|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
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 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2010 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0004623
Predictors of academic achievement among students at hillsborough community college :
b can school engagement close the racial gap of achievement?
h [electronic resource] /
by Warren Smith.
[Tampa, Fla] :
University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains X pages.
Thesis (MA)--University of South Florida, 2010.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
ABSTRACT: ABSTRACT In the United States today, significant gaps exist among the races along a variety of measures of academic success, including standardized test scores, grade point averages, and drop-out and graduation rates. In recent decades, social scientists and educators alike have sought to uncover the reasons for these gaps, and many have focused on the role of cultural and institutional factors within the school setting. In recent years, researchers have examined such factors as a students' school identification (Osborne 1997; Voelkl 1997), students' opportunities to learn and the classroom climate (Oakes 1985), students' sense of school belonging (Goodenow 1993), and of particular interest to this researcher, sense of school engagement (Fredricks, Blumenfeld and Paris, 2004). Using data drawn from the Community College Survey on Student Engagement (CCSSE) administered by Hillsborough Community College (HCC) in the spring semester of 2007, I explore (1) whether students' levels of academic achievement, as measured by grade point average, vary across racial groups, as much of the literature has shown; and (2) whether any of the observed racial differences in academic achievement can be explained by differing levels of school engagement. Results show that black students at HCC do, in fact, report lower academic achievement compared to their white counterparts, but that these racial differences persist even after controlling for levels of school engagement. In other words, school engagement predicts academic achievement for all students, blacks as well as whites. The strongest predictors of academic achievement for students at HCC are class attendance, quality of student-faculty relations, and hours spent studying.
Advisor: James Cavendish, Ph.D.
Keywords: academic achievement
t USF Electronic Theses and Dissertations.
Predictors of Academic Achievement among Students at Hillsborough Community College : Can School Engagement Close the Rac ial Gap of Achievement ? by Warren T. Smith This thesis is submitted in partial fulfillment of the requirements for the degree of Masters of Arts Department of Sociology College of Arts and Sciences University of South Florida Major Professor: James C. Cavendish, Ph.D. Donileen Loseke, Ph.D. Maralee Mayberry, Ph.D. Date of Approval: June 2 2 0 10 Keywords: academic achievement, school e ngagement race Copyright 20 10 Warren T. Smith
DEDICATION I dedicate this thesis to my wife two daughters mother and brother without their patience, understanding, support and most of all love the completion of this project would not have been possible. I would also like to thank Dr. James Cavendish for his tireless support and confidence in me as a graduate student. He is truly a credit to his profession. Lastly, an un payable d ebt of gratitude goes to my mother, who taught me to be independent, suspicious of the status quo, and ambitious. She assumed the challenging responsibility of raising two boys, alone fighting the influences of the streets and she prevailed. And above al l I want to thank God who alone is responsible for all that is right and good!
ACKNOWLEDGMENTS I wish to express my sincere and profound gratitude to the following people who in one way or another have generously contributed to the successful compl etion of this study. I am heartily thankful to my graduate advisor, Dr. James Cavendish, whose encouragement, guidance and support from the initial to the final level enabled me to develop an understanding of the subject in the complet ion of th is project. I offer my regards and blessings to all of those who supported me in any respect durin g the completion of the project, particularly Dr. Isaac Williams who instilled in me eighteen years ago that I was worthy of academic success beyond a Bachelors Degree. Warren T. Smith
i TABLE OF CONTENTS LIST OF TABLES ii ABSTRACT iii CHAPTER 1: INTRODUCT ION 1 CHAPTER 2 : REVIEW OF THE LITERATURE ON SOCIAL CAPITAL AND SCHOOL ENGAGEMENT 6 Social Capita l 6 5 School Engagement 12 Behavioral Engagement 1 3 Emotional Engagement 1 4 C ognitive Engagement 1 8 Summary and Hypotheses 1 8 CHAPTER 3: RESEARCH METHOD OLOGY 21 The Sample 21 Survey Administration 21 Representativeness of Respondents 22 The Dependent Vari able 2 6 Independent Variables 2 7 Demographic Variables 2 9 Interaction Variables 30 Data Analysis 30 CHAPTER 4: RESULTS 3 2 CHAPTER 5: DISCUSSION AND CONCL USION 36 REFERENCES 39 APPENDICES 4 3 APPENDI X A : Question Items Wordings and Coded Values for V ariables Used in Analysis 4 4 APPENDIX B : Descriptive Statistics for Variables Used in Analysis 4 6 APPENDIX C : Z ero Order Pearson Correlations for Variables Used in Analysis 51 ABOUT THE AUTHOR END PAGE
ii LIST OF TABLE S Table 1: 2007 Demographic Profile of Students an d Survey Respondents at Hillsborough Community College and Community Colleges Nationally 25 Table 2: Mean Values on Self Reported Grade Point Averages for Students of Different Races/Ethnicities at Hillsborough Community College 32 Table 3: Net Relationships (Bs and Betas) Between Demographic Variables (Model I), Dimensions of School Engagement (Model II), and Interaction Variables (Model III and Academic Achi evement among Students at Hillsborough Community College (N=1,296) 34
iii PREDICTORS OF ACADEMIC ACHIEVEMENT AMONG STUDENTS AT HILLSBORO UGH COMMUNITY COLLEGE : CAN SCHOOL ENGAGEMENT CLOSE THE RAC IAL GAP OF ACHIEVEMENT ? Warren T. Smith ABSTRACT In the United States today, significant gaps exist among the races along a variety of measures of academic success, including standardized test sc ores, grade point average s and drop out and graduation rate s In recent decades, social scientists and educators alike have sought to uncover the reasons for these gaps, and many have focused on the role of cultural and institutional factors within the s chool setting In recent years, r esearchers have examined such factors as a students school identification (Osborne 1997; Voelkl 1997), students opportunit ies to learn and the classroom climate (Oakes 1985) students sense of school belonging (Goodenow 1993), and of particular interest to this researcher, sense of school engagement (Fredricks Blumenfeld and Paris, 2004). Using data drawn from the Community College Survey on Student Engagement (CCSSE) administered by Hillsborough Community College (HCC ) in the spring semester of 2007, I explore (1) whether students levels of academic achievement, as measured by grade point average, vary across racial groups, as much of the literature has shown; and (2) whether any of the observed racial differences in academic achievement can be explained by differing levels of school engagement. Results show that black students at HCC do, in fact, report lower academic achievement compared to their white
iv counterparts, but that these racial differences persist even aft er controlling for levels of school engagement. In other words, school engagement predicts academic achievement for all students, blacks as well as whites The strongest predictors of academic achievement for students at HCC are class attendance, quality of student faculty relations, and hours spent studying.
1 CHAPTER 1: INTRODUCTION America s educational system has purported to treat students the same by offering them the same opportunity for achievement, however education al outcomes are very different depending on their race, social class, and gender. Sociologist W.E.B. DuBois (19 98 ) observed this with respect to race when he predicted in 1903 that the color line would be the central problem of the twentieth century. Ove r a hundred years later, despite enormous change in the American educational system including the desegregation of schools significant gaps persist between the races along a variety of measures of academic success, including standardized test scores, gra de point averages, and drop out and graduation rates ( Hallinan 2001). These r acial disparities are of particular concern to leaders of colleges and universities because they have seen the enrollment of black and Hispanic/Latino students increase in recor d numbers over the past few decades. In 1976 15.4 percent of college students were ethnic minorities born in the United States; by 2000 that number had risen to 28.2 percent ( National Center for Education Statistics 2004). Despite the success of racial and ethnic minorities in gaining admission to colleges and universities, African American and Hispanic students continue to lack the proportional representation in higher education graduation rates (Johnston 2006) The challenge for post secondary ed ucational institutions today, is not primarily one of recruitment and admission of racial and ethnic minorities, but rather one of the retention and graduation of racial and ethnic minorities from their institutions. As lenge for students has gradually shifted from that of gaining access to higher education to that of persistence and achieving their academic
2 rom the minority students perspective, the challenge is less about gaining admission to college s and universities and more about performing well in their courses and ultimately graduati ng According to the National Center for Public Policy and Higher Education (2005) t hese challenges are seen most dramatically in the ever widening gap among racial groups in the level of education completed. Of the working age population, from 1980 to 2000 whites and Asian Americans made the most progress in attaining a bachelor s degree or higher, while African Americans, Native Americans, and Hispanics/Lat inos made the least progress. In fact, when comparing the percentage of the working age population that has attained a bachelor s degree or higher within each racial group, the gap between whites and Hispanics/Latinos has almost doubled over the last two decades growing from 12 percentage points in 1980 to 19 percentage points in 2000. The achievement gap between whites and African Americans has expanded from 11 percentage points in 1980 to 15 percentage points in 2000. The result of this widening gap is that in 2000, whites aged 25 to 64 were twice as likely as African Americans to have a bachelor s degree, and almost three times as likely as Hispanics/Latinos. The long term effects of this racial gap in academic achievement must not be under estimated. Future demographic shifts are expected to make the United States a non white majority nation around the year 2050, which is driving a consensus among policy makers, scholars and educators to shape school reform in ways to close the racial achievement gap Some observers even say the long term prospects for a healthy national economy and social stability depend on boosting the achievement levels and closing the gap for all students. Additionally, the National Center for Public Policy and Higher
3 Education (2005) states the greatest increase in population growth in the United States workforce is occurring among those racial/ethnic groups with the lowest level of education, while the group reaching retirement age is predominantly white with higher levels of e ducation. For these reasons, it has become vitally important to know why t hese racial disparities in academic achievement continue to persist e specially considering all of the manpower and resources targeted at closing the gap. Researchers have made a number of inroads in this area, but more work needs to be done, especially in the area of understanding what factors impact or influence students academic achievement. The purpose of this thesis is to investigate the factors that predict levels of acade mic achievement among students at Hillsborough Community College (HCC) in Tampa, Florida, using data drawn from the Community College Survey on Student Engagement (CCSSE) administered by HCC during the spring semester of 2007. Specifically, I explore (1) whether students levels of academic achievement, as measured by grade point average, vary across racial groups, as much of the literature has shown; and (2) whether any of the observed racial differences in academic achievement can be explained by differi ng levels of school engagement. I have chosen to focus on a community college for a variety of reasons. First, c ommunity colleges are a distinct institutional type in the United States post secondary education system. They maintain an integral part of t he educational and economic landscape in this country. Public two year colleges represent more than one fourth of all post secondary educational institutions in the United States and enroll more than one third of all college students, including a disprop ortionate number of nontraditional, part
4 time, and low socioeconomic status students ( Cohen and Brawer 2003). Secondly, d espite the voluminous empirical literature on the predictors of academic achievement among students attending four year colleges and u niversities, relatively few studies have achievement of students attending community colleges (Pascarella and Terenzini 2005). This dearth of research on community colle ges limits our understanding of the factors that may influence academic achievement among the populations that enroll in them. Finally, a focus on community colleges is important because of the large number of their students who are racial and ethnic mino rities or who come from lower socioeconomic backgrounds, a reality driven by the fact that the admissions standards and costs of attending a four year college or university ha ve risen considerably in the last decade. I have chosen to focus on school enga gement as a predictor of academic achievement because although school engagement is an important academic outcome in its own right, it has been identified by numerous researchers (e.g., Furrer and Skinner 2003) as a strong predictor of a variety of other o utcomes According to Finn and Voelkl (1993) and Goodenow and Gady (1993), the term school engagement is used broadly to academic s schooling. As such, school engagement is more than simply a measure of involvement in activities; it entails the extent to which emotional component, which relates Furrer and Skinner (2003) report that e ngagement in school improves performance and validates positive expectations about academic abilities. Moreover,
5 engagement seems to serve as an important social signal; when stud ents are engaged, they are provided with more motivational support by their teachers (Skinner and Belmont 1993). This motivational support improves the level of comfort students have in approaching faculty with questions about the course content. I t is no surprise therefore, that e ngagement is a strong predictor of students long term academic achievement and their eventual completion of school ( Furrer and Skinner 2003). More specific to my research interest in racial disparities in academic achieve ment, I have chosen to focus on school engagement because a variety of research has suggested that school engagement may be a n important mechanism for reducing the persistent racial gap in academic achievement. A ccording to Talbert Johnson (2004), many bl ack and Latino students come from social settings where their lives and experiences are vastly different from those of their middle class, monolingual, English speaking white classmates and teachers. Because they may experience a cultural disconnect and/o r language barrier, these students may have particular difficulty establishing the kind of trusting relationships with their peers and teachers which are necessary for developing a sense of belonging or feeling of community that has been identified in rese arch as an important prerequisite for academic achievement. When these kinds of trusting relationships are established, however, schools can become o ne of the few social settings where racial minorities by interact ing with people of different background s, can enjoy heightened levels of self esteem, positive academic outlooks, and higher levels of academic achievement.
6 CHAPTER 2: REVIEW OF THE LITERATURE ON SOCIAL CAPITAL AND SCHOOL ENGAGEMENT The literature relevant to this study is reviewed in th e following order: a) A conceptual model for understanding the role social capital plays in enhancing school engagement ; b) Th eories used by scholars to explain variations in academic achievement in general and the gap in academic achievement between raci al groups. These sections are followed by a summary of the reviewed literature. Social Capital Recent theoretical and empirical work on social capital offers a useful starting point for understanding how networks and relationships can foster the academic achievement of students, particularly African American and Latino students. James Coleman (1988) was among the first social scientists to theorize about the importance of social capital in educational settings. Although students have various sources of capital available to them that may influence their success in college -including financial capital, human capital and social capital -Coleman maintain ed that soci al capital lies at the intersection of networks, norms, and academic achievement. The c entral notion of social capital is that social networks have value. Social capi t al refers to the collective value of all social networks (i.e., who people know ) and the inclinations that arise from these networks to do things for each other ( norms of reci procity ) Accordingly, Bourdieu (1986) asserts that social capital consists of two key components: social relationships and the resources available because of those connections. While some scholars, such as Stanton Salazar (1997), focus on the relationsh ips among institutional agents, and the networks that weave these relationships into units, other scholars, such as Lin (2001) Burt (1997), and Fukuyama (1997), focus
7 on the informal values and norms of reciprocity and trustworthiness that arise from thos e social relationships which can be used to access resources. Lin (2001), for instance, asserts that social capital is the capacity of a person or group to utilize social relationships to mobilize resources embedded in a network structure. In other words capital can be described as investment in social relations with expected returns. Burt (1997) argue s that social capital is the opportunity (as opposed to the ability) to access various resources Fukuyama (1997) de fines social capital as the existence of a certain set of informal values or norms shared among members of a group that permit cooperation among them. Lin (2001) further contends that the quantity of social capital a person possesses depends on the size of the network connections they can mobilize and the amount of resources each person in that network possesses. Putnam (2000) identifies two types of social capital: bonding and bridging. Both might be useful in explaining the link between student engag ement and social capital. According to Putnam, bonding social capital constitutes close knit ties among similar individuals or groups. These ties provide support within the bonded group and within group solidarity, but they often present barriers to the formation of relationships outside of the close knit group. Putnam (2000) contends that bonding social capital is inward looking and tends to reinforce exclusive identities and homogeneous groups. Conversely, bridging social capital constitutes the relati ons between heterogeneous individuals or groups. Bridging social capital does not produce very strong ties, but they are more likely to be inclusive and allow for the transfer of resources across heterogeneous groups Putnam (2000) suggests that bridging social capital is the key to mobilizing community resources, acquiring wider variety of
8 resources and sharing diverse information. For the purpose of this research, both bonding bridging soc ial capital (e.g., is perhaps more relevant in generating the resources necessary to succeed in college. Although the possession of social capital is important for all stu dents, it may be particularly critical to the academic survival of those students who are more likely to feel alienated and unaccepted in an environment or by an institution whose values, beliefs, attitudes and culture seem incompatible with their own. Wi thout social capital, trust can erode, and at a certain point this erosion begins to manifest itself in perceived atypical behavior or behavior that the dominant culture perceives as deviant. This can prompt those individuals or groups who feel culturally isolated or disengaged from the campus community to potentially drop out of school. In theory social capital contends that building or rebuilding community and trust requires face to face encounters between students and teachers and students and their pe ers. A significant body of research has suggested that students from different social and cultural backgrounds look at schooling in vastly different ways (Eckert, 1989; Farrell, 1990; Weis, 1990). Students who identify with the more conventional expecta tions of working hard on abstract tasks (e.g., developing verbal and mathematical skills ) to obtain educational credentials are more likely to invest themselves in academic work than those students who have little hope in the future rewards promised by the educational system. Students who are more interested in developing their manual and physical competenc ies to enter a particular trade may be less inclined to invest themselves in academic work From this p er spective, the social and cultural orientations
9 that students bring to school which are largely shaped by their family backgrounds, are the most important factors affecting their engagement. Although the family bears the responsibility of transmitting the desirable motivational dispositions, educatio nal institutions and the ir agents can either enhance those dispositions or retard their development. The fact that educational institutions can enhance or retard certain dispositions in students by channeling them into different learning opportunities ca n be understood by habitus. Bourdieu (1971) defines habitus as a system of lasting dispositions, which integrate past experiences into a matrix of perceptions, appreciation and actions. This would help explain th e variability in how individual students and groups of students make sense of the world and the opportunities presented to them. Bourdieu (1971) further highlights the various ways in which students the educational system. Bourdieu and Passeron (1979) contend that some students try to bluff their way through, picking up bits and pieces of the valued capital along the way; other students simply give up when they realize that they lack the disposition s that schools require and reward. Children who grow up in broken homes, improvised communities or racially segregated neighborhoods often do not acquire the necessary social capital from membership in social networks that can translate into the desired values, norms and behavior which can be used as currency for academic achievement and upward mobility. Because children in these communities are often not expose d to children or adults of different races, ethnicities, or social classes, they may not deve lop the same types of dispositions norms, and aspirations and as a result, suffer in classroom settings. Cobb,
10 Wood and Yackel (1 996), for instance, state that classroom norms, which they define as pted ways of behaving, can have an effect on student learning opportunities and ultimately their academic achievement. According to Cobb et al. (1996), norms can influence the nature of discussions in the classroom, lassroom discussions ability to track student understanding. For students who enter educational institutions without the desired norms of classroom participation and performance, the school setting ities, behaviors and attitudes can change as a result of interactions with peers and teachers. According to Bourdieu, the transmission and effect of social capital in educational institutions may vary depending on the status of the teacher and his/her so cial network T eachers who possess a different value system or who value different styles of engagement and modes of communication may not be willing to solicit the membership or participation of those students who are different from the ones they are acc ustomed to interacting with in their social network. According to Bourdieu, w ithin the field of education, students who have certain dispositions that are shared by the teacher and the dominate culture and are thus seen as valued are more likely to be a ble to trade such dispositions for status or favor. Because social capital accrues from membership in social networks, black students may suffer from not being members of the same social networks ing different values on what is being taught, or even question what is learned and the reasons for learning it. It is also well documented that as students matriculate through the educational system their contact and interaction with teachers can signifi cantly decrease or change.
11 This decrease in interaction can profound ly limit t he transmission of specific social and cultural values skills codes and motivational dispositions valued by institutional agents. Furthermore, t he decrease in contact with in stitutional agents can negatively impact the performance of students, particularly African American and Latino students whose c ultural orientation promotes the establishment of relationships and networks as a support system for their retention and academic success. According to Nichols (1976) the axiology or the highest value for African Americans lies in interpersonal relationships. Hence, the maintenance and enhancement of the interpersonal relationship is considered the most preeminent value in Africa n American communities, a value that is highlighted in the human relations model of organizations, with its emphasis on interpersonal competence, warm personal ties, and collegial relationships (Kaplan and Tausky 1977; Litwak 1978). In summary, social int eraction in educational settings enables students to build social capital, those relationships of trust and support that can provide access to opportunities and resources necessary for academic achievement. Some students, however, are faced with challenge s in developing relationships of trust and support because of their varying norms, values, and aspirations. This can be especially true for students who are racial minorities or who are from economically disadvantaged families because schools tend to rewa rd white, middle class values ( Ogbu 1978 2003 ; Lareau 1989) The challenge for many black students who attend Hillsborough Community College (and other community colleges across the nation) is that their social networks are often localized and insular, thereby depriving them of exposure to the different values, norms, and expectations they may experience in college.
12 School Engagement Related to the theoretical concept of social capital is the concept of school engagement. The following review presents a definition of the concept of school engagement and describes three dimensions or types of engagement identified in the existing literature. This is followed by a review of social science literature that investigates empirically whether and how school e ngagement is related to academic achievement. Baumeister and Leary (1995) argue that social engagement is a basic human need, which individuals seek to satisfy through their social relationships. Because of its essential quality, human beings who do n ot experience a sense of engagement or who are not engaged socially might experience negative outcomes such as alienation or withdrawal. Willms (2000) defines school engagement as the extent to which students identify with and value schooling outcomes and participate in academic and non academic school activities. At its highest level, school engagement is seen in the form of membership conspicuous part of the school envi ronment, and the school is an important aspect of their own experience. Hallinan (2001) states that schools are second only to the family as a socializing agent of children. In synthesizing some of the recent research on school engagement, Fredricks, Blum enfeld, and Paris (2004) and Finlay and NCSE (2006) have identified what they consider to be three types of engagement:
13 1) academic, social, and extra curricular activities. 2) Emotio positive and negative reactions to teachers, classmates, academics, and school environment. This type of engagement is believed to create feelings of belonging to the institution and to ingness to do their work (Fredericks, Blumenfeld, and Paris 2004:60) 3) study, to comprehend complex ideas, and to master difficult concepts or skills (Fredericks, Blumenfeld and Paris 2004:60) According to Fredricks et al. (2004), s tudies on school engagement often include one or perhaps two of these types of engagement but rarely do they include all three components or deal with engagement as a multifaceted construct A brief review of this literature helps to illustrate this point. Behavioral Engagement In a report to the National Center for Education Statistics, Finn (1993) summarized studies that utilized measures of behavioral engagement. These studies, u sing a na tional sample of eight grade students from a survey sponsored by the U.S. Department of Education, found that s tudents who participated in school and classroom activities scored significantly higher on achievement tests than students who did not participat e. In addition, the results of the National Survey of Student Engagement (2009) show that students who indicated that they participated in college sponsored activities
14 that involved interacting with other students and members of the campus community demons trated a significantly higher grade point average than those who indicated they did not participate in college sponsored activities. In general, the more interaction students have with their peers and teachers, the more likely they are to learn effectivel y and persist toward the achievement of their educational goals. Personal interaction with faculty academic progress. Working with an instructor on a project or serving with faculty members on a college committee lets students see first hand how experts identify and solve practical problems. Through such interactions, faculty members become role models, mentors, and guides for continuous, lifelong learning. Emotional Eng agement The quantity of social interactions would be meaningless, however, if these interactions were not of sufficient quality to foster feelings of belonging or a sense of community Focusing on what might be described as emotional engagement, several s tudies have examined the influence of sense of belonging or sense of community on academic achievement. Although it is met outside of the classroom and school, these studies show that when these basic em otional needs are fulfilled in the school, students develop a sense of belonging that reduces the likelihood of dropping out (Tinto 1987; Bryk and Driscoll 1988) and enhances academic achievement (Finn and Voelkl 1993; Bryk and Driscoll 1988) According t belonging to the school community is derived largely from having positive relationships with peers and faculty at the school.
15 What are some of the mechanisms by which sense of belonging, or sense of community, might sign ificantly enhance academic achievement? Finn and Voelkl (1993) identification with the school as an important mechanism by which sense of community among peers maximiz es student learning outcomes. Bryk and Driscoll (1988) found tha t communally organized schools have fewer problems with student misbehavior (e.g., class cutting, absenteeism, etc.) and drop out, and greater levels of student interest in academics, than do other schools. These factors, according to Bryk and Driscoll (1 988), get translated into high levels of academic achievement Fine (1991) was significant ly and positively related to a cademic achievement and significantly negatively relate d to their likelihood of drop ping out of school. In a review of studies that examined the link between supportive schools and academic success, Schaps (200 5 ) reports that a sense of community at school positively a ffect s ademic motivation educational aspirations, and tendency to stay in school. According to Schaps (200 5 ) these positive outcomes can, in turn, lead to higher levels of academic achievement, as measured by grades and standardized test scores. Schaps (200 5 ) conclude s, however, that building a sense of community may not be sufficient for some students particularly low income students and s of strong norms and expectations at the school that encourage academic effort and achievement. The challenge with presenting these norms, however, is that schools are not necessarily neutral institutions, and may favor the norms, attitudes, and behaviors commonly associated with the dominant class. John Ogbu (1978) made this point when he
16 contended that blacks are not socialized to succeed in an educational system dominated by whites; rather they are trained to cope with their lower status in society that limits their opportunities. Having a sense of bel onging therefore, may be particularly important for racial minorities because, according to Talbert Johnson (2004), many black and Latino students come from social settings where their lives and experiences are vastly different from those of their middle class, monolingual, English speaking white teachers. In fact, the National Center for Educational Statistics (2004) reports that there is a high probability that black students in grades K 14 will have mostly white teachers throughout their educational ca reer. The under representation of black teachers can have a profound impact on student teacher relations further widening the achievement gap due in part to the cultural disconnect and potential language barrier. Johnson et al. (2001) state that beyond t he composition of the student body, the composition of the teaching staff may level of school engagement. For many students, race is perhaps the initial dimension along which students can identify with teachers and thus fe e l a sense of be l onging At the same time, students who come from low income households or single parent households are at greater risk of having their emotional and social needs neglected in their families. As a consequence, they may view school as an alternative place where they can get their social needs addressed. Booker (2004) and Sanchez Colon and Esparza (2005) argue that affirmative, meaningful interactions with teachers and other students in and out of the classroom are especially critical to the academic succe ss of black students because black students may seek relationships with teachers and students to substitute for
17 what they are missing at home. In the absence of the traditional two parent households, many black students may envision their school as an ext ension of their community and their teachers as surrogate parents. Although this can be a challenge for teachers who are not familiar with the nuances of the black family or black culture, the more interaction African American students have with their tea chers, the more likely they are to learn effectively and persist toward achievement of their educational goals. One of the challenges faced by those seeking to educate black students is the perception among some black students that the school culture is simply a reflection of the domina n t culture. Ogbu (1978) was one of the first to observe that students from historically oppressed minority groups may resist school goals as a way of opposing the culture and values of the dominant society. Describing thi s cultural disconnect as it is manifested today Witherspoon, Speight and Thomas (1997) and Steele (1992) observe that African American students particularly males who get good grades are often accused of trying to act white, because performing well in school has been deemed as a white behavior. Faced with the prospect of losing the ir connection with their friends, many black students invariably choose peer relations or group identity over academic achievement. DeRosier, Kupersmidt, and Patterson (1994 ) observations confirmed this by finding t hat students who are accepted by in group and out group peers experience more positive academic outlooks and higher levels of academic achievement than students who are accepted by in group peers but rejected or n ot highly accepted, by out group peers This reinforces the notion that because schools are one of the few social settings where racial minorities can interact with people of different backgrounds, school based relationships play a critical role in enhanc
18 Black students may also perceive differential expectations from their instructors based on their appearance, their perceived intellectual capabilities, or their perceived possession of different values, interests, a nd modes of communication. These perceived differential expectations can impact the ways in which students perceive their own academic abilities, which in turn influences their academic achievement. Many researchers have noted that black students need to have strong, positive relationships with faculty to promote their self esteem and to overcome the perception, held by the students or the faculty, that college is not a domain in which they can ex cel ( Wentzel and Wigfield 2009 ). Cognitive Engagement Fin ally, some studies (Newmann et al. 1992; Marks 2000) focus on the cognitive dimension of school engagement study habits and the discipline and dedication it takes to comprehend complex ideas and master di fficult concepts or skills. Marks (2000) conceptualized cognitive engagement as the attention, interest, investment and effort students expend in the work of learning. investment in and effort directed toward learning, understanding, mastering the These studies find a positive relationship between cognitive engagement and academic achievement. S ummary and Hypotheses This r eview of the literature highlights the fact that school engagement is a multi dimension al concept involving behavioral, emotional, and cognitive dimensions. The behavioral dimension, as measured by tion in a variety of
19 school activities and social networks in the educational setting, provides the kinds of social experiences that tie students to the norms and values of the educational context. ns and feelings about their relationships with their peers and their teachers, enables students to develop feelings of warmth and a sense of belonging to the educational community. Finally, the cognitive d dedication to learning, fosters students comprehension and mastery of important concepts and skills. Together, these dimensions of school engagement can enhance motivation to learn, their incentives to prepare for class and impress their tea chers, and their overall enjoyment of school. They can also reduce their likelihood of skipping classes and dropping out of school. According to Royal and Rossi (1997) these positive benefits get translated into higher academic achievement for all stude nts who are highly engaged. In addition, these dimensions of school engagement, particularly emotional engagement, may be especially critical in predicting the academic achievement of racial and ethnic minorities and students from low income families G uthrie and Wigfield (2000) for instance, recently found that highly engaged students from low income and minority families scored higher on reading tests than less engaged students from high income or white families. This suggests that increasing a stude nt s may be one way to close the racial academic achievement gap. According to the 2007 National Survey of Student Engagement (NSSE) not only does engagement work for minority and under represented students, but such practices make a bigger difference for such students than for students in general. Steele (1992)
20 contends that dis identification with the school is the root of black students academic achievement problems. The literature reviewed in this chapter included the litera ture on social capital and the literature on school engagement. The se literature s punctuate the importance of examining social capital and sense of engagement as predictors of academic achievement for students in general, and for African American students in particular On the basis of this review, I seek to test the following hypotheses: Hypothesis 1: African American students at Hillsborough Community College will report lower levels of academic achievement than will white students at Hillsborough Comm unity College. Hypothesis 2: Variables measuring school engagement will emerge as significant predictors of academic achievement among all students at Hillsborough Community College. Hypothesis 3: Variables measuring school engagement particularly e motional engagement, will have a greater influence on academic achievement among African American students than among White students. achievement at Hillsborough Community C ollege, I hope to be able to offer recommendations for the steps that HCC and other community colleges can take to ensure that all students particularly at risk students, c an enjoy increased rates of completion and improved levels of academic achievement
21 CHAPTER 3: RESEARCH METHODOLOGY The Sample This study uses survey data gathered from students who attend Hillsborough Community College (HCC), a mid sized public community college established in 1968 and located in Hillsborough County (Tampa), Flori da. In the spring of 200 7 Hillsborough Community College (HCC) a greed to participate in a nationwide survey of community colleges designed to evaluate the performance of colleges in meeting the needs of their constituents (students). The survey, titled the Community College Survey of Student Engagement (CCSSE) was administered to 310,013 students at 525 two year colleges in 48 states, the British Columbia, Nova Scotia and the Marshall Islands using a uniform sampling and administrative procedure To a ssist in the administration of the survey at HCC, HCC administrators submitted a copy of its class schedule of credit generating courses for the spring semester to CCSSE staff and CCSSE randomly selected the courses that would receive the survey. Remedial college preparatory courses were excluded from the sample. The required number of course sections to be surveyed was determined by the total sample size needed to reduce sampling error and to ensure valid results. The sampling and survey strategy produc ed a usable N of 1 2 52 community college students from four campuses for a response rate of 83% Survey Administration Survey administration took place in classrooms during regularly scheduled class meeting times and was not announced to the stud ents in advance. The survey was administered by either the faculty member teaching the course or by a campus
22 representative. Survey administrators were given a script that they read to students in each classroom. The script instructed students to comple te all items on the survey and reminded them that the survey is about their experience at the college where the survey is being administered. Because sampling units were classrooms and not individuals, some students were in more than one sampled course. Students who were attending more than one course in which the surveys were administered were asked to complete the survey again even if they had completed one in another class. CCSSE administered a test retest reliability on respondents who took the surv ey more than once during the same administration year. While only the first completed surveys were included in the analyses reported here CCSSE has used the second completed survey to examine test retest reliability. Representativeness of Respondents I n order to gauge how representative the HCC survey respondents were of the larger Hillsborough Community College student population, d escriptive analyses of the demographic data from the CCSSE survey were performed and the results compared to the instituti onal data collected by HCC of its overall student population. A similar comparison was then made between the demographic profile of HCC respondents and the demographic profile of respondents to the nationwide CCSSE survey. Table 1 report s the results of these comparisons. As Table 1 shows survey participants at HCC are very similar to the overall HCC student population except in one major respect survey respondents are much more likely than non respondents to be full time students. While 66 % of HCC survey respondents reported being full time students and 34 % reported being part time students, institutional dat a o btained from HCC about
23 its student population show that of the overall student population at HCC only 32 % is enrolled full time and 68 % is enrolled part time. This disparity between the percent of full time students among survey respondents and the overall HCC student population can be explained by the fact that classes were sampled rather than students. Because full time students enroll i n more courses than do part time students, full time students had a greater chance of being selected to participate in the survey. This disparity is important to note because the results of the analyses reported herein will be less generalizable to the pa rt time student population at HCC than to the full time student population Despite this difficulty in being able to generalize my findings to the part time student population, I can be fairly confident that the results of my analyses are generalizable to HCC students of different races and genders because, as Table 1 also reports, there is little difference between HCC survey respondents and HCC students in terms of gender and race. While 39 % of HCC students are male and 61 % are female, 40 % of survey r espondents report being male and 60 % report being female. Furthermore, of those surveyed, 48 % (or 598 students) reported being White/Non Hispanic, 17 % (or 230 students) reported being Black or African American, 22 % (or 275 students) reported being Hispani c, Latino, Spanish, 6 % (or 65 students) reported being Asian, Asian American or Pacific Islander, 1.2 % (or 16 students) reported being American Indian or Native American and less than 1 % (or one student) reported being of another race. These figures are very similar to the institutional data provide d by the HCC administration, which show that 55 % of HCC students are White/Non Hispanic, 19 % are Black or African American, 20 % are Hispanic, Latino or Spanish and 4 % are Asian.
24 The demographic profile of t hose who responded to the survey at Hillsborough Community College is also very similar to the demographic profile of those who responded to the nationwide 2007 CCSSE survey except in respect to race and age As Table 1 illustrates, the two groups of resp ondents are similar with respect to full time or part time status and gender. Th e HCC distribution reveals that 34 % of the surveyed respondents identified as part time students while 66 % identified as full time students. The percent of full time students in the national survey is slightly higher than the HCC survey at 70 % and the number of part time students is slightly lower at 40% Of the respondents to the HCC survey, 40% identified themselves as male while 60% identified as female. Again, these fi gures are comparable to those reported in the national survey, where 41% reported being male and 5 9% reported being female.
25 Table 1 : 2007 Demographic Profile of Students and Survey Respondents at Hillsborough Community College an d Community Colleges Nationally HCC respondents are slightly different from CCSSE member survey respondents, however, in terms of race and age. Whereas only 48% of HCC respondents report being white, non Hispanic, 64% of CCSSE member college respondents report being white, non Hispanic. Furthermore, 70% of HCC respondents (or 911 students) are between the ages of 18 and 24, 25% (or 320 students) are between the ages of 25 and 49 and the 1% (or 18 students) are between the ages of 50 and 65. This distribution reveals that the HCC respondents are somewhat younger than those who responded to the nationwide 2007 CCSSE survey, where 55% reported being between the ages of 18 and 24, 39% between the ages of 25 and 49, and 5% between the ages of 50 and 65. Based on these data, we can say that HCC respondents are similar to CCSSE survey respondents except Student Characteristic s All HCC Students HCC Respondents All CCSSE Students CCSSE Respondents Enrollment Status Full time students Part time students 32% 68% 66% 34% 38% 62% 70% 30% Gender Male Female 3 9% 61% 40% 60% 41% 59% 41% 59% Race/Ethnicity White, Non Hispanic Black Non Hispanic Hispanic, Latino, Spanish Asian or Pacific Islander Native American Other 55% 19% 20% 4% 0% 1% 48% 17% 22% 6% 1% 6% 58% 13% 15% 6 % 1 % 64% 12% 12% 6 % 2 % Age 18 24 years old 25 49 years old 50 64 years old 65+ years old 70% 25% 1% 55% 39% 5%
26 that HCC respondents are slightly younger and more likely to be racial minorities compared to CCSS E respondents. The Dependent Variable : Academic Achievement (Grade Point Average) Ac ademic achievement can be measured in a variety of ways and can be based on several factors. Many colleges use grade point average s standardized test scores and grad uation rates as measures of academic achievement. In college academic achievement entails fulfilling academic requirements resulting in a grade point average and ultimately a degree or certificate. The primary dependent variable in my analyses is acade mic achievement which is operationalized by the use of grade point average Young and Fry (2008) contend that grade point average (or GPA) is a broad and progressive measure of academic achievement over the duration of enrollment i n a course or at a college. It provides an overall view of a student s performance and it is an internationally recognized. While the use of one variable to measure of academic achievement is not ideal, given the limitations of the CCSSE survey data it is necessary. Additionally, GPA is the only direct, traditional, universal and quantifiable measure available that provides a reasonable ordinal ranking of academic achievement. In the CCSSE survey, GPA was assessed by asking the student respondents to self report their grade s in response to the following question : I n what range is your overall college grade average ( GPA ) ? Respondents were asked to select from the responses: C or lower (coded 1), C (coded 2), C+ to B (coded 3), B (coded 4), B+ to A (coded 5) or A (coded 6) (See Appendix for variable names, question wordings, and actual coded values) Forty one students ( or 3 .2 % ) indicated they had a grade point
27 average of a C or lower, 103 students ( or 7 .9 % ) indicated they had a grade point average of a C, 2 51 students ( or 19.4 % ) indicated they had a grade point average of C+ to B 3 15 ( or 24% ) indicated they had a grade point average of B, 3 45 ( 26.6 % ) indicated they had a grade point average of B+ to A and 1 53 ( or 11 .8 % ) indicated they had a gra de point average of A. There were 44 ( or 3% ) missing cases Independent Variable s The primary aim of this study is to examine the relationship between sense of school engagement and academic achievement as measured by using the proxy GPA am ong African Americans at Hillsborough Community College. The CCSSE survey consisted of 38 questions designed to measure the students level of involvement in various aspects of Hillsborough Community College. Of the se 38 questions 6 independent variable s were identified as measures of the three dimensions of school engagement behavioral engagement, emotional engagement, and cognitive engagement -described in the existing literature. The variables identified as measuring were frequency of skipping class, hours spent in college activities, and use of academic advising. In your experiences at this college during the current school year, about how often h ave you done each of the following? 1=Never, 2=Sometimes, 3=Often, and 4=Very Often. To measure the number of hours About how many hours do you spe nd in a typical 7 day week doing each of the following? Participating in college sponsored activities (organizations, campus publications, student govt., intercollegiate or intramural
28 None, 1=1 5 hours, 2=6 10 hours, 3=11 20 hours, 4=21 30 hours, and 5=More than 30 hours. Because preliminary analyses revealed that this variable was highly skewed (i.e., with 144 respondents reporting that they spent 1 5 hours but only 30 respondents indicating th at they spent 6 10 hours, 13 respondents reporting that they spent 11 20 hours, 8 reporting 21 30 hours, and only 5 reporting more than 30 hours), this variable was recoded into a dichotomous variable in which 0= No hours spent on college sponsored activit ies, and 1= 1 or more hours spent on college sponsored activities. To measure the use of academic advising, advising/planning 2=Sometimes, and 3=Often. were quality of peer relationships and quality of faculty relationships. To measure the the box that best represents the quality of your relationships with people at this college. Your relationship with: The variable identified as measurin the simple meas About how many hours do you spend in a typical 7 day week doing each of the following? Preparing for class
29 (studying, reading, writing, rehearsing doing homework, or other activities related to 5 hours, 2=6 10 hours, 3=11 20 hours, 4=21 30 hours, and 5=more than 30 hours. Demographic Variables ace, gender, age, marital status, and having a child can influence their academic performance, all of these variables were included as control variables in my analyses. Race is included because previous studies have shown that significant gaps persist bet ween the races along a variety of measures of academic success, including grade point averages, with blacks having significantly lower levels of academic achievement than whites (Hallinan 2001; National Center for Public Policy and Higher Education 2005). To measure race, respondents to were used to create dummy variables for each racial group (i.e., Native American, Asian, Black, Hispanic, White), and in the regression analyses White is omitted to serve as a reference category. The literature on college academic achievement also reports the presence of gender differences n are le ss favored in college and university admissions today than they were in the past. While of the ratio of male to female college admissions favored men during the 1970s and achieved equilibrium during 1980s, the ratio favored women (54 to 46) during the 199 0s. The same type of shift i.e., from ratios favoring men to ratios favoring women can be seen in college graduation rates. In the 1970s, women were less likely than men to receive bachelors and masters degrees, whereas in
30 the 1990s, women made up th e majority of bachelors and masters degree recipients. In light of this literature which documents disparities between the genders, I include gender (coded 0=male, 1=female) as a control variable in my analyses. Age, marital status, and having a child we re also included in my analyses. The response categories for age presented to the respondents of the survey were: 1=under 20 years old; 2=20 to 21; 3 =22 to 24; 4 =25 to 29; 5 =30 to 39; 6 =40 to 49; 7 =50 to 64; and 8 =65+. Marital status was measured by ask Interaction Variables Because the literature on academic achievement sug gests that school engagement, particularly emotional engagement, may be especially critical in predicting the academic achievement of racial and ethnic minorities (Guthrie and Wigfield 2000; Steele 1992), leading some to argue that school engagement might actually be one of the best strategies for closing the racial academic achievement gap, I have included in my analyses two interaction terms Black X Quality of Peer Relationships and Black X Quality of Faculty Relationships to measure whether either or both of these two measures of emotional engagement is a better predictor of GPA for black students than for white students. Data Analysis First, to test whether black students at Hillsborough Community College report lower levels of academic achievement than white students I conducted an Independent Sample t test of mean scores on the dependent variable (i.e., Grade Point Average) for the black and white subsamples. This provided a direct test of the hypothesized black
31 white difference in academic achi evement based on the prior literature. This t test was followed by a Ordinary Least Squares (OLS) multiple regression to examine whether any differences in academic achievement based on race stand up while controlling for other demographic variables. To test my second hypothesis, in which I predicted that v ariables measuring school engagement will emerge as significant predictors of academic achievement among all students at Hillsborough Community College, I conducted a second OLS regression in which the variables measuring school engagement were added to the model in which only the demographic variables were included. By observing the change in values of the Adjusted R square statistic, I am able to determine the extent to which adding the variables mea suring school engagement increases the predictive capacity of the model. This is because the Adjusted R square statistic can be interpreted as the percent of variation in the dependent variable (i.e., GPA) that is attributable to the independent variables in the model. Finally, to test my third hypothesis, in which I predicted that v ariables measuring school engagement, particularly emotional engagement, will have a greater influence on academic achievement among black students than among w hite s tudents, I added two additional variables the regression model the interaction terms Black X Quality of Peer Relationships and Black X Quality of Faculty Relationships.
32 CHAPTER 4 : RESULTS Table 2 displays the mean values on self reported grade points averages for students of different races and ethnicities at Hillsborough Community College. As n other groups of students. Furthermore, independent sample t tests of difference between black and white students at H CC revealed that a statistically significant difference exists between black and white students (p<.05). Table 2: Mean Values on Self Reported Grade Point Average s for Students of Different Races/Ethnicities at Hillsborough Community College Mean GPA N Native American 4.13 16 Asian or Pacific Islander 4.20 51 Black, Non Hispanic 3.87 230 White, Non Hispanic 4.11 598 Hispanic, Latino 4.09 275 Other 4.06 65 Table 3 reports the results of three OLS regression models, the first of which examines whether the difference in academic achievement observed between black and white students holds up when controlling for demographic variables. Because black students at HCC, in comparison to other students in the HCC sample, report being older and more lik ely to have a child, it is conceivable that once age and the presence of a child are controlled, that the significant, zero order race difference in academic achievement will no longer be apparent. As Model I in Table 3 shows, black students (b= .250, p<. 05) report significantly marital status, and child care obligations. In addition to this significant relationship
33 between race and GPA, Model I also reveals that age ( b=.148) gender (b=.220) and marital status (b=.357) are significant net predictors of academic achievement. Those who are older (p<.001) female (p<.01) or married (p<.01) report significantly higher As the beta values (or standardized regression coefficients) in this model show, of the various demographic characteristics of respondents, age (beta=.179) and marital status (beta=.103) appear to be ment. Model II in Table 3 adds the variables measuring school engagement. As the model shows, variables measuring each of the dimensions of school engagement behavioral, emotional, and cognitive emerge as significant predictors of academic achievemen t. Those who report skipping classes (and the frequency of that skipping) b= .292, p<.001). Moreover, those who report spending some time in college sponsored activities the other measure of behavioral
34 Table 3 : Net Relationships (Bs and Betas) Between Demographic Variables (Model I), Dimensions of School Engagement (Model II), and Interaction Variables (Model III) and Academic Achievement among Students at Hillsborough Community College (N=1,296) Model I Model I I Model II I B Beta B Beta B Beta Demographic Variables Native American .086 .008 .030 .003 .026 .002 Asian .122 .018 .088 .013 .090 .013 Black .250* .073* .292** .085** .242 .071 Hispanic .060 .019 .007 .002 .007 .002 Age .148*** .179*** .096*** .115*** .095*** .114*** Female .220** .082** .148 .055 .150 .056 Married .357** .103** .237* .069* .234* .068* Has Child (1=yes; 0=no) .078 .025 .094 .031 .090 .029 Behavioral Engagement Frequency of skipping class .292*** .149*** .294*** .150*** Spent time in college activities .224* .062* .228* .063* Use of academic advising .061 .031 .060 .031 Emotional Engagement Quality of Peer Relationships .053 .054 .062 .063 Quality of Faculty Relationships .180*** .181*** .173*** .174*** Cognitive Engagement Hours Spent Studying .146*** .113*** .146*** .113*** Interaction Variables Black X Qual. of Peer Relations .059 .097 Black X Qual. of Fac. Relations .048 .081 Constant 3.385 2.680 2.670 Adjusted R .060 .151 .150 engagement t spent time in these types of activities (b=.224, p<.05). Model II also shows that emotional engagement is a significant predictor of academic achievement, but only in terms of the quality of the with the faculty, not his/her pee rs. HCC students who report that
35 than students who report that their instructors are u navailable, unhelpful, or unsympathetic ( b=.180, p<.001). Cognitive engageme nt also matters i n terms of academic achievement, with time spent studying emerging as a significant, net predictor of academic achievement. The longer students spend p reparing for class (studying, reading, writing, rehearsing, doing homework, or other a ctivities) the higher their reported grade point average s ( b=.146, p<.001). Examining the beta values (or standardized coefficients) in Model II further reveals that of the variables measuring school engagement, the quality of faculty relations (beta= .181) appears to be the most influential although attending class and studying hard are also very important Overall, students who feel engaged by their teachers attend class regularly, and spend time studying reported having higher grades th a n those st udents who report they were not engaged by instructors skipped classes, and did not study. Model III tests whether emotional engagement might be more influential and predicting academic achievement among black students than among white students. As the model shows, neither of the two interaction variables emerges as a significant predictor of GPA. In other words, although emotional engagement is a significant predictor of academic achievement among students in general, it is not more influential for black students than for white students, as I had hypothesized. Examination of the R square values of these three models reveals that Model II does the best job explaining variation in academic achievement. Slightly over fifteen percent of the va riation in the dependent variable (GPA) is explained by the variables in this model.
36 CHAPTER 5 : DISCUSSION AND CONCLUSION Th e central finding of this analysis is that school engagement matters for student success at HCC. The strongest predictors of ac ademic achievement for students at HCC are class attendance, quality of student faculty relations, and hours spent studying. These three factors can be conceived as measuring each of the three dimensions of school engagement behavioral engagement, emoti onal engagement, and cognitive engagement -identified by previous researchers ( Fredricks, Blumenfeld, and Paris 2004) faculty emerges as such a strong predictor of academ ic achievement in this research confirms the results of Talbert Johnson (2004) who observed that the most important variable in determining academic achievement in a school environment is the quality of teachers. Such relat ionships are seen as indicators of the extent to which students have integrated themselves into the academic and social aspects of a college community, which Tinto (1987) showed are year persistence decisions and long term a cade mic achievement. Although black students at HCC do, in fact, report lower academic achievement compared to their white counterparts, these racial differences persist even after controlling for ound characteristics and for levels of behavioral, emotional, and cognitive school engagement. In other words, whatever is causing the racial academic achievement gap at HCC (and I suspect other community colleges), it appears NOT to be linked to the back ground characteristics or levels of school engagement included in my analyses. S chool engagement predicts academic achievement equally well for all stude nts, blacks as well
37 as whites, and while all students, including black students, should avail themselv es of opportunities for engagement and supportive relationships with their teachers, researchers and college administrators must look beyond these factors if they want to close the academic achievement gap. One factor which our society must examine in gr eater detail is a factor which I was unable to incorporate into my analyses because of its absence in the CCSSE survey that is, readers may wonder if the academic achievement g ap between whites and blacks is better quality of the childhood schools, ancillary analyses of the CCSSE survey data for HCC is significantly related I believe that studying a community college like HCC is adds an important dimension to the existing literature on school engagement and academic achievement because, as men tioned earlier, community colleges are a distinct institutional type in the United States post secondary education system. Because community colleges in general enroll a disproportionate number of nontraditional, part time, and low socioeconomic status s t udents (Cohen and Brawer 2003), and because HCC in particular enrolls a disproportionate number of racial and ethnic minorities, these contexts are important sites for observing a high degree of variability in academic achievement. As admission standards and tuition costs continue to climb and four year colleges and universities, we
38 continue to attract and enroll highly diverse groups of students with highly diverse aca demic abilities.
39 REFERENCES Anderman E.M. 2002 Journal of Educational Psychology 94:795 809. Astin A. 1993. What Matters in College? Four Critical Years Revisited San Francisco: Joss ey Bass. Baumeister R. F., and Leary M. R. 1995 Psychological Bulletin 117:497 529. in African Curriculum and Teaching Dialogue 6:133 146. 258 in Handbook of Theory and Research for the Sociology of Education John G. Richardson, (ed.). New York: Greenwood Press. Administrative Science Quarterly 42: 339 365. Bryk, A.S., and M. E. Driscoll. 1988 The High School as Community: Contextual Madison Wisconsin: National Center on Effective Secondary Schools, University of Wisconsin. Cohen, A. M., and F. B. Brawer. 2003. The American Community College San Francisco: Jossey Bass. The American Journal of Sociology 94:S95 S120. Educational Psychology 18:88 108. DeRosier, M. E., J. B. Kupersmidt, and C. J. Patterson. 19 Child Development 65:1799 1813. Dubois, W.E.B. 1998. The Souls of Black Folk New York: Bantam (originally published in 1903). Fairc Journal of Youth and Adolescence 34(4):293 309. Fine, M. 1991. Framing Dropout State University of New York Press.
40 Finlay, K. A. and NCSE. 2 006. Quantifying School Engagement: Research Report Denver, CO: The National Center for School Engagement (NCSE). Retrieved from http://www.peecworks.org/PEEC/PEEC_Inst/S035FA6B1 035FA923 Review of Educati onal Research 59:117 142. _____. 1993. School Engagement and Students at Risk Washington, DC: National Center for Education Statistics. Journal of Negro Educatio n 62(3):249 268. New York, NY: Eric Clearinghouse on Urban Education 3, 1 5. Retrieved February 23, 2006, from Eric Educational Database. Fredericks, J.A., Blumenfeld, Review of Educational Research 74: 59 109. Academic Engagement and Perform Journal of Educational Psychology 95(1):148 162. Journal of Experimental Education 62:60 71. Gre Outcome Gap: Differences for African American & Hispanic Community College Students in Student Engagement The Journal of Higher Education 79: l Perspectives on Black White Inequalities in American Sociology of Education 74:50 70. Sociology of Education 7 4(4): 318 340. Johnston, Candice Globuschutz. 2006. Predictors of College Success Among African American, Caucasian, and Hispanic Students Ph.D. Dissertation, Texas Tech University. Public Administration Review 37(5):171 180. Lareau, A. 1989. Home Advantage: Social Class and Parental Intervention in Elementary Education New York: Falmer Press.
41 Lin, N. 2001. Social Capital. A Theory of Social Structure and Action Cambridge University Press. National Center for Education Statistics. 2004. Mini Digest of Education Statistics 2003 U.S. Department of Education, Institute of Education Sciences, Washington, D.C.: National Center for Education Statistics, October 2004. Groups that are the Least Educated are the Fastest Gr owing. Policy Alert. November. National Survey of Student Engagement. 2009. Assessment for I mprovement: Tracking Student E ngagement over T ime Annual R esults 2009 Bloomington, IN: Indiana University Center for Postsecondary Research. Newman n F. M., and 39 in Student Engagement and Achievement in American Secondary Schools Fred M. Newmann, (ed.). New York: Teachers College Press. manuscript presented at the meeting of the World Psychiatric Association, Ibadan, Nigeria, Nov. 1976. Sociology of Education 55:197 212. Ogbu, J. 1978. Minority Education and Caste: The American System in Cross cultural Perspective New York: Academic Press. _____ 2003. Black American Students in an Affluent Surburb: A Study of Academic Disengagemen t. Mahwan, N.J.: Lawrence Erlbaum Associates. The Civil Rights Project, Harvard University. Putnam, R. D. 2000. Bowling Alone: The Collapse and Revival of American Community New York: Simon Schuster. Fulfilling Harvard Educational Review 40(3):411 451. Leve l Correlates of Sense of Journal of Community Psychology 24(4):395 416.
42 Journal of Youth and Adolsecence 6:619 628. Getting Results: Developing Safe and Healthy Kids, Update 5: Student Health, Supportive Schools, and Aca demic Success Developed by the Safe and Healthy Kids Program Office, California Department of Education. Effects of Teacher Behavior and Student Engagement Across the Sc Journal of Educational Psychology 85(4):571 581. Stanton Harvard Educational Review 67(1):1 38. Steele, C. M. 1992. Atlantic 269(4):68 78. Talbert Education and Urban Society 37(1):22 36. Tinto, V. 1993. Leaving College: Rethinking the Causes and Cures of Student Attrition. 2 nd Ed. Chicago: University of Chicago Press. United States Department of Education, 2004. National Center for Education Statistics. Digest for Educational Statistics. Remedial Education at Higher Education Institutions W ashington DC: US Government Printing Office American Journal of Education 105(3):294 318. Wehlage, G. G. 1989. Reducing the Risk: Schools as Communities of Support New York: The Falmer Press. Wentz el, K. R. and A. Wigfield. 2009. Handbook of Motivation at School New York: Routledge. ISUMA: Canadian Journal of Policy Research 1(2):64 70. Witherspoon, K., Speight, S. L., and Thomas, A. J. 1997. School Achievement, and Academic Self efficacy among African American High School Journal of Black Psychology 23: 344 357.
43 A PPENDICES
44 APPENDIX A: QUESTION ITEM WORDINGS AND CODED VALUES FOR VA RIABLES USED IN ANALYSES Dependent Variable Wording of Question Item Actual Coded Values Used Academic Achievement (GPA) At this college, in what range is your overall college grade average? 1=C or lower; 2=C; 3=C+ to B ; 4 = B; 5= B+ to A 6 = A Demographic Variables Wording of Questions Items Actual Coded Values Used Native American What is your racial identification? 1=American Indian or Native American Asian What is your racial identification? 1=Asian, Asian American or Pacific Islander Black What is your racial identification? 1= Black or African American, Non Hispanic Hispanic What is your racial identification? 1= Hispanic, Latino, Spanish Age Mark your age group. 1= Less than 20; 2=20 to 21; 3=22 to 24; 4=25 to 29; 5=30 to 39; 6=40 to 49; 7=50 to 64; 8= 65+ Female Your sex. 0=Male; 1=Female Married Are you married? 0=No 1=Yes Has Child Do you have children who live with you? 0=No 1=Yes Time Spent Studying ( ACADPR01) About how many hours do you spend in a typical 7 day week? 0=None 1=1 5 hours 2=6 10 hours 3=11 20 hours 4=21 30 hours 5=More than 30 hours Independent Variables Wording of Questions Items Actual Coded Values Used Behavioral Engagement Frequ ency of skipping class In your experiences at this college during the current school year, about how often have you done each of 1=Never; 2=Sometimes; 3=Often; 4=Very often Hours spent in college activities About how man y hours do you spend in a typical 7 day week doing each of the following? college sponsored activities (organizations, campus publications, student govt., intercollegiate or intramural sports, etc.) Original: 0=None; 1=1 5 hours; 2=6 10 hours; 3=11 20 hours; 4=21 30 hours; 5=More than 30 hours Recoded as: 0=None; 1=1 or more hours Use of academic advising Indicate how often you use the following 1=Rarely/nev er; 2=Sometimes; 3=Often
45 APPENDIX A (CONTINUED) : QUESTION ITEM WORDINGS AND CODED VALUES FOR VARIABLES USE D IN ANALYSES Independent Variables Wording of Questions Items Actual Coded Values Used Emotional Engagement Quality of Peer Relationships Mark the box that best represents the quality of your relationships with people at this college. Scale from 1=Unfriendly, unsupportive, sense of alienation; to 7=Friendly, supportive, sense of belonging Quality of Fac ulty Relationships Scale from 1=Unavailable, unhelpful, unsympathetic; to 7 = Available, helpful, sympathetic Cognitive Engagement Hours spent studying About how many hours do you spend in a typical 7 day week doin g each of the following? Preparing for class (studying, reading, writing, rehearsing, doing homework, or other activities related to your program) 0=None; 1=1 5 hours; 2=6 10 hours; 3=11 20 hours; 4=21 30 hours; 5=More than 30 hours
46 APPENDIX B : DESCRIPTIVE STATISTICS FOR VARIABLES USED IN ANALYSES N Mean Median Mode Std. Deviation Skewness Kurtosis Valid Missing GPA 1208 88 4.0588 4.0000 5.00 1.28235 .379 .483 Age 1249 47 3.63 3.00 2 1.571 .809 .207 Female 1252 44 .6318 1.0000 1.00 .48251 .547 1.703 Married 1252 44 .1701 .0000 .00 .37590 1.758 1.092 Has Child 1256 40 .2309 .0000 .00 .42157 1.279 .365 Frequency of Skipping Class 1285 11 1.68 2.00 2 .667 .891 1.272 Hours Spent in College Activities 1276 20 .1 567 .0000 .00 36370 1 .89 1 1. 577 Use of Academic Advising 1177 119 1.67 2.00 2 .659 .484 .728 Quality of Peer Relationships 1275 21 5.31 5.00 6 1.358 .606 .066 Quality of Faculty Relationships 1275 21 5.48 6.00 6 1.294 .753 .272 Hours Spent Studying 1278 18 1.85 2.00 1 1.000 .907 .308 GPA Frequency Percent Valid Percent Cumulative Percent Valid 1 = C or lower 41 3.2 3.4 3.4 2 = C 103 7.9 8.5 11.9 3 = C+ to B 251 19.4 20.8 32.7 4 = B 31 5 24.3 26.1 58.8 5 = B+ to A 345 26.6 28.6 87.3 6 = A 153 11.8 12.7 100.0 Total 1208 93.2 100.0 Missing System 88 6.8 Total 1296 100.0
47 APPENDIX B (CONTINUED) : DESCRIPTIVE STATISTICS FOR VARIABLES USED IN ANALYSES RACE Frequency Percent Valid Percent Cumulative Percent Valid Native American 16 1.2 1.3 1.3 Asian or Pacific Islander 50 3.9 4.0 5.3 Black, Non Hispanic 230 17.7 18.6 24.0 White, Non Hispanic 598 46.1 48.4 72.5 Hispanic, Latino, Spanish 275 21.2 22.3 94.7 Ot her 65 5.0 5.3 100.0 Total 1235 95.3 100.0 Missing System 61 4.7 Total 1296 100.0 AGE Frequency Percent Valid Percent Cumulative Percent Valid 1= Less than 20 years old 379 29.2 30.3 30.3 2= 20 21 years 319 24.6 25.5 55.9 3= 22 24 years 213 16.4 17.1 72.9 4= 25 29 years 151 11.7 12.1 85.0 5= 30 39 years 111 8.6 8.9 93.9 6= 40 49 years 58 4.5 4.6 98.6 7= 50 64 years 17 1.3 1.4 99.9 8= 65+ years 1 .1 .1 100.0 Total 1249 96.4 100.0 Missing System 47 3.6 Total 1296 100.0
48 APPENDIX B (CONTINUED) : DESCRIPTIVE STATISTICS FOR VARIABLES USED IN ANALYSES F EMALE Frequency Percent Valid Percent Cumulative Percent Valid 0= Male 461 35.6 36.8 36.8 1= Female 791 61.0 63.2 100.0 Total 1252 96.6 100.0 M issing System 44 3.4 Total 1296 100.0 M ARRIED Frequency Percent Valid Percent Cumulative Percent Valid 0= No 1039 80.2 83.0 83.0 1= Yes 213 16.4 17.0 100.0 Total 1252 96.6 100.0 Missing System 44 3.4 Total 1296 100.0 H AS CHILD F requency Percent Valid Percent Cumulative Percent Valid 0= No 966 74.5 76.9 76.9 1= Yes 290 22.4 23.1 100.0 Total 1256 96.9 100.0 Missing System 40 3.1 Total 1296 100.0
49 APPENDIX B (CONTINUED) : DESCRIPTIVE STATISTICS FOR VARIABLES USED IN ANALYSES SKIPS CLASS Frequency Percent Valid Percent Cumulative Percent Valid 1= Never 536 41.4 41.7 41.7 2= Sometimes 656 50.6 51.1 92.8 3= Often 67 5.2 5.2 98.0 4= Very Often 26 2.0 2.0 100.0 Total 1285 99.2 100.0 Missing System 11 .8 Total 1296 100.0 HOURS IN COLLEGE ACTIVITIES Frequency Percent Valid Percent Cumulative Percent Valid 0= None 1076 83.0 84.3 84.3 1= 1 or more hours 200 15.4 15.7 100.0 Total 1276 98.5 100.0 Missing System 20 1.5 Total 1296 100. 0 USE OF ACADEMIC ADVISING Frequency Percent Valid Percent Cumulative Percent Valid Rarely/never 518 40.0 44.0 44.0 Sometimes 535 41.3 45.5 89.5 Often 124 9.6 10.5 100.0 Total 1177 90.8 100.0 Missing System 119 9.2 Total 1296 100.0
50 APPENDIX B (CONTINUED) : DESCRIPTIVE STATISTICS FOR VARIABLES USED IN ANALYSES QUALITY OF PEER RELATIONSHIPS Frequency Percent Valid Percent Cum. Percent Valid 1 9 .7 .7 .7 2 38 2.9 3.0 3.7 3 67 5.2 5.3 8.9 4 229 17.7 18.0 26.9 5 316 24.4 24.8 51.7 6 325 25.1 25.5 77.2 7 291 22.5 22.8 100.0 Total 1275 98. 4 100.0 Missing 21 1.6 Total 1296 100.0 QUALITY OF FACULTY RELATIONSHIPS Frequency Percent Valid Percent Cum. Percent 9 .7 .7 .7 21 1.6 1.6 2.4 60 4.6 4.7 7 .1 188 14.5 14.7 21.8 299 23.1 23.5 45.3 374 28.9 29.3 74.6 324 25.0 25.4 100.0 1275 98.4 100.0 21 1.6 1296 100.0 HOURS SPENT STUDYING Frequency Percent Valid Percent Cumulative Percent Valid 0= None 15 1.2 1.2 1. 2 1= 1 5 hours 576 44.4 45.1 46.2 2= 6 10 hours 373 28.8 29.2 75.4 3= 11 20 hours 229 17.7 17.9 93.3 4= 21 30 hours 65 5.0 5.1 98.4 5= More than 30 hours 20 1.5 1.6 100.0 Total 1278 98.6 100.0 Missing System 18 1.4 Total 1296 100.0
51 APPENDIX C : ZERO ORDER PEARSON CORRELATIONS FOR VARIABLES USED IN ANALYSES GPA Black Age Female Married Has Child Skips Class College Activitie s Uses Advising Qual. of Peer Rel. Qual. of Fac. Rel. Hours Studying GPA 1 .067 .192 ** .089 ** .1 80 ** .078 ** .227 ** .069 .031 .154 ** .239 ** .199 ** Black .067 1 .100 ** .020 .013 .146 ** .056 .064 .158 ** .014 .053 .012 Age .192 ** .100 ** 1 .071 .451 ** .422 ** .275 ** .049 .089 ** .066 .135 ** .203 ** Female .089 ** .020 .071 1 .091 ** .154 ** .134 ** .057 .070 .068 .067 .100 ** Married .180 ** .013 .451 ** .091 ** 1 .357 ** .193 ** .035 .013 .047 .089 ** .186 ** Has Child .078 ** .146 ** .422 ** .154 ** .357 ** 1 .156 ** .045 .057 .043 .112 ** .122 ** Skips Class .227 ** .056 .275 ** .134 ** .193 ** .156 ** 1 .003 .114 ** .118 ** .165 ** .236 ** College Activities .069 .064 .049 .057 .035 .045 .003 1 .116 ** .074 ** .008 .059 Uses Advising .031 .158 ** .089 ** .070 .013 .057 .114 ** .116 ** 1 .109 ** .123 ** .079 ** Qual ity of Peer Rel ations .154 ** .014 .066 .068 .047 .043 .118 ** .074 ** .109 ** 1 .411 ** .120 ** Quality of Fac. Rel ations .239 ** .053 .135 ** .067 .089 ** .112 ** .165 ** .008 .123 ** .411 ** 1 .086 ** Hours Studying .199 ** .012 .203 ** .100 ** .186 ** .122 ** .236 ** .059 079 ** .120 ** .086 ** 1
52 ABOUT THE AUTHOR Warren Timothy Smith was born to Olive B. Smith and James Warren Smith on June 8, 1966 in Queens New York. He was reared in the Rochdale Village Housing Development in South Jamaica. He atte nded St. Agnes Cathedral High School in Rockville Center Long Island where he was a three year letterman in basketball Warren has one younger brother who he considers to be his best friend since December 22, 1969. Warren graduated from high school in 19 84 and went on to play college basketball for Central Florida Community College in Ocala Florida where he earned an Associate of Arts Degree. He continued his education as a student athlete at Charleston Southern University in Charleston South Carolina, w here he earned a Bachelor of Science Degree in Political Science and Sociology in 1990 After completing his degree, Warren returned to Central Florida Community College to work with student athletes at the behest of the Vice President Dr. Issac Williams Warren considers Dr. Williams to be his first mentor For the past twenty years Warren has worked for the College Reach Out Program (CROP) at three community colleges in the state of Florida. He is currently manager of the C ollege R each O ut P rogram at Hillsborough Community College in Tampa Florida. Warren has assisted more than 100 high school students earn college admission to the universities of their choice Success is a Habit!