Blackwell. Weinstein, N. D., Klotz, M.L., & Sandman, P. M. (19 88). Optimistic biases in public perceptions of the risk from radon. American Journal of Public Health, 78 (7), 796-800.

PAGE 161

149 Weinstein, N. D., Rothman, A. J., & Sutton, S. R. ( 1998). Stage theories of health behavior: conceptual and methodological issues. Health Psychology, 17 (3), 290299. Weinstein, N. D., Sandman, P. M., & Roberts, N. E. (1990). Determinants of selfprotective behavior: Home radon testing. Journal of Applied Social Psychology, 20 (10), 783-801. Weinstein, N. D., & Sandman, P. M. (1992). A model of the precaution adoption process: Evidence from home radon testing. Health Psychology, 11 (3), 170-180. Weinstein, N. D., & Sandman, P. M. (2002). The prec aution adoption process model and its application. In R. J. DiClemente, R. A. Crosby, & M. C. Kegler (Eds.). Emerging theories in health promotion practice and research: Strategies for improving public health. (pp. 16-39). San Francisco: Jossey-Bass. West, R., & Hall, J. (1997). The role of personalit y and attitudes in traffic accident risk. Applied Psychology: An International Review, 46 (3), 253-264. Whissell, R. W., & Bigelow, B. J. (2003). The speed ing attitude scale and the role of sensation seeking in profiling young drivers at ris k. Risk Analysis, 23 (4), 811820 Whiting, J.D., Dunn, K., March, J.A., & Brown, L.H. (1998). EMT knowledge of ambulance traffic laws. Prehospital Emergency Care, 2 (2), 136-140 Williams, A. F. (2003). Views of U.S. drivers about driving safety. Journal of Safety Research, 34 (5), 491-494. World Health Organization (2004a). Injuries and violence prevention databases. Retrieved September 15, 2004, from

PAGE 162

150 http://www.who.int/violence_injury_prevention/surve illance/databases/injurydata/ en/ World Health Organization (2004b). World report on road traffic injury prevention Retrieved September 15, 2004, from http://www.who.int/world-healthday/2004/infomaterials/world_report/en/index.html World Health Organization (2004c). Injury: A leading cause of global burden of disease 2000 Retrieved June 20, 2005, from http://whqlibdoc.who.int/publications/2002/92415623 23.pdf World Health Organization (2004d). Guidelines for conducting community surveys on injuries and violence Retrieved September 15, 2004, from http://www.who.int/world-healthday/2004/infomaterials/world_report/en/index.html Yu, C.-Y. (2002). Evaluating cutoff criteria of model fit indices for latent variable models with binary and continuous outcomes. Doctoral dissertation, Univesity of California, Los Angeles. Zhang, W., Huang, Y. H., Roetting, M., Wang, Y., & Wei, H. (2006). Driver's views and behaviors about safety in china--what do they NOT k now about driving? Accident; Analysis and Prevention, 38 (1), 22-27.

PAGE 163

151 Appendices

PAGE 164

152 Appendix A: IRB Approval Letter Ismael Hoare, M.P.H. and Robert McDermott, Ph.D. Community and Family Health P.O. Box 48034 Tampa, Fl 33647 RE: Exempt Certification for Application for Exemption IRB#: 104876 Title: Attitudinal Factors Related to Driving Behaviors of Young Adults in Belize: An Application of the Precaution Adoption Process M odel Dear Mr. Hoare and Dr. McDermott: On July 26, 2006, the Institutional Review Board (I RB) determined that your Application for Exemption MEETS FEDERAL EXEMPTION CRITERIA two (2). PLEASE NOTE: If revisions are made to the study instrumen t as a result of the pilot testing, the finalized document must be submitted to the IRB via a Modification Request prior to implementation. It is your responsibility to ensure that this resea rch is conducted in a manner consistent with the ethical p rinciples outlined in the Belmont Report and in compliance with USF IRB policies and procedures. Please note that changes to this protocol may disqu alify it from exempt status. It is your responsibility to notify the IRB prior to implement ing any changes. The Division of Research Compliance will hold your exemption application for a period of five years from the date of this letter or until a Final Review Report is received. If you wish to continue this protocol beyond the five-year exempt certification period, you will need to submit an Exemption Certification Request form at least 30 days before this exempt certification expires. The IRB will send yo u a reminder notice prior to expiration of the certification; therefore, it is important th at you keep your contact information current. Should you complete this study prior to t he end of the five-year period, you must submit an Application for Final Review Please reference the above IRB protocol number in a ll correspondence to the IRB or the Division of Research Compliance. In addition, we have enclosed an Institutional Review Board (IRB) Quick Reference Guide providing guidelines and resources to assist you in meeting your responsibilities when conductin g human subjects research. Please read this guide carefully.

PAGE 165

153 Appendix A: Continued We appreciate your dedication to the ethical conduc t of human subject research at the University of South Florida and your continued comm itment to the Human Research Protections Program. If you have any questions reg arding this matter, please call 813974-9343. Sincerely, Paul G. Stiles, J.D., Ph.D. USF Institutional Review Board Enclosures: IRB Quick Reference Guide Cc: Brenda Kuska, USF IRB Professional Staff IA-EC-05-01

PAGE 166

154 Appendix B: IRB Modification Approval Letter

PAGE 167

155 Appendix B: Continued

PAGE 168

156 Appendix C: Permission Letter Sacred Heart Junior C ollege Appendix

PAGE 169

157 Appendix D: Permission Letter University of Belize

PAGE 170

158 Appendix E: Driving Behavior Survey First Draft Attitudinal Factors Related to Driving Behaviors of Young Adults in Belize: An Application of the Precaution Adoption Process Mode l. By Ismael Hoare, M.P.H. INSTRUCTIONS Your class has been selected to participate in a di ssertation research study on risky driving behaviors, risk-taking attitudes, risk perc eption and knowledge of Belizean road law and signs among 18-24 University of Belize stud ents. The survey is completely voluntary. You may choose to leave certain questions unanswered and you may stop at any time. The survey does not require you to put your name or provide any information that may reveal you r identity. Your responses will be kept strictly confidential a nd available only to the researchers. Thank you for filling out this survey!

PAGE 171

159 Appendix E: Continued Section I. Risky Driving Behavior Questions In the past year, how often if ever do you do the f ollowing activities while driving. Please circle your choice. Speeding 1 How often do you exceed the speed limit of 25 mph in the villages, towns or cities (by more than 10 mph)? 1 Often 2 Sometimes 3 Rarely 4 Never 2 How often do you exceed the speed limit on the hi ghway (by more than 10 mph)? 1 Often 2 Sometimes 3 Rarely 4 Never 3 How often do you drive fast to show others that y ou can speed and still keep the car under control? 1 Often 2 Sometimes 3 Rarely 4 Never 4 How often do you drive fast to show off to passen gers in the car? 1 Often 2 Sometimes 3 Rarely 4 Never 5 How often do you worry that you will be caught sp eeding? 1 Often 2 Sometimes 3 Rarely 4 Never 6 How often have you raced another driver on the hi ghway? 1 Often 2 Sometimes 3 Rarely 4 Never Distracted Driving This next section is about behaviors that drivers may do while driving. 7 How often, if ever, do you talk to other passenge rs while driving? 1 Often 2 Sometimes 3 Rarely 4 Never 8 How often, if ever, do you read, such as a book, newspaper, mail, or notes while driving? 1 Often 2 Sometimes 3 Rarely 4 Never 9 How often, if ever, do you eat or drink while dri ving? 1 Often 2 Sometimes 3 Rarely 4 Never 10 How often, if ever, do you deal with children in the back seat while driving? 1 Often 2 Sometimes 3 Rarely 4 Never 11 How often, if ever, do you talk on a cellular ph one while driving? 1 Often 2 Sometimes 3 Rarely 4 Never

PAGE 172

160 Appendix E: Continued 12 How often, if ever, do you do personal grooming (such as, combing hair, shaving, putting on makeup) while driving? 1 Often 2 Sometimes 3 Rarely 4 Never 13 How often, if ever, do you change radio stations CDs or tapes while driving? 1 Often 2 Sometimes 3 Rarely 4 Never 14 How often, if ever, do you use a PDA, such as a Palm Pilot, while driving? 1 Often 2 Sometimes 3 Rarely 4 Never Aggressive Driving 15 How often, if ever, do you cut in front of anoth er driver? 1 Often 2 Sometimes 3 Rarely 4 Never 16 How often, if ever, do you use the shoulders on the road to pass traffic? 1 Often 2 Sometimes 3 Rarely 4 Never 17 How often, if ever, do you make an angry, insult ing or obscene gesture or comment toward another driver such that they hear o r see it? 1 Often 2 Sometimes 3 Rarely 4 Never 18 How often, if ever, do you pass a vehicle on a c urve? 1 Often 2 Sometimes 3 Rarely 4 Never 19 How often, if ever, do you pass a vehicle on a h ill? 1 Often 2 Sometimes 3 Rarely 4 Never 20 How often, if ever, do you pass a bus letting of f passengers without slowing down? 1 Often 2 Sometimes 3 Rarely 4 Never 21 How often, if ever, do you tailgate another vehi cle? 1 Often 2 Sometimes 3 Rarely 4 Never Not adhering to traffic laws 22 How often, if ever, do you drive through an inte rsection without stopping? 1 Often 2 Sometimes 3 Rarely 4 Never 23 How often, if ever, do you slow down, but not st op completely at a stop sign? 1 Often 2 Sometimes 3 Rarely 4 Never 24 How often do you ignore traffic laws to get ahea d in traffic? 1 Often 2 Sometimes 3 Rarely 4 Never

PAGE 173

161 Appendix E: Continued 25 How often do you break traffic laws due to peer pressure? 1 Often 2 Sometimes 3 Rarely 4 Never 26 How often do you drive the wrong way down a oneway street? 1 Often 2 Sometimes 3 Rarely 4 Never 27 How often do you yield to pedestrians at a pedes trian crossing? 1 Often 2 Sometimes 3 Rarely 4 Never 28 How often do you reverse your vehicle the wrong way down a one-way street? 1 Often 2 Sometimes 3 Rarely 4 Never 29 How often do you reverse your vehicle the on the highway? 1 Often 2 Sometimes 3 Rarely 4 Never Section II. Precaution Adoption Process Model Quest ions Read each item carefully and respond by circling on e of the following. If you are not sure about an answer, do not leave the item blank, but c hoose the best possible response. 30 Have you ever heard about young people being hurt i n motor vehicle crashes? 1 Yes 2 No 3 Don’t Know 31 Have you ever heard about the need to use seatbe lts to prevent injury from motor vehicle crashes? 1 Yes 2 No 3 Don’t Know 32 Do you ever travel in a car (your car, someone e lse’s car, or taxi)? 1 Yes 2 No 3 Don’t Know 33 Do you travel in….. Your car most often Someone else’s car most often Taxi most often Don’t know

PAGE 174

162 Appendix E: Continued 34 When you travel in a car, or taxi, do you use a sea tbelt? 1 Yes (Skip to question 36) 2 No 3 Don’t Know 35 Have you thought of using a seat belt? 1 Yes 2 No 3 Don’t Know 36 How often do you use a seat belt when you travel in your car, someone else’s car or taxi? Would you say….. 1 Never 2 Almost never 3 Sometimes 4 Usually 5 Almost always 6 Always 37 Do you plan to use a seat belt more often? 1 Yes 2 No 3 Don’t Know Section III. Risk-taking Attitudes Read each item carefully and respond by circling on e of the following. If you are not sure about an answer, do not leave the item blank, but g ive a response as close as possible to how you really feel. Riding with an unsafe driver 38 I would get in the car with a driver who has bee n drinking if I knew and trusted him or her. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 39 I would get into my friend's car even though she /he is known to be an unsafe driver. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 40 I would get into the car with a careless driver if I had no other way to get home. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree

PAGE 175

163 Appendix E: Continued 41 I would ask my friend to let me out of the car i mmediately if she/he drove recklessly. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 42 I would not even consider riding with a drunk pe rson. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 43 I might get in the car with a driver who has bee n drinking. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 44 I would rather walk a hundred miles than get int o a car with an unsafe driver. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree Speeding 45 It’s alright to race when driving. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 46 If you have good driving skills, speeding is O.K 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 47 I usually (or will usually) drive faster when my friends are in the car. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 48 It is fun to drive fast. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 49 Driving 5 or 10 miles above the speed limit is O .K. because everyone does it. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 50 I think it is O.K. to speed if traffic condition s allow you to do so. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree

PAGE 176

164 Appendix E: Continued 51 I like (or will like) to show off my skill by dr iving fast. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree Concern for others 52 It makes me feel good when I am courteous to oth er drivers. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 53 Hurting someone else with my car would scar me f or life. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 54 I don't think of others because if I did that's when I would get into a crash. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 55 I couldn't live with myself if I hurt another hu man being. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 56 It usually doesn't pay to be concerned about oth ers because most others aren't concerned about me. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 57 If I cause a motor vehicle crash because of stup idity, I hope I'm the one who gets hurt. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 58 I obey (or would obey) all laws when kids are wi th me because I want them to grow up to be safe drivers. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 59 It is inexcusable to take a human or animal life because of one's carelessness. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree

PAGE 177

165 Appendix E: Continued 60 I hope I never get into a crash in which someone is hurt and it is my fault. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 61 If I hurt someone because of my driving, I never want to drive again. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree Drinking and driving 62 I don't need anybody to tell me when they think I've had enough to drink. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 63 If you have just one or two beers while driving, it's O.K. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 64 It's O.K. to drive if you have one or two drinks and you feel in control. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 65 Drunk driving is unlawful and whoever doesn't ob ey this law should be punished severely. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 66 Even though I know it can be dangerous to drink and drive, I would do so anyway in most cases. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 67 If I had a lot on my mind, a drink or two would help me get my head together before driving. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree 68 I think they exaggerate the risk of getting into a crash due to drinking and driving. 1 Strongly agree 2 Agree 3 Disagree 4 Strongly disagree

PAGE 178

166 Appendix E: Continued Section IV. Risk Perception Questions Read each item carefully and respond by circling on e of the following. If you are not sure about an answer, do not leave the item blank, but g ive a response as close as possible to how you really feel. Emotion-based risk perception 69 How often are you feeling unsafe that you could be injured in a traffic crash? 1 Often 2 Sometimes 3 Rarely 4 Never 70 How often are you worried that you could be inju red in a traffic crash? 1 Often 2 Sometimes 3 Rarely 4 Never 71 How often are you feeling unsafe that a young ad ult could be injured in a traffic crash? 1 Often 2 Sometimes 3 Rarely 4 Never 72 How often are you worried that a young adult cou ld be injured in a traffic crash? 1 Often 2 Sometimes 3 Rarely 4 Never Cognition-based risk perception 73 How probable do you think it is for a young adul t to be injured in a traffic crash? 1 Very probable 2 Probable 3 Somewhat probable 4 N ot probable 74 How probable do you think it is that you could b e injured in a traffic crash? 1 Very probable 2 Probable 3 Somewhat probable 4 N ot probable Concern 75 How concerned are you about traffic risks in gen eral? 1 Very worried 2 Worried 3 Somewhat worried 4 Not worried 76 How concerned are you about traffic risks for yo ung adults in general? 1 Very worried 2 Worried 3 Somewhat worried 4 Not worried 77 How concerned are you that a young adult could b e injured in a traffic crash? 1 Very worried 2 Worried 3 Somewhat worried 4 Not worried 78 How concerned are you that you could be injured in a traffic crash? 1 Very worried 2 Worried 3 Somewhat worried 4 Not worried

PAGE 179

167 Appendix E: Continued Section V. Knowledge of road laws Read each item carefully and respond by circling on e of the following. If you are not sure about an answer, do not leave the item blank, but c hoose the best possible response. 79 On what side of vehicle should a driver keep whe n passing another vehicle? 1 right side 2 left side 80 What should a driver do before passing another v ehicle? 1 ensure that no pedestrian is crossing the road 2 ensure that there is no traffic coming from the o pposite direction 3 signal to the other driver 4 all of the above 81 What should a driver or rider do before he/she p roceeds into a major road from a minor road? 1 continue driving at the same speed 2 slow down 3 come to a complete stop 4 come to a complete stop and yield to the driver on the major road 82 On what side of the road should a driver pick up and drop off passengers? 1 left side 2 right side Knowledge of Road Signs 83 This road sign means 1 yield to pedestrians 2 yield to the vehicle in front of you 3 yield to vehicles on the main road 4 both 1 and 3

PAGE 180

168 Appendix E: Continued 84 This road sign means 1 slow down before proceeding 2 yield to the other driver before proceeding 3 come to a complete stop 85 This road sign means 1 winding road, do not pass 2 keep straight ahead 3 divided highway ahead 86 This road sign means 1 road curves at left 2 road slippery when wet 3 vehicle ahead is speeding Section VI. Socio-demographic Questions The following questions are for statistical purpose s. Age 87 How old are you? _____ years Ethnicity 88 Circle the ethnic group in which you belong. 1 Chinese 5 Maya 2 Creole 6 Mennonite 3 East Indian 7 Mestizo/Spanish 4 Garifuna 8 Other

PAGE 181

169 Appendix E: Continued Sex 89 What is your gender? Male_____ Female _____ Education 90 What is the highest diploma or degree that you h ave earned? 1 High school diploma 2 Sixth form degree Enrolment status 91 Are you a full-time or part-time student? 1 full-time student 2 part-time student 92 Do you drive? 1 Yes 2 No If yes, how often? 1 Once a month or less 2 Two or three times a week 3 Four or five times a week 4 Daily 93 Do you have a valid driver’s license? 1 Yes 2 No 3 Don’t Know 94 How many years have you been driving? __________ ___________ 95 In which district do you live? _________________ ___ Crash Experience 96 In the past year, have you been in a crash? 1 Yes 2 No 3 Don’t Know 97 How many times has this happened to you in the p ast year? _______________

PAGE 182

170 Appendix E: Continued 98 Were you driving or was someone else driving? 1 I was driving 2 Someone else was driving 3 Don’t Know 99 Was anyone injured in the crash? 1 Yes 2 No 3 Don’t Know

PAGE 183

171 Appendix F Driving Behavior Survey Final Draft Driving Behavior Survey FINAL DRAFT INSTRUCTIONS This survey is about driving behavior. It has been developed so you can tell us about how you drive, your attitudes about d riving, and your understanding of traffic laws. The information you give us will be used to develop better defensive driving programs for young drivers. DO NOT write your name on this survey. The answers you give will be kept private. No one will know what your write. Please answer the questions based on what you really do and how you really feel Your honest answers will help us make better decisions. Completing this survey is voluntary. If you are no t comfortable answering a question, just leave it blank. The questions that ask about your background will b e used only to describe the types of students completing the surve y. Please do not write your name or provide any information about yourself on the survey. Make sure you read every question and the instructi ons to all sections. When you are finished, follow the instructions of t he person giving you the survey. Thanks for your help.

PAGE 184

172 Appendix F: Continued Section A. For each question, please circle the response that best applies to you. 1 Almost never (0-20% of the time) 2 Seldom (21-40% of the time) 3 Sometimes (41-60% of the time) 4 Often (61-80% of the time) 5 Almost always (81-100% of the time) During the past year, how often, did you do the fol lowing activities while driving. Question Almost Never Seldom Sometimes Often Almost Always A1. How often do you exceed the speed limit while driving within villages or towns or city limits? 1 2 3 4 5 A2. How often do you exceed the speed limit while driving on the highway? 1 2 3 4 5 A3. How often do you drive fast to show off? 1 2 3 4 5 A4. How often do you worry that you will be caught speeding? 1 2 3 4 5 A5. How often have you raced another driver on the highway? 1 2 3 4 5 A6. How often do you talk to other passengers while driving? 1 2 3 4 5 A7. How often do you read (such as a book, newspaper, mail, or notes) while driving? 1 2 3 4 5 A8. How often do you eat or drink while driving? 1 2 3 4 5 A9 How often do you talk on the phone while driving? 1 2 3 4 5 A10 How often do you do personal grooming (such as, combing hair, shaving, putting on makeup) while driving? 1 2 3 4 5

PAGE 185

173 Appendix F: Continued Question Almost Never Seldom Sometimes Often Almost Always A11 How often do you change radio stations, CDs or tapes while driving? 1 2 3 4 5 A12 How often do you cut in front of other drivers? 1 2 3 4 5 A13 How often do you pass on the right of the road to overtake traffic? 1 2 3 4 5 A14 How often do you make angry, insulting or obscene gestures toward other drivers that they see? 1 2 3 4 5 A15 How often do you make angry, insulting, or obscene statements to other drivers that they hear? 1 2 3 4 5 A16 How often do you pass vehicles on a curve? 1 2 3 4 5 A17 How often do you pass vehicles on a hill? 1 2 3 4 5 A18 How often do you pass buses letting off passengers without slowing down? 1 2 3 4 5 A19 How often do you tailgate other vehicles? 1 2 3 4 5 A20 How often do you drive through intersections without slowing down? 1 2 3 4 5 A21 How often do you cruise through stop signs? 1 2 3 4 5 A22 How often do you ignore traffic laws to get ahead in traffic? 1 2 3 4 5 A23 How often do you break traffic laws because of peer pressure? 1 2 3 4 5

PAGE 186

174 Appendix F: Continued A24 How often do you drive the wrong way down one-way streets? 1 2 3 4 5 A25 How often do you stop for pedestrians in a pedestrian crossing? 1 2 3 4 5 A26 How often do you use your turning signal indicator when making right and left turns? 1 2 3 4 5 A27 How often do you blow your horn when you are upset at the driving behaviors of other drivers? 1 2 3 4 5

PAGE 187

175 Appendix F: Continued Section B. For each question, please circle the response that best applies to you. B1 During the past 12 months, have you ever heard a bout a young person being hurt in a motor vehicle crash? 1 Yes 2 No 3 Don’t Know B2 During the past 12 months, have you ever heard a bout the importance of using seat belts to prevent injury resulting from m otor vehicle crash? 1 Yes 2 No 3 Don’t Know B3 During the past 30 days, have you traveled in a car? 1 Yes 2 No 3 Don’t Know B4 When you traveled in a car during the past 30 da ys, how did you travel? 1 I did not travel in a car during the past 30 days 2 I usually drove my own car. 3. I usually rode in a car driven by someone else. 4. I usually rode in a taxi B5 During the last 5 times you rode in a car, how m any times did you wear a seat belt? 1 0 times 2 1-2 times 3 3-4 times 4 5 times

PAGE 188

176 Appendix F: Continued B6 When you ride in a car driven by someone, how of ten do you wear a seat belt? 1 Never 2 Seldom 3 Sometimes 4 Often 5 Always B7 The next time you drive a car, do you think you will wear a seat belt? 1 Yes 2 No 3 Don’t Know B8 The next time you ride in a car driven by someon e else, do you think you will use a seat belt? 1 Yes 2 No 3 Don’t Know

PAGE 189

177 Appendix F: Continued Section C. For each statement, please circle the response that best applies to you. Statement Strongly Agree Agree Disagree Strongly Disagree C1 I would ride in a car or other vehicle driven by someone who had been drinking alcohol. 1 2 3 4 C2 It is fun to drive fast. 1 2 3 4 C3 I feel good when I am courteous to other drivers. 1 2 3 4 C4 The risk of crashing a car after drinking alcohol is exaggerated. 1 2 3 4 C5 I am concerned about the safety of others when I drive. 1 2 3 4 C6 I would ride in a car driven by someone I did not know. 1 2 3 4 C7 I am a better driver after drinking one or two alcoholic drinks. 1 2 3 4 C8 It is okay to race when driving. 1 2 3 4 C9 I would rather stay where I was than get into a car with an unsafe driver. 1 2 3 4 C10 I would never drink alcohol and drive. 1 2 3 4 C11 If I injured someone because of my driving, I will never drive again. 1 2 3 4 C12 It is okay to drive above the speed limit. 1 2 3 4 C13 I would feel guilty if one of my passengers was injured in a car accident when I was driving. 1 2 3 4 C14 People who drink alcohol and drive should be punished. 1 2 3 4 C15 Having one or two beers before driving, is no big deal. 1 2 3 4 C16 I would ask to be let out of car driven recklessly by a friend. 1 2 3 4

PAGE 190

178 Appendix F: Continued Section D. For each statement, please circle the response that best applies to you. Statement Strongly Agree Agree Disagree Strongly Disagree D1 I am concerned about being in a car accident. 1 2 3 4 D2 I am concerned about being injured in a car accident. 1 2 3 4 D3 I am concerned about how I drive. 1 2 3 4 D4 I will likely be injured in a car accident sometime during my life. 1 2 3 4 D5 I am concerned about people my age being in a car accident. 1 2 3 4 D6 I am concerned about people my age being injured in a car accident. 1 2 3 4 D7 I am concerned about how people my age drive. 1 2 3 4 D8 It is likely that most people my age will be injured in a car accident sometime during their life. 1 2 3 4 D9 Most people my age are concerned about being in a car accident 1 2 3 4 D10 Most people my age are concerned about being injured in a car accident. 1 2 3 4 D11 Most people my age are concerned about how they drive. 1 2 3 4 D12 Most people my age believe that it is likely that they will be injured in a car accident sometime during their life. 1 2 3 4

PAGE 191

179 Appendix F: Continued Section E. For each question, please circle the most correct a nswer. E1 At the junction of two equal roads the driver of a vehicle should yield to the vehicle on his left. 1 True 2 False 3 Don’t Know E2 In any city, town or village, the speed limit fo r vehicles except trucks, bus and tractors is 45 MPH. 1 True 2 False 3 Don’t Know E3 When approaching other vehicles at night, a driv er should use low beam light only. 1 True 2 False 3 Don’t Know E4 Before moving forward onto a roadway, a driver s hould look to his/her right to see if his/her lane is clear of all traffi c. 1 True 2 False 3 Don’t Know E5 Drivers must yield to pedestrians standing by to enter a pedestrian crosswalk.. 1 True 2 False 3 Don’t Know

PAGE 192

180 Appendix F: Continued E6 When approaching a curve a driver should dip the headlights to alert oncoming traffic of its presence. 1 True 2 False 3 Don’t Know E7 A driver must ensure that there is no traffic co ming from the opposite direction before passing another vehicle. 1 True 2 False 3 Don’t Know E8 Which side of the road should a driver pass anot her vehicle? 1 Right side 2 Left side 3 Don’t Know E9 Before proceeding onto a major road from a minor road, which of the following should a driver do? 1 Continue driving at the same speed 2 Slow down 3 Come to a complete stop 4 Come to a complete stop and yield to the driver on the major road 5 Don’t Know E10 Which side of the road should a driver pick up and drop off passengers? 1 Left side 2 Right side 3 Don’t know

PAGE 193

181 Appendix F: Continued E11 What does this road sign mean? 1 Yield to pedestrians 2 Yield to the vehicle in front of you 3 Yield to vehicles on the main road 4 don’t know E12 What does this road sign means? 1 Slow down before proceeding 2 Yield to the other driver before proceeding 3 Come to a complete stop before proceeding 4 Don’t know E13 What does this road sign mean? 1 Winding road, do not pass 2 Keep straight ahead 3 Divided highway ahead 4 Don’t know E14 What does this road sign mean? 1 Road curves at left 2 Road slippery when wet 3 Vehicle ahead is speeding 4 Don’t know

PAGE 194

182 Appendix F: Continued Section F. Read each item carefully and fill in the blank spac e or circle the appropriate response. F1 How old are you? _____ years F2 What is your gender? Male_____ Female _____ F3 How do you describe yourself? 1 Chinese 5 Maya 2 Creole 6 Mennonite 3 East Indian 7 Mestizo/Spanish 4 Garifuna 8 Other F4 Are you a full-time or part-time student? 1 full-time student 2 part-time student F5 Do you drive? 1 Yes 2 No F6 Do you currently have a valid driver’s license? 1 Yes 2 No 3 Don’t Know F7 How many years have you been driving e.g. car, p ickup, motorcycle or other? _____ years F8 In which district do you live? 1 Corozal 4 Cayo 2 Orange Walk 5 Stann Creek 3 Belize 6 Toledo

PAGE 195

183 Appendix F: Continued F9 During the past 12 months, how many times were y ou in a car accident while driving? 1 0 2 1 time 3 2 times 4 3 times 5 4 or more times F10 During the past 12 months, how many times were you in a car accident when someone else was driving. 1 0 2 1 time 3 2 times 4 3 times 5 4 or more times F11 During the past 12 months, how many times were you in a car accident in which you were injured and had to be treated by a d octor or nurse? 1 0 2 1 time 3 2 times 4 3 times 5 4 or more times F12 During the past 12 months, how many times were you in a car accident in which someone else was injured and had to be treate d by a doctor or nurse? 1 0 2 1 time 3 2 times 4 3 times 5 4 or more times F13 Overall, how would you rate your driving skills ? 1 Fair 2 Good 3 Very Good 4 Excellent

PAGE 196

184 Appendix F: Continued THANK YOU FOR PARTICIPATING.

PAGE 197

185 Appendix G External Panel List Name Abbreviated Research Interest Dr. Julie Baldwin* Research focused on developing, implementing, and evaluating for adolescents and young adults Dr. Niki Harr* Design and evaluation of injury pre vention interventions; youth driving attitudes Dr. Robert McDermott* Dr. Dale O. Ritzel* Dr. Robert M. Weiler* Adolescent health; planning a nd evaluation Dr. Neil Weinstein Health psychology; Risk percepti ons, health-protective behavior Dr. Charles Basch Health education program planning and evaluation Dr. Brian Jonah Road Safety Programs Dr. Steve Brown Health psychology; risk perception Dr. Daniel V. McGehee Human factors design, test an d evaluation Hilde Iversen Risk and safety research Agreed to participate

PAGE 198

186 Appendix H: External panel review guide Evaluation Guide Please review the questionnaire based on the follow ing guide. 1. Does the survey appear to measure Risk-taking Attit udes, Risk Perceptions, Knowledge of road laws and road signs of Belize and Risky Driving Behaviors? 2. Does the instrument appear to be appropriate for 18 to 24 year old students? 3. Are there questions that are redundant? 4. Are the response options appropriate? 5. Are there any other questions you would like to add to the questionnaire?

PAGE 199

187 Appendix H: Continued Evaluation Criteria: 2nd Review Instructions: Please provide a summary of your over all assessment of the instrument based on the following criteria. 1. Are the directions clear and concise? 2. Are questions appropriate for this target audience? Are there items that are inappropriate? Are response options types appropriate? 3. Format: Is the survey easy to navigate? 4. Are there any other comments you would like to add?

PAGE 200

188 Appendix I: Research Question Table Research Questions Domain Survey Questions Analysis 1. To what extent are the specific stages of the PAPM observed in the study population? Staging questions B1-B8 Staging Algorithm Research Question Variable Indicators Survey Questions Analysis Socio-demographic Questions F1-F8 Demographic Factors Crash Experience F9-F12 Speeding A1-A5 Distracted Driving A6-A11 Aggressive Driving A12-A14, A16, A18-A20 2. What is the relationship between selected demographic factors and risky driving behavior? Risky Driving Behaviors Not adhering to traffic laws A21-A27 CFA Structural Equation Modeling Research Question Variable Indicators Survey Questions Analysis Riding with an unsafe driver C1, C6, C9, C10, C16 Speeding C2, C8, C12 Concern for others C3, C5, C11, C13 Risk-taking Attitudes Drinking and driving C4, C7, C10, C14, C15 3. What is the relationship between young adults’ risk-taking attitudes and risky driving behaviors? Risky Driving Behaviors Speeding, Distracted Driving, Aggressive Driving, Not adhering to traffic laws A1A27 CFA Structural Equation Modeling

PAGE 201

189 Appendix I: Continued Research Questions Variable Indicators Survey Questions Analysis Knowledge Of Road Laws E1-E6, E8-E10 Knowledge Of Road Laws Knowledge of Road Signs E11-E14 4. What is the relationship between young adults’ knowledge of road laws and risky driving behaviors? Risky Driving Behaviors Speeding, Distracted Driving, Aggressive Driving, Not adhering to traffic laws A1A27 CFA Structural Equation Modeling Research Questions Variable Indicators Survey Questions Analysis Cognition-based Perception D4, D8, D12 Emotion-based Perception D3, D7, D11 Risk Perceptions Concern D1, D9, D10 5. What is the relationship between young adults’ risk perceptions and risky driving behaviors? Risky Driving Behaviors Speeding, Distracted Driving, Aggressive Driving, Not adhering to traffic laws A1A27 CFA Structural Equation Modeling

PAGE 202

190 Appendix J Pilot Test Review Guide Evaluation Criteria: Instructions: Please provide an assessment of the instrument based on the following criteria. 1. Are the directions clear and concise? 2. Are questions appropriate for this target audience? Are there items that are inappropriate? Are response options types appropriate? 3. Format: Is the survey easy to navigate? 4. Are there any other comments you would like to add?

PAGE 203

191 Appendix K SEM Output Mplus VERSION 3.0 MUTHEN & MUTHEN 06/26/2007 1:06 PM INPUT INSTRUCTIONS TITLE: Path Analysis for Attitudinal Factors DATA: FILE IS "F:\Entire scales JUNE 20.sps"; format (17F8.4); VARIABLE: NAMES ARE B1 B2 B3 B4 A1 A2 A3 A4 P1 P2 P3 K1 K 2 Male DrvY Crsh Age; MISSING ARE B1 B2 B3 B4 A1 A2 A3 A4 P1 P2 P3 K1 K2 Male DrvY Crsh Age (99); USEVARIABLES ARE B1 B2 B3 B4 A1 A2 A3 A4 P1 P2 P3 K1 K2 Male DrvY Crsh Age; Analysis: iterations=10000; TYPE=MISSING H1; MODEL: RTA by A1 A2 A3 A4; RP by P1 P2 P3; KLS by K1 K2; RDB by B1 B2 B3 B4; RTA with RP; KLS with RTA; KLS with RP; RP ON Male DrvY Crsh Age; RTA ON Male DrvY Crsh Age; KLS ON Male DrvY Crsh Age; RDB ON RTA RP KLS Male DrvY Crsh Age; OUTPUT: SAMPSTAT MODINDICES (0) STANDARDIZED; INPUT READING TERMINATED NORMALLY Path Analysis for Attitudinal Factors SUMMARY OF ANALYSIS Number of groups 1 Number of observations 532 Number of dependent variables 13

PAGE 204

192 Number of independent variables 4 Number of continuous latent variables 4 Observed dependent variables Continuous B1 B2 B3 B4 A1 A2 A3 A4 P1 P2 P3 K1 K2 Observed independent variables MALE DRVY CRSH AGE Continuous latent variables RTA RP KLS RDB Estimator ML Information matrix OBSERVED Maximum number of iterations 10000 Convergence criterion 0.500D-04 Maximum number of steepest descent iterations 20 Maximum number of iterations for H1 2000 Convergence criterion for H1 0.100D-03 Input data file(s) F:\Entire scales JUNE 20.sps Input data format (17F8.4) SUMMARY OF DATA Number of patterns 48 COVARIANCE COVERAGE OF DATA Minimum covariance coverage value 0.100 PROPORTION OF DATA PRESENT Covariance Coverage B1 B2 B3 B4 A1 ________ ________ ________ ________ ________ B1 0.874 B2 0.872 0.874 B3 0.868 0.870 0.872 B4 0.867 0.868 0.870 0.870 A1 0.874 0.874 0.870 0.868 0.994 A2 0.848 0.846 0.842 0.840 0.945 A3 0.868 0.868 0.867 0.865 0.953

PAGE 205

193 A4 0.870 0.868 0.865 0.863 0.985 P1 0.808 0.806 0.805 0.803 0.902 P2 0.842 0.842 0.840 0.838 0.900 P3 0.846 0.846 0.842 0.840 0.959 K1 0.868 0.868 0.867 0.865 0.983 K2 0.867 0.867 0.865 0.863 0.979 MALE 0.874 0.874 0.872 0.870 0.994 DRVY 0.872 0.872 0.870 0.868 0.992 CRSH 0.861 0.861 0.861 0.859 0.977 AGE 0.874 0.874 0.872 0.870 0.994 Covariance Coverage A2 A3 A4 P1 P2 ________ ________ ________ ________ ________ A2 0.945 A3 0.919 0.953 A4 0.942 0.947 0.985 P1 0.867 0.882 0.898 0.904 P2 0.870 0.898 0.897 0.855 0.900 P3 0.914 0.923 0.951 0.898 0.887 K1 0.936 0.944 0.976 0.897 0.895 K2 0.932 0.942 0.972 0.898 0.897 MALE 0.945 0.953 0.985 0.904 0.900 DRVY 0.944 0.951 0.983 0.904 0.900 CRSH 0.929 0.938 0.968 0.893 0.889 AGE 0.945 0.953 0.985 0.904 0.900 Covariance Coverage P3 K1 K2 MALE DRVY ________ ________ ________ ________ ________ P3 0.961 K1 0.951 0.989 K2 0.953 0.983 0.985 MALE 0.961 0.989 0.985 1.000 DRVY 0.961 0.987 0.983 0.998 0.998 CRSH 0.947 0.974 0.972 0.983 0.983 AGE 0.961 0.989 0.985 1.000 0.998 Covariance Coverage CRSH AGE ________ ________ CRSH 0.983 AGE 0.983 1.000 SAMPLE STATISTICS ESTIMATED SAMPLE STATISTICS Means

PAGE 206

194 B1 B2 B3 B4 A1 ________ ________ ________ ________ ________ 1 2.153 2.314 1.682 2.509 1.763 Means A2 A3 A4 P1 P2 ________ ________ ________ ________ ________ 1 1.996 1.909 1.882 2.435 2.101 Means P3 K1 K2 MALE DRVY ________ ________ ________ ________ ________ 1 2.061 42.450 16.143 0.419 0.716 Means CRSH AGE ________ ________ 1 1.099 20.105 Covariances B1 B2 B3 B4 A1 ________ ________ ________ ________ ________ B1 0.655 B2 0.312 0.609 B3 0.245 0.234 0.300 B4 0.249 0.253 0.187 0.355 A1 0.065 0.047 0.038 0.033 0.282 A2 0.205 0.127 0.098 0.085 0.091 A3 0.016 -0.002 0.025 -0.001 0.042 A4 0.085 0.088 0.048 0.067 0.122 P1 -0.079 -0.053 -0.041 -0.041 -0.015 P2 0.024 0.000 0.029 0.020 0.064 P3 -0.035 -0.005 0.001 -0.019 0.033 K1 2.230 1.617 0.598 1.821 0.738 K2 1.144 1.086 0.219 0.694 0.821 MALE 0.125 0.014 0.048 0.046 0.069 DRVY 0.103 0.100 0.035 0.095 0.041 CRSH 0.028 0.044 0.030 0.022 -0.001 AGE -0.009 0.013 0.011 0.005 0.028 Covariances A2 A3 A4 P1 P2 ________ ________ ________ ________ ________ A2 0.348 A3 0.037 0.164 A4 0.112 0.043 0.288 P1 -0.035 0.018 -0.014 0.325 P2 0.060 0.052 0.069 0.055 0.283 P3 0.021 0.037 0.014 0.089 0.089 K1 1.229 -0.582 1.052 -0.621 0.678

PAGE 207

195 K2 0.723 -0.237 0.417 -0.049 0.356 MALE 0.072 0.027 0.065 0.002 0.042 DRVY 0.048 -0.021 0.040 -0.025 0.017 CRSH 0.009 -0.004 -0.005 -0.018 -0.004 AGE -0.135 -0.055 -0.013 0.039 -0.063 Covariances P3 K1 K2 MALE DRVY ________ ________ ________ ________ ________ P3 0.321 K1 -0.168 187.456 K2 0.233 43.572 58.640 MALE 0.014 1.755 1.079 0.243 DRVY -0.002 2.851 1.234 0.059 0.203 CRSH -0.004 -0.013 -0.023 -0.002 0.010 AGE -0.023 2.193 1.791 -0.037 0.055 Covariances CRSH AGE ________ ________ CRSH 0.065 AGE 0.045 3.248 Correlations B1 B2 B3 B4 A1 ________ ________ ________ ________ ________ B1 1.000 B2 0.495 1.000 B3 0.553 0.547 1.000 B4 0.516 0.545 0.572 1.000 A1 0.151 0.114 0.131 0.103 1.000 A2 0.430 0.276 0.304 0.243 0.290 A3 0.048 -0.008 0.112 -0.005 0.195 A4 0.196 0.210 0.162 0.210 0.428 P1 -0.172 -0.120 -0.131 -0.120 -0.051 P2 0.057 0.000 0.098 0.063 0.225 P3 -0.076 -0.011 0.003 -0.057 0.109 K1 0.201 0.151 0.080 0.223 0.101 K2 0.185 0.182 0.052 0.152 0.202 MALE 0.314 0.037 0.178 0.157 0.264 DRVY 0.282 0.285 0.143 0.355 0.171 CRSH 0.136 0.221 0.216 0.143 -0.007 AGE -0.006 0.009 0.011 0.004 0.030 Correlations A2 A3 A4 P1 P2 ________ ________ ________ ________ ________ A2 1.000 A3 0.155 1.000 A4 0.355 0.199 1.000 P1 -0.104 0.078 -0.045 1.000

PAGE 208

196 P2 0.192 0.240 0.242 0.183 1.000 P3 0.062 0.163 0.044 0.277 0.295 K1 0.152 -0.105 0.143 -0.080 0.093 K2 0.160 -0.077 0.101 -0.011 0.087 MALE 0.247 0.135 0.245 0.007 0.162 DRVY 0.180 -0.115 0.164 -0.099 0.070 CRSH 0.060 -0.042 -0.038 -0.121 -0.027 AGE -0.127 -0.076 -0.013 0.038 -0.066 Correlations P3 K1 K2 MALE DRVY ________ ________ ________ ________ ________ P3 1.000 K1 -0.022 1.000 K2 0.054 0.416 1.000 MALE 0.049 0.260 0.285 1.000 DRVY -0.007 0.462 0.357 0.265 1.000 CRSH -0.026 -0.004 -0.012 -0.012 0.085 AGE -0.022 0.089 0.130 -0.041 0.068 Correlations CRSH AGE ________ ________ CRSH 1.000 AGE 0.098 1.000 MAXIMUM LOG-LIKELIHOOD VALUE FOR THE UNRESTRIC TED (H1) MODEL IS 9505.737 THE MODEL ESTIMATION TERMINATED NORMALLY TESTS OF MODEL FIT Chi-Square Test of Model Fit Value 309.869 Degrees of Freedom 95 P-Value 0.0000 Chi-Square Test of Model Fit for the Baseline Model Value 1724.325 Degrees of Freedom 130 P-Value 0.0000 CFI/TLI CFI 0.865 TLI 0.816

PAGE 209

197 Loglikelihood H0 Value -9660.671 H1 Value -9505.737 Information Criteria Number of Free Parameters 61 Akaike (AIC) 19443.343 Bayesian (BIC) 19704.218 Sample-Size Adjusted BIC 19510.586 (n* = (n + 2) / 24) RMSEA (Root Mean Square Error Of Approximation) Estimate 0.065 90 Percent C.I. 0.057 0.073 Probability RMSEA <= .05 0.001 SRMR (Standardized Root Mean Square Residual) Value 0.062 MODEL RESULTS Estimates S.E. Est./S.E. Std StdYX RTA BY A1 1.000 0.000 0.000 0.309 0.581 A2 1.111 0.138 8.048 0.343 0.580 A3 0.424 0.076 5.565 0.131 0.323 A4 1.128 0.117 9.630 0.348 0.648 RP BY P1 1.000 0.000 0.000 0.147 0.259 P2 2.740 1.080 2.537 0.403 0.759 P3 1.575 0.386 4.077 0.232 0.410 KLS BY K1 1.000 0.000 0.000 9.459 0.691 K2 0.488 0.052 9.446 4.618 0.603 RDB BY B1 1.000 0.000 0.000 0.582 0.716 B2 0.964 0.074 13.091 0.561 0.719 B3 0.729 0.052 14.054 0.424 0.767 B4 0.753 0.056 13.377 0.438 0.741 RDB ON RTA 0.873 0.195 4.485 0.463 0.463 RP -0.723 0.350 -2.067 -0.183 -0.183 KLS 0.006 0.006 0.939 0.098 0.098

PAGE 210

198 RP ON MALE 0.054 0.024 2.235 0.368 0.182 DRVY 0.004 0.021 0.207 0.029 0.013 CRSH -0.030 0.037 -0.813 -0.204 -0.052 AGE -0.005 0.005 -1.050 -0.035 -0.062 RTA ON MALE 0.232 0.037 6.266 0.753 0.372 DRVY 0.103 0.038 2.745 0.334 0.150 CRSH -0.013 0.062 -0.201 -0.041 -0.010 AGE -0.011 0.009 -1.193 -0.034 -0.062 KLS ON MALE 5.159 0.952 5.419 0.545 0.269 DRVY 11.852 1.198 9.896 1.253 0.565 CRSH -2.631 1.772 -1.485 -0.278 -0.071 AGE 0.742 0.252 2.939 0.078 0.141 RDB ON MALE -0.006 0.071 -0.085 -0.010 -0.005 DRVY 0.229 0.104 2.194 0.393 0.177 CRSH 0.519 0.109 4.753 0.893 0.228 AGE -0.007 0.016 -0.396 -0.011 -0.020 RTA WITH RP 0.017 0.006 2.945 0.368 0.368 KLS WITH RTA 0.174 0.163 1.070 0.060 0.060 RP 0.097 0.088 1.099 0.070 0.070 Intercepts B1 1.389 0.328 4.241 1.389 1.710 B2 1.574 0.316 4.983 1.574 2.016 B3 1.105 0.238 4.640 1.105 1.998 B4 1.941 0.246 7.892 1.941 3.285 A1 1.819 0.186 9.754 1.819 3.422 A2 2.057 0.208 9.885 2.057 3.482 A3 1.927 0.081 23.838 1.927 4.758 A4 1.944 0.210 9.251 1.944 3.623 P1 2.534 0.108 23.476 2.534 4.454 P2 2.404 0.273 8.819 2.404 4.527 P3 2.233 0.163 13.699 2.233 3.947 K1 19.782 5.314 3.723 19.782 1.444 K2 5.078 2.749 1.847 5.078 0.663 Residual Variances B1 0.321 0.027 11.828 0.321 0.487 B2 0.295 0.025 11.835 0.295 0.484 B3 0.126 0.012 10.703 0.126 0.411 B4 0.157 0.014 11.379 0.157 0.451 A1 0.187 0.015 12.316 0.187 0.663 A2 0.232 0.019 11.915 0.232 0.664 A3 0.147 0.010 15.053 0.147 0.896

PAGE 211

199 A4 0.167 0.016 10.719 0.167 0.580 P1 0.302 0.022 13.821 0.302 0.933 P2 0.119 0.048 2.498 0.119 0.423 P3 0.266 0.025 10.634 0.266 0.832 K1 98.097 10.582 9.271 98.097 0.523 K2 37.355 3.097 12.061 37.355 0.637 RTA 0.077 0.014 5.632 0.805 0.805 RP 0.021 0.012 1.747 0.957 0.957 KLS 45.023 9.040 4.980 0.503 0.503 RDB 0.219 0.031 7.190 0.648 0.648 R-SQUARE Observed Variable R-Square B1 0.513 B2 0.516 B3 0.589 B4 0.549 A1 0.337 A2 0.336 A3 0.104 A4 0.420 P1 0.067 P2 0.577 P3 0.168 K1 0.477 K2 0.363 Latent Variable R-Square RTA 0.195 RP 0.043 KLS 0.497 RDB 0.352 MODEL MODIFICATION INDICES Minimum M.I. value for printing the modification in dex 0.000 M.I. E.P.C. Std E. P.C. StdYX E.P.C. BY Statements RTA BY B1 10.249 0.443 0. 137 0.168 RTA BY B2 2.166 -0.196 -0. 060 -0.077 RTA BY B3 0.083 -0.026 -0. 008 -0.015 RTA BY B4 1.804 -0.134 -0. 041 -0.070 RTA BY P1 22.406 -0.605 -0. 187 -0.328 RTA BY P2 29.742 1.438 0. 444 0.835 RTA BY P3 3.547 -0.284 -0. 088 -0.155

PAGE 212

200 RTA BY K1 1.445 -3.706 -1. 143 -0.083 RTA BY K2 1.445 1.809 0. 558 0.073 RP BY B1 0.244 0.127 0. 019 0.023 RP BY B2 2.058 -0.353 -0. 052 -0.066 RP BY B3 1.390 0.200 0. 029 0.053 RP BY B4 0.113 -0.062 -0. 009 -0.015 RP BY A1 0.213 0.111 0. 016 0.031 RP BY A2 2.990 -0.469 -0. 069 -0.117 RP BY A3 14.557 0.708 0. 104 0.258 RP BY A4 0.621 -0.197 -0. 029 -0.054 RP BY K1 0.904 -5.912 -0. 870 -0.064 RP BY K2 0.904 2.886 0. 425 0.055 KLS BY B1 7.522 0.011 0. 107 0.131 KLS BY B2 0.043 -0.001 -0. 008 -0.010 KLS BY B3 16.697 -0.011 -0. 105 -0.191 KLS BY B4 3.489 0.006 0. 052 0.088 KLS BY A1 2.441 0.005 0. 046 0.087 KLS BY A2 1.606 0.004 0. 042 0.072 KLS BY A3 25.159 -0.012 -0. 115 -0.285 KLS BY A4 0.005 0.000 -0. 002 -0.004 KLS BY P1 5.337 -0.008 -0. 072 -0.127 KLS BY P2 4.992 0.011 0. 106 0.199 KLS BY P3 1.132 -0.003 -0. 032 -0.057 RDB BY A1 7.253 -0.145 -0. 084 -0.159 RDB BY A2 29.192 0.327 0. 190 0.322 RDB BY A3 6.773 -0.107 -0. 062 -0.154 RDB BY A4 1.233 -0.063 -0. 036 -0.068 RDB BY P1 18.201 -0.214 -0. 125 -0.219 RDB BY P2 12.977 0.270 0. 157 0.296 RDB BY P3 2.067 -0.070 -0. 041 -0.072 RDB BY K1 0.020 0.219 0. 127 0.009 RDB BY K2 0.020 -0.107 -0. 062 -0.008 WITH Statements B2 WITH B1 2.174 -0.031 -0. 031 -0.048 B3 WITH B1 0.974 0.015 0. 015 0.033 B3 WITH B2 0.028 0.002 0. 002 0.006 B4 WITH B1 1.841 -0.022 -0. 022 -0.045 B4 WITH B2 0.519 0.011 0. 011 0.024 B4 WITH B3 0.771 0.010 0. 010 0.030 A1 WITH B1 1.139 -0.015 -0. 015 -0.034 A1 WITH B2 0.061 -0.003 -0. 003 -0.008 A1 WITH B3 0.362 -0.005 -0. 005 -0.018 A1 WITH B4 1.718 -0.013 -0. 013 -0.041 A2 WITH B1 27.999 0.082 0. 082 0.170 A2 WITH B2 0.368 0.009 0. 009 0.019 A2 WITH B3 0.133 0.004 0. 004 0.011 A2 WITH B4 2.711 -0.018 -0. 018 -0.052 A2 WITH A1 4.167 -0.028 -0. 028 -0.089 A3 WITH B1 0.102 -0.004 -0. 004 -0.011 A3 WITH B2 2.192 -0.016 -0. 016 -0.051 A3 WITH B3 7.227 0.020 0. 020 0.089 A3 WITH B4 1.771 -0.011 -0. 011 -0.045 A3 WITH A1 0.153 0.003 0. 003 0.016

PAGE 213

201 A3 WITH A2 0.724 -0.008 -0. 008 -0.035 A4 WITH B1 3.823 -0.026 -0. 026 -0.060 A4 WITH B2 3.825 0.025 0. 025 0.060 A4 WITH B3 4.332 -0.018 -0. 018 -0.062 A4 WITH B4 1.779 0.013 0. 013 0.041 A4 WITH A1 10.616 0.044 0. 044 0.153 A4 WITH A2 0.865 -0.014 -0. 014 -0.044 A4 WITH A3 0.026 -0.001 -0. 001 -0.007 P1 WITH B1 3.380 -0.031 -0. 031 -0.067 P1 WITH B2 0.096 0.005 0. 005 0.011 P1 WITH B3 0.483 -0.008 -0. 008 -0.024 P1 WITH B4 0.130 0.004 0. 004 0.013 P1 WITH A1 1.999 -0.017 -0. 017 -0.056 P1 WITH A2 5.569 -0.032 -0. 032 -0.096 P1 WITH A3 1.597 0.013 0. 013 0.055 P1 WITH A4 0.953 -0.012 -0. 012 -0.038 P2 WITH B1 0.021 0.002 0. 002 0.005 P2 WITH B2 2.752 -0.023 -0. 023 -0.056 P2 WITH B3 2.756 0.016 0. 016 0.054 P2 WITH B4 0.583 0.008 0. 008 0.025 P2 WITH A1 0.000 0.000 0. 000 0.000 P2 WITH A2 0.072 0.003 0. 003 0.011 P2 WITH A3 4.062 0.018 0. 018 0.083 P2 WITH A4 0.240 0.006 0. 006 0.020 P2 WITH P1 2.296 -0.036 -0. 036 -0.119 P3 WITH B1 3.907 -0.031 -0. 031 -0.068 P3 WITH B2 1.577 0.019 0. 019 0.043 P3 WITH B3 0.416 0.007 0. 007 0.021 P3 WITH B4 0.868 -0.010 -0. 010 -0.031 P3 WITH A1 0.450 0.008 0. 008 0.026 P3 WITH A2 0.131 -0.005 -0. 005 -0.014 P3 WITH A3 3.988 0.019 0. 019 0.083 P3 WITH A4 3.776 -0.023 -0. 023 -0.075 P3 WITH P1 25.238 0.078 0. 078 0.243 P3 WITH P2 11.606 -0.150 -0. 150 -0.500 K1 WITH B1 0.048 0.074 0. 074 0.007 K1 WITH B2 0.056 -0.077 -0. 077 -0.007 K1 WITH B3 1.317 -0.255 -0. 255 -0.034 K1 WITH B4 1.915 0.335 0. 335 0.041 K1 WITH A1 2.813 -0.412 -0. 412 -0.057 K1 WITH A2 0.634 0.222 0. 222 0.027 K1 WITH A3 4.880 -0.446 -0. 446 -0.080 K1 WITH A4 1.227 0.274 0. 274 0.037 K1 WITH P1 2.230 -0.441 -0. 441 -0.057 K1 WITH P2 1.237 0.344 0. 344 0.047 K1 WITH P3 1.982 -0.395 -0. 395 -0.051 K2 WITH B1 0.409 0.125 0. 125 0.020 K2 WITH B2 4.862 0.413 0. 413 0.069 K2 WITH B3 3.268 -0.231 -0. 231 -0.054 K2 WITH B4 0.953 -0.136 -0. 136 -0.030 K2 WITH P2 0.176 -0.068 -0. 068 -0.017 K2 WITH P3 1.184 0.175 0. 175 0.040

PAGE 214

202 K2 WITH A1 6.961 0.370 0. 370 0.091 K2 WITH A2 0.624 0.126 0. 126 0.028 K2 WITH A3 2.369 -0.180 -0. 180 -0.058 K2 WITH A4 2.245 -0.210 -0. 210 -0.051 K2 WITH P1 0.226 0.081 0. 081 0.019 Beginning Time: 13:06:49 Ending Time: 13:06:52 Elapsed Time: 00:00:03

PAGE 215

About the Author Ismael Hoare received his Bachelor’s of Science in Medical Technology degree from Saint Louis University in 1991. He was awarded a Fulbright Scholarship in 1994. In 1995, Ismael received his Master of Public Health d egree from the School of Public Health and Hygiene at The Johns Hopkins University. In August 2000, Ismael was appointed as the inaugu ral Dean of the Faculty of Nursing Health Sciences and Social Work at the Univ ersity of Belize. He is currently a Senior Lecturer in the Faculty of Nursing, Allied H ealth and Social Work at the University of Belize. His interest is to continue r esearch activities on risk factors affecting young adults in Belize. Ismael is married to Olda and they have three children, Alyssa, Ismael II and Kieran.


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200409Ka 4500
controlfield tag 001 001988967
005 20091216162726.0
007 cr mnu|||uuuu
008 090216s2007 flu s 000 0 eng
datafield ind1 8 ind2 024
subfield code a E14-SFE0002190
035
(OCoLC)307589669
040
FHM
c FHM
049
FHMM
090
RA425 (Online)
1 100
Hoare, Ismael A.
0 245
Attitudinal factors related to driving behaviors of young adults in Belize :
b an application of the precaution adoption process model
h [electronic resource] /
by Ismael A. Hoare.
260
[Tampa, Fla] :
University of South Florida,
2007.
500
Title from PDF of title page.
Document formatted into pages; contains 202 pages.
Includes vita.
502
Dissertation (Ph.D.)--University of South Florida, 2007.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
3 520
ABSTRACT: Young adults' risk-taking attitudes, risk perception, and knowledge of road laws and signs influence their driving behaviors. The adoption of risky driving behaviors increases young adults' risk of motor vehicle crashes. The purpose of this study was to increase the understanding of the factors that lead to increased risks of MVC-related mortality and morbidity for young adults in Belize, to provide support for the development of evidence-based programs, and, more importantly, to investigate the relationships involving young adults' risk-taking attitudes, risk perception, and knowledge of road laws and signs and their relation to driving behaviors. The Precaution Adoption Process Model provided the theoretical foundation for this study and was used as the framework to investigate the variables of interest. This study used a nonexperimental, cross-sectional research design to examine the relationships between the latent variables. A convenience sample of 532 students enrolled at the University of Belize participated in this study. Data were collected through the completion of the Driving Behavior Survey. Structural equation modeling was used to examine the strength and direction of relationships among these latent variables and provide a better understanding of the relationships among these latent variables. The study found that the majority of students were in the final stages of the Precaution Adoption Process Model and were exhibiting the safest behaviors. However, the risk-taking attitudes significantly contributed to the manifestation of risky driving behavior and to a lesser extent so did risk perception. The study's findings suggest that interventions should focus on lowering young adults' risk-taking attitudes and raising risk perception to reduce risky driving behaviors.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
590
Co-advisor: Robert J. McDermott, Ph.D.
Co-advisor: Wayne W. Westhoff, Ph.D.
2 650
Accidents, Traffic.
Motor Vehicles.
Automobile Driving.
Attitude.
Risk-Taking.
Risk Factors.
653
Risk perception
Risk-taking attitudes
Risky driving behaviors
Knowledge of road laws and signs
Motor vehicle crashes
690
Dissertations, Academic
z USF
x Community & Family Health
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.2190