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 Ka
controlfield tag 001 001916966
007 cr mnu|||uuuuu
008 071115s2006 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001806
Maximizing the educational effects of collaborative learning :
b the role of vested interest
h [electronic resource] /
by Christina Partin.
[Tampa, Fla] :
University of South Florida,
ABSTRACT: This study, using a quasi-experimental research design, investigates connections between pedagogy and social psychology by applying social psychological theories of group work and interaction to collaborative learning, a current trend in pedagogical techniques. It was hypothesized that by creating a setting in which students would be evaluated based in part on the performance of their peers would improve their individual performance. The incentive (a percentage of their grade) would hypothetically motivate students to teach their peers effectively; thus they would be taking a vested-interest role in becoming a co-teacher to their partner. This study was implemented by examining two sections of Introduction to Sociology which were taught concurrently and in exactly the same manner, with the only difference between the classes being the vested-interest feature present in the experimental class and absent from the control class. While this technique was determined not to have any statistical significance on the students' final grades, it did indicate that other factors involved in group work and collaborative learning might influence student outcomes or perceived student outcomes. Students in the experimental course exhibited more signs of anxiety about their grades, expressed more concern about their partners' abilities, and gave the instructor significantly lower ratings than the control class. However, students in the experimental class also came to class more often. These findings may indicate that placing a grade on group work, while effective in encouraging attendance, does not significantly alter the output of the group. Instead, this increased pressure about partners' performance may diminish the effectiveness of the group as students tend to see that the performance of their partners as outside of their own control.
Thesis (M.A.)--University of South Florida, 2006.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 56 pages.
Adviser: Maralee Mayberry, Ph.D.
t USF Electronic Theses and Dissertations.
Maximizing the Educational Effects of Collaborative Learning: The Role of Vested Interest by Christina Partin A thesis 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: Mara lee Mayberry, Ph.D. Michael Kleiman, Ph.D. James Cavendish, Ph.D. Date of Approval: November 3, 2006 Keywords: education, teaching techniques, social psychology, peer-learning, pedagogy Copyright, 2006, Christina Partin
ACKNOWLEDGEMENTS I would like to say thanks to the many people who have offe red their insights in this project. I specifically want to thank Mara lee Mayberry for her continued encouragement and support of this research. Without her, ne ither this project nor my confidence in my own abilities could have been brought to fr uition. Additionally, I would like to thank Mike Kleiman for helping me with the c onstruction of this study and for his early confidence in its importance, Jim Cavendish for his continued enthusiasm in working with me to interpret results, and Richard Gagan for always being available to bounce ideas back and forth. IÂ’m also pleased to thank my mom, Robbie, and the rest of my family who have been behind me, reminding me that a researcher can only become successful by keeping a positive attitude. Last ly, I am most grateful and would like to offer my sincerest thanks to Larry, my best friend who has been a pi llar of strength to me not only in my academic endeavors but throughout my life as well.
i TABLE OF CONTENTS LIST OF TABLES iii LIST OF FIGURES iv ABSTRACT v INTRODUCTION 1 Classroom Organization 1 Collaborative Learning 3 THEORETICAL BACKGROUND/REV IEW OF LITERATURE 6 Educational Pedagogy 6 Social Psychology 7 Matching Students 10 RESEARCH QUESTIONS 12 RESEARCH METHOD AND DESIGN 14 Participants and Design 14 Control/Experimental Group Demographics 15 Pre-test 17 Matching Students 18 Control Group 19 Experimental Group 19 Dependent Variables 20 Independent Variables 21 Control Variables 21
ii Ensuring Internal Validity 22 Analysis of Data 22 FINDINGS 24 Quantitative Findings 24 Qualitative Findings 30 DISCUSSIONS AND CONCLUSIONS 33 REFERENCES 35 APPENDICES 37
iii LIST OF TABLES Table 1: Sex of Students Based on Group Placement 16 Table 2: Race of Students Based on Group Placement 16 Table 3: College of Students (A rts & Sciences or Other) Based on Group Placement 16 Table 4: Previous Number of Sociology Classes Taken by Student Based on Group Placement 17 Table 5: Net Relationships Between Vested-Interest Learning (Model I), Demographic Variables (Model II), Condition Variables (Model III), and Learning Outcomes Attained During First Half of the Semester 27 Table 6: Net Relationships Between Vested-Interest Learning (Model I), Demographic Variables (Model II), Condition Variables (Model III), and Learning Outcomes Attained During the Second Half of the Semester 28 Table 7: Net Relationships Between Vested-Interest Learning (Model I), Demographic Variables (Model II), Condition Va riables (Model III), and Learning Outcomes Attained During the Full Semester 29
iv LIST OF FIGURES Figure 1: Box Plot Showing Distri bution of Pretest Scores Based on Group Placement 18 Figure 2: Amount of Improvement from Pretest to Midterm; Pretest to Final Based on Group Placement 25 Figure 3: Absences Based on Group Placement 26
v Maximizing the Educational Effects of Collaborative Learning: The Role of Vested Interest Christina Partin ABSTRACT This study, using a quasi-experimental re search design, investigates connections between pedagogy and social psychology by appl ying social psychological theories of group work and interaction to collaborative learning, a cu rrent trend in pedagogical techniques. It was hypothesized that by creating a setting in which students would be evaluated based in part on the performance of their peers would improve their individual performance. The incentive (a percentage of their grade) would hypothetically motivate students to teach their peers effectively; thus they would be taking a vested-interest role in becoming a co-teacher to their partner. This study wa s implemented by examining two sections of Introduction to Sociology which we re taught concurrently and in exactly the same manner, with the only difference betw een the classes being the vested-interest feature present in the experimental cla ss and absent from the control class. While this technique was determined not to have any statistical significance on the studentsÂ’ final grades, it did indicate th at other factors invol ved in group work and collaborative learning might influence student outcomes or perceived student outcomes. Students in the experimental c ourse exhibited more signs of anxiety about their grades, expressed more concern abou t their partnersÂ’ abilities, and gave the instructor
vi significantly lower ratings than the control clas s. However, students in the experimental class also came to class more often. Thes e findings may indicate that placing a grade on group work, while effective in encouraging a ttendance, does not sign ificantly alter the output of the group. Instead, this increased pressure about partne rsÂ’ performance may diminish the effectiveness of the group as st udents tend to see that the performance of their partners as outside of their own control.
1 INTRODUCTION Collaborative learning is a t ype of cooperative classroom learning mechanism that has been widely employed in the last seve ral decades (Cohen, 1994). Many articles and publications have reported glowing results in learning outcomes from using cooperative techniques (Cohen, 1994; Johnson & Johnson, 1998). There is a gap in the literature regarding how theories from social psychology ca n be applied to colla borative learning to help us understand learning outcomes. The pres ent research serves to apply theories from social psychology to a collabora tive classroom setting to ex plore potential connections. Included are a thorough review of the coopera tive/collaborative literature, as well as theories and research from social psychol ogy that will enable me to address possible oversights in past research. Specificall y, this research will address the following questions: Is vested-interes t (an original concept with a background from social psychology, explained herein) an effective tech nique in increasing the potential benefits of collaborative learning? Ar e there particular circumst ances where vested-interest impacts learning outcomes? Classroom Organization Not all classrooms are the same. Teachers implement different classroom management styles based on their own pe rsonal pedagogy and methodology. In the classroom setting, the students are individuals but depending on the manner in which the classroom is arranged, the students can relate to each other in different ways. According
2 to Johnson and Johnson (1998), th ere are three ways that an individualÂ’s actions can relate to the actions of others: competition, individualism, or cooperation. These various patterns of contact affect how students interact, how students feel about the interaction, and the outcome of the interaction. When classrooms are organized in ways that encourage competition, as they traditionally have been, the success of an individual is dependent on the failure of others. This is the case when teachers grade on a curve or allow the best student to set the standard for the rest. In these class setti ngs, students do not see each other as coworkers or classmates, but as competitors who need to be eliminated to ensure individual success. This leads to a negative correlation between studentsÂ’ successÂ—as some studentsÂ’ grades increase, others are left performing (or at l east being evaluated) more poorly than they would have done in isolation from other st udents, or in a different class population (Johnson & Johnson, 1998). In such a setting, st udents may feel pressured to obstruct the success of others for personal be nefit and individual success. In comparison, when classrooms are organi zed in an individualistic manner, an individualÂ’s evaluation is irre levant to the goals of others. Student success is not contingent upon the success of others. While this technique may be easy to implement, it does not encourage teamwork. An individual w ill make decisions and exert effort only to the extent necessary to achieve his or her personal goals withou t any interest of others in mind. Students in this scenario are aware th at their performance will not influence the success of others, nor will they be influen ced by the performance of others. Students learn quickly that interactions are ancillary, or possibly even bothersome, and work alone to accomplish isolated goals (Johnson & Johnson, 1998).
3 Finally, in a cooperative sett ing students work together to achieve educational goals. With this technique, there is a positive correlation between studentsÂ’ success: an individual can succeed only if others in th e group are also succeeding. In this case, individuals think of their goa ls in terms of the greater good of the group instead of thinking of themselves only. An individualÂ’s actions can promote the success of others, and vice-versa. Students realize that the best way to succeed is to work together and to encourage all group members to achieve (Johnson & Johnson, 1998). Collaborative Learning Traditionally in college teaching, the focus of the class has been on the lectures of the professor. In competitive classroom settings, which are common in higher education, the students are graded accordi ng to their ability to retain information better than their peers. Recently, especially in the last tw o decades, researchers and educators have begun to challenge lecture-based classrooms that encourage competition and are utilizing other teaching methods in an effort to increase class performance as well as to help students gain life-skills. One method that has m oved into the spotlight of instruction is collaborative learning (Slavin, 1982; Wh itman, 1988; and Michaelsen, 2002). Collaborative learning is a va riation of cooperative le arning, which exists when students work together in groups to ac hieve common goals. Collaborative learning differs from cooperative learning in that it gives students more freedom to work independently from the instructor, within their group, while s till adhering to the principles of cooperative learning. Thus collaborative learning is a subgroup of cooperative learning. Most research has been done on cooperative learning, which
4 emphasizes more direct instruction from the teach er, but the principals are the same as in collaborative learning. For this reas on, the terms Â“cooperative learningÂ” and Â“collaborative learningÂ” are used interchangeably in this rese arch based on the previous research being cited. Johnson and Johnson (1998) have found that this collaborative technique creates higher achievement in the classroom, higher productivity, positive relationships between students, and increased psychologi cal health. With the influx of attention turned toward collaborative learning in education, it is first important to understand exactly what collaborative learning is and how it differs from traditional class settings. Since its onset, cooperative learning has generated a debate in education and research. Some believe that c ooperative learning is a way to shift the burden of preparing a class from the teacher and onto the stude nts (Felder, 1996). Others believe that cooperative learning is simply another wa y of reproducing the dominant discourse without implementing analytical skill bui lding into the curr icula (Mayberry, 1998). However, many people believe that cooperati on enhances the lear ning process and the education that a student receives (S lavin, 1982; Whitman, 1988, Johnson & Johnson, 1998; and Michaelsen, 2002). The research su pporting cooperative le arning as a viable or useful instructional method, as shown in Johnson and JohnsonÂ’s meta-analysis of instructional outcomes (1989), indicates that cooperative learners outperform learners from competitive or individualistic envir onments. The study reports that students in cooperative learning environments demonstrat e greater retention, gr eater willingness to take on difficult tasks, creative thinking (defin ed by an ability to generate new ideas and strategies that one would not have created on his or her own) and an increased ability to
5 apply newly learned information to prev iously learned information (Johnson and Johnson, 1998). It appears that the benefits to cooperative and colla borative learning are noteworthy. However, are the possible benefi ts of cooperative learning being maximized to their fullest potential?
6 THEORETICAL BACKGROUND/REVIEW OF EXISTING LITERATURE Educational Pedagogy Â“Vested-interest learningÂ” is a term that I have created to describe the techniques I propose implementing in the classroom. The term is original, but its foundations are based in constructivist theo ries of education (Habron, 2005; Milbrandt, 2004; Smith 1992), and interdependence theories from social psychology (Lewin, 1947). These theories share in common the emphasis on working together to achieve, but all are lacking in one essential area: studentsÂ’ material or objective interests in participation. Collaborative learning is one of many techniques along with cooperative learning, peer-led guided inquir y, and group-based or team-based learning, that share in common the goal of joint intellectual effort by students. These pedagogies are based in constructivist approaches to education where there is an Â“emphasis [placed] on the active social participation of the learner with th e environmentÂ” (Milbrandt, 2004). Students in collaborative learning environments are not assumed to be passive recipients of knowledge, but rather they are co-creators of knowledge. Th e constructivist approach also takes into account the adage that Â“we le arn best by teaching othe rs.Â” According to Whitman (1988), teaching others is an incredib le way for students to fully understand and gain insight into knowledge, because during th e process of teaching others, students must explain and reword meanings and give multiple examples to their peers. Whitman (1988) refers to this process as learning the materi al a second time, which leads to a deeper and fuller understanding and processing of mate rials covered in the classroom.
7 Constructivist approaches also consider the application of active learning. Active learning is a teaching style that encourag es students to look beyond their books and develop their critical thinki ng skills. Students are not re cipients of knowledgeÂ—they are asked to seek out the answers they look for and to question their own thoughts and beliefs, as well as to question the discourse presented by their instructors. Active learning ties inquiry into the classroom, and in conjunction with collaborative learning, helps students realize their fuller poten tials as students and as thinkers. Social Psychology Aside from the field of education, applic ations from social ps ychological theories are credited with much of the widespread su ccess of cooperative learning. Many theories from social psychology have been aptly applie d to cooperative learning to help explain and enhance its effects. Theories such as social interdependence, positive interdependence, and promotive interaction all have their pla ces in collaborative learning (Lewin, 1947; Johnson & Johnson, 1998). Social interdependence theory was deve loped initially by Kurt Koffka, a Gestalt Psychologist from the early 1900s, and later was expounded upon by one of his colleagues, Kurt Lewin (1947). Lewin believed that interd ependence is the product of common goals within a group, a nd that this interdependen ce creates a Â“dynamic whole,Â” which in essence characterizes a group as a system by which offsetting one member of the system can jeopardize or offset other me mbers in the group as well. Lewin (1947) also noted a tension which is present in gr oups and group members, which can lead to an
8 increase in group productivity and motivation. Group members may feel pressured to contribute to group work, and this would lead to an increase in productivity. According to Deutsch (1949), Â“social in terdependence exists when individuals share common goals and each individualÂ’s outcomes are affected by the actions of others.Â” Accordingly, Johnson and Johnson (19 98), differentiated social interdependence from social dependence (whereby the outco mes of one person are affected by the outcomes of a second person but not vice-vers a), and social independence (whereby the individualsÂ’ outcomes are affected by each othe rÂ’s actions). Social interdependence is present is competitive and cooperative situa tions, and the absence of either social interdependence or dependence leads to an individualistic situation (Johnson & Johnson, 1998). Positive interdependence Â“exists when one perceives that one is linked with others in a way so that one cannot succeed unless th e others do (and vice vers a) and/or that one must coordinate oneÂ’s efforts with the effort s of others to complete a taskÂ” (Johnson and Johnson, 1989). In other words, group members must realize that they not only have to perform on an individual level, but they also have a responsibility to be sure that their partner or group is performing as well. This is much aligned with the notion of Â“vestedinterestÂ” learning. Positive interdependence emphasizes individual accountability and personal responsibility. Individu al accountability incorporates an individualÂ’s personal responsibility for contributing a fair share of work, but it also includes that personÂ’s willingness to work in a group setting and to help promote group cohesiveness.
9 Promotive interaction (Johnson & Johnson, 1989) furthers our understanding of group dynamics and interaction among students in a cooperative learning environment. Promotive interaction can effectively be characterized as a process whereby: students [are] providing ot hers with efficient and effective help and assistance, as well as exchanging in formation and materials, providing each other with feedback, challenging each othersÂ’ conclusions in order to create the best final answer, work ing to achieve mutual goals, being trustworthy, being motivated to strive for mutual benefit, and work to decrease anxiety and st ress (Johnson & Johnson, 1989). Promotive interaction can efficiently be c onsidered an additive type task, where the performance of the group is depende nt on the sum of all group membersÂ’ contributions. Since individuals are not compe ting against each other, but rather working toward a common goal, the overall motivating factor of the group will be group success through which individual success will be achieved (Johnson and Johnson, 1998). Vested-interest learning is related to the c oncept of constructivist learning in that students will be actively working to gain know ledge instead of being passive recipients. They will use collaborative learning techniques to achieve this. Social interdependence comes into play when we examine the dynamics of the groups, and how they will work together to achieve group goals knowing th at individual outcomes are affected by the actions of others. Positive interdependen ce will give students a feeling of personal responsibility to their group, which will enhanc e the group effort. Promotive interaction helps guide the grade assignment of the gr oup work, since students will perform for group success as opposed to individual success. My specific research focus will be how social interdependence theory, along with positive interdepen dence and promotive
10 interaction, can be applied to the constructivist approach to education by placing a vested-interest on group work and success, th ereby giving a tangible and immediate value to group performance. By applying these specific theories to collaborative learning in a controlled setting, we should be able to see whether so cial psychological theories actually impact the collaborative learning environment, or the result from the environment as measured by the learning outcomes. One specific ar ea of social psychology that is only peripherally discussed in educat ion literature is group formation. This area is important to social psychology, and should help measur e understand what, if any, circumstances make vested-interest learning is effective. Matching students Previous literature on group work has deba ted the effectiveness of collaborative learning, specifically in questioning the a ssignment of group members and its effects on learning outcomes. Some researchers have suggested that students should be partnered with other students in mixed ability groups (Cumming, 1983). Theoretically, this would allow the weaker students to catch up to th e stronger students. Other researchers have suggested that students should be Â“streamed,Â” or paired with students of similar abilities (McKeachie, 1974). Based on ideologies in th e field of social psychology this method may be effective because it eliminates an academic burden from the stronger students (having to teach others) and from the weaker students (f eeling that they are holding others back). Additionally, studies have show n that in mixed ability groups, the stronger students tend to improve while the weaker stude nts remain at a constant (lower) level, as
11 group work facilitates the stronger studentsÂ’ ability to solidify information, while the weaker students do not have ample opportunities to explore and improve. In order to test both of these methods for effectiveness, this experiment includes both mixed and streamed ability groups. Another important issue that needs to be addressed in any group setting is social loafing. Social interdependen ce, positive interdependence, or promotive interactions can not take place in groups with social loafi ng problems. Overall group success can not be accomplished if some members are not contri buting. Social loafing is the phenomenon whereby a person in a group will contribute less since others in the group can compensate, will be effectively eliminated in this situation by using only two members, so that over or under compensation cannot go unnoticed. According to the results of a meta-analysis study of group performance (Kar au & Williams, 1993), social loafing is a pervasive phenomenon, but it does not occur when group members feel that the task or the group itself is important. The groups in th is vested-interest lear ning experiment will be made of co-peers (teachers who are at th e same level as their learners) who will work together and share responsibility for one another (Whitman, 1988). This, Karau and Williams (1993) suggest, will increa se the likelihood of group success.
12 RESEARCH QUESTIONS The research on collaborative learning show s that as students work together, their class performance increases as a result of constructive group experiences (Johnson & Johnson, 1998; Whitman, 1988; and Michaelsen, 2002). However, there seems to be a gap in the literature that examines studentsÂ’ desire to work in groups or their level of involvement and participation. It is assumed in the literature that st udents who are placed in groups will have the desire to work within that group, although ofte n times, there is no real desire, or reason to work together. Could it be that the students will do just enough group work to get by in the course, without maximizing their efforts? If group work becomes unfairly divided among stud ents, as is sometimes the case, or if work is parceled out to minimize actual group interaction, is group or peer learning effective? It is commonly said by teachers that they never truly learned a given material until they tried to teach it. The act of teaching other students may well be the biggest possible benefit of group work, but in half-hearted groups, is this stage being achieved? Will students put the extra effort into their group to create a learning/teaching environment if they see no tangible reward in doing so? W ill group participation and effort increase if students are offered a tangible and immediate reward? The literature from education maintains that collaborative learni ng is effective in enhancing student learning as measured by le arning outcomes. Theories from social
13 psychology can be applied, which, based on previ ous literature, should show that with the right conditions (social interdependence, appropriate group pairing, and promotive interations) collaborative learning should improve studentÂ’s grades. In order to explore these questions and to potentially offer tec hniques which may help maximize the benefits of collaborative learning, I propose a quasi-experimental fi eld study that introduces the concept of vested-interest learning. In what I call vested-interest learning, each student is paired with another student and a portion of the class time is allotted for the pairs to work together. In pairs, students can go through cl ass material to encour age understanding of the content. A studentÂ’s overall performa nce evaluation will reflect their degree of engagement in facilitating thei r partnerÂ’s performance. Thus, each individual will have a vested-interest in making sure that their partne r is learning and comprehending the course. Vested-interest learning makes each student responsible for teaching another, and by teaching another, it is hypothesized th at students will gain a better understanding of the information taught in class than w ill their counterparts in traditionally taught collaborative learning classrooms. Applying principals from social psychology, in the form of vested-interest learning, to the coll aborative learning da ta can strengthen the literature in collaborative le arning by offering practical in formation on if, and under what conditions, collaborative l earning is effective.
14 RESEARCH METHOD AND DESIGN In order to explore the infl uence of vested-interest le arning as an instructional technique on learning outcomes in collabo rative settings, I have conducted a quasiexperiment in which two sections of intr oductory level sociology classes were taught concurrently by the same instructor. These classes were taught at a large, urban public university with a diverse enrollment of ove r 40,000. The classes consisted of identical instructional material and c ourse layout and design, with the exception of the vestedinterest feature present in the experimental group and absent from the control group. Both groups used course material from a typical and approved Introduction to Sociology textbook. The control group was taught using st andard collaborative learning techniques, as described above, and the experimental group was set up in an identical fashion. However, in the experimental group, the studen ts were informed that their performance evaluation would be based, in part, on how well they facilitated partnerÂ’s learning, similar to traditional group work in which students mu st perform cooperatively to earn a shared grade. This gives them a stake in the performance of their peer. Participants and Design Subjects in this study consisted of two gr oups of participants from Introduction to Sociology classes at a large, urban univers ity. Each group (experimental and control) initially consisted of 45 students, but due to at trition the class sizes were reduced to 43 in
15 the experimental group and 42 in the contro l group (n=42 control, n=43 experimental, n=85 overall sample). It is important to note that while the participants self-selected their courses during an open enrollment period, ther e was no knowledge of course content or experimental/control group diffe rences prior to commencement of the course. Further, after explaining the syllabus and grading policies, no unusua l drop rate was observed. Additionally, these classes met on the same days, twice per week in the afternoon, temporally separated by only one and a half hours, so there isnÂ’t any reason to believe that classes would differentially attr act particular Â“typesÂ” of students. Control/Experimental Group Demographics The control group consisted of 42 studen ts. In the control group there were 13 males and 29 females, 29 of whom were wh ite students and 13 of whom were non-white students. Of these students 23 were majoring within the College of Arts and Sciences while 19 were majoring in other colleges. 7 st udents in this group had taken at least one other college level sociology class. The experimental group consisted of 43 st udents. In the experimental group there were 14 males and 29 females, 24 of whom were white students and 19 of whom were non-white students. Of these students, 24 were majoring within the College of Arts and Sciences while 19 were majoring in other colle ges. Six students in this group had taken at least one other college level sociology class (See Tables 1-4).
16 Table 1: Sex of Students Based on Group Placement Sex Male Female Total Count 13 29 42 Control % within grouping 31.0% 69.0% 100.0% Count 14 29 43 Grouping Experimental % within grouping 32.6% 67.4% 100.0% Count 27 58 85 Total % within grouping 31.8% 68.2% 100.0% Table 2: Race of Students Based on Group Placement Race White Non-white Total Count 29 13 42 Control % within grouping 69.0% 31.0% 100.0% Count 24 19 43 Grouping Experimental % within grouping 55.8% 44.2% 100.0% Count 53 32 85 Total % within grouping 62.4% 37.6% 100.0% Table 3: College of Students (Arts & Scie nces or Other) Based on Group Placement College Arts & Sciences Other Total Count 23 19 42 Control % within grouping 54.8% 45.2% 100.0% Count 24 19 43 Grouping Experimental % within grouping 55.8% 44.2% 100.0% Count 47 38 85 Total % within grouping 55.3% 44.7% 100.0%
17 Table 4: Previous Number of Sociology Classes Taken by Student Based on Group Placement Number of Previous Sociology Classes 0 1 2 3 4 Total Count 35 6 1 42 Control % within grouping 83.3% 14.3% 2.4% 100.0% Count 37 4 1 1 43 Grouping Experimental % within grouping 86.0% 9.3% 2.3% 2.3% 100.0% Count 72 10 1 1 1 85 Total % within grouping 84.7% 11.8% 1.2% 1.2% 1.2% 100.0% Pre-test Students were assigned a teammate based on their scores on a preliminary exam, which served to determine their ability level prio r to instruction in this course. This exam was multiple choice format, covering questions re lated to definitions or understandings of Introduction to Sociology level knowledge. Application questions were excluded from the exam on the basis that students could not adequately answer t hose questions without having knowledge of concepts first. The ex am consisted of 37 questions. The scores on the pre-test ranged from (13-29) correct answers in the experimental class, and (14-28) in the control class (See Figure 1). In compar ison to the control class, the experimental class had a slightly larger range of scores. The distribution of scores of the experimental class was negatively skewed, whereas the dist ribution of scores of the control group was positively skewed. However, this difference was not determined to be statistically significant.
18 Figure 1: Box Plot Showing Distributio n of Pretest Scores Based on Group Placement Matching Students Approximately 50% of the groups were made up of heterogeneous or mixed ability pairs, and the other 50% were made up of homogenous or similar (Â“streamedÂ”) ability pairs. To determine th e studentsÂ’ ability level, I l ooked at the statistics from the pre-test. Using simple sta tistical analysis such as m ean and standard deviation, I measured Â“how much differenceÂ” constitute s a heterogeneous pair. The standard deviation for both classes was between 3 a nd 4 points, so homogenous pairs had two students that scored within 3 points differen ce from one another, and heterogeneous pairs 43 41 N =groupingexperimental controlpretest40 30 20 10 32 31 13 34 1 79 62
19 consisted of partners who were more than 4 points difference from one another. Based on this criterion, partners were selected ra ndomly. Dividing students into these types of pairs provided a potential to indicate whether this technique would be beneficial to all students, or only students who were paired in a particular manner. Control Group Success in the course is based on student participa tion in group activities (determined by studentsÂ’ completing assignments given to the groups relating to students working with each other to teach [or help with understanding] material from the course); attendance (to ensure that group members are attending classes regularly); and two multiple choice (post) exams which will m easure learning. The exams will determine student success in terms of this experiment. The exam questions will be taken from a test bank that accompanies the book of choice. E ach post exam contains questions from the pretest, as well as others. The improvemen t on these repeat questi ons will show learning and retention. Experimental Group The experimental group was been taught exactly the same as the control group, with all assignments, activities, and exams identical in conten t. However, the grading of these classes was slightly different. In the e xperimental class, 15 pe rcent of the studentsÂ’ overall course grade comes from the pe rformance of his or her partner. Control Class Exam Score = _______ X 100% Experimental Class Exam Score = _______ X 85% PartnerÂ’s Score = _______ X 15%
20 The groups were treated differently in no ot her way. An outsider who was not aware of the experimental differences observed th ese classes on one occasion, and reported no differences in style or content. Dependent Variables Dependent variables in this study are pr imarily outcome variables in relation to the effect of vested-interest learning. The pretest used for each class was identical (see Appendix B). Although there is inevitably a chance in educ ation that students from classes will gain information fr om others, I do not feel a thre at to validity for the pretest because students were not anticipating th e test, it came unannounced, and they were informed that they were being graded on completion only, so improving oneÂ’s test score would not have improved that personÂ’s grade. The post tests were identical in content, although the ordering and formatting of the exams were different. Further, the experimental class was tested before the contro l class, so if exam information sharing was an issue, it would have skew ed results in favor of the control groupÂ—however, due to the close proximity of these classes in terms of time, I did not find this to be a problem. Thus, the dependent variables analyzed are: pretest score (coded as number correct), pretest part 1 and pretest part 2 (since the final was not cumulative, I divided the pretest into 2 partsÂ—the first part has questions that reappear on the midterm, the second part has questions that reappear on the final), midterm score (coded as number of questions taken from pretest part 1 that th e student answered correctly), final exam score (coded as number of questions taken from pretest part 2 that the student answer ed correctly), pretest to midterm difference (coded as the actual numer ical difference from pretest to midterm),
21 pretest to final exam difference (coded as the actual numerical difference from pretest to final), and final course grade (coded as actual final grade, in percentages). Independent Variables Specific research questions have been propos ed in this study: Is vested-interest an effective technique in increasing the potential benefits of collabor ative learning? Are there particular circumstances where vested-i nterest impacts learning outcomes? In order to address these questions, I am analyzing data from the three following independent variables: group placement (coded as experime ntal=1, control=2), placement of students in pairs (coded as homogenous=1, heterogene ous=2), and how much the participant liked their partner (coded as 1= Â“nothing/not at a llÂ”, 2= Â“not very muchÂ”, 3= Â“somewhatÂ”, 4= Â“quite a bitÂ”, 5= Â“very muchÂ”) against whet her or not the vested-interest impacted the grade as a measure of positive interd ependence and group overall performance. Control Variables In addition to grouping and types of pairs, other control variables considered and evaluated in this study include demographi c data, such as gender (coded males=1, females=2), race/ethnicity (coded white=1, non-white=2), college (coded other major major=1, Arts & Science =2) as well number of previous sociology classes taken (coded as actual number of classes), to determine if there is a relatio nship with the grouping variable. By looking at these variables, it is possible to determine if vested-interest impacts certain categories of students more th an others. Evaluating these variables may
22 also give teachers who use group assignmen ts valuable information about pairing strategies that may or may not work best. Ensuring Internal Validity Several steps were taken throughout this study to maintain as much internal validity as possible during the experiment. First, the exam questions were objective, since objective grading is more quantifiable and standard in grading. Second, the exam questions were selected from a test bank. Th is essentially eliminated any possible threat to internal validity due to experimenter eff ect, since the instructor effectively removed herself from the process of determining the knowledge that needs to be demonstrated for successful course completion. Additionally, the control group a nd the experimental group have been taught using PowerPoint te chnology, so that all written information expressed in the classes is the same and no group has an advantage of more or different information. Lastly, as a protective measure looking to attest to internal validity, an external observer sat through each class on one occasion without knowing which was the experimental group and which was the control group, and reported no differences in style or information distributed. Analysis of Data Data were analyzed using the Statistical Software Package for the Social Sciences (SPSS 14.0). The following independent variab les have been examined: group placement (experimental or control), pairing (heterog eneous ability pairs or homogeneous ability pairs), and partner liking in relation to the dependent variables (namely, the outcome
23 measures). Additionally, the following cont rol variables have been analyzed: sex, race/ethnicity, college, number of previ ous sociology classes ta ken, and number of absences. Using crosstabs, T-tests, ANOVAs and multiple regression analyses, I have analyzed these independent, dependent, and c ontrol variables to dete rmine whether or not there is a statistical signif icance in student performance based on the teaching method I have proposed, or under what ci rcumstances this method may be effective. The pre-test administered at the beginning of the course to determine student ability prior to instruction, and based on this I have been able to determine the amount of variance between the groups, and the amount of improvement over the span of the course. I tested the null hypothesis that instru ctional technique (c ollaborative versus collaborative with vested-interest) has no impact on student out comes, which has given me the ability determine the effectiveness (or lack thereof) of the vested-interest learning technique and make recommendations about its useful ness or potential for further study.
24 FINDINGS Quantitative Findings This vested-interest learning technique wa s determined not to have any statistical significance on the studentsÂ’ final grades t (83)=-0518, p =.606. However, it did indicate that other factors involved in group work and collaborative learning might influence student outcomes or perceived student outcome s. Students in the experimental course exhibited more signs of anxiety about their grades, expressed more concern about their partnersÂ’ abilities, and gave the instructor significantly lower ratings than the control class. However, students in the experimental class also came to class more often. These findings may indicate that pl acing a grade on group work, wh ile effective in encouraging attendance, does not significantly alter the output of the group. In the beginning of this experiment, students in the experimental group worked rigorously with their partners in an attempt to increase th eir own grades. There was a greater improvement in the experimental classÂ’ test scores from the pretest to the midterm exam t (82)=-2.521, p =.014 than the control classÂ’ test scores. The difference from the pretest to the final exam, however, did not show any significant differe nce (see Figure 2).
25 Figure 2: Amount of Improvement from Fi rst Half of Course; Second Half of Course, Based on Group Placement This could be either because the experime ntal group became less concerned about their partnersÂ’ performance, or the control group b ecame more concerned. Nonetheless, it seems to indicate that this vested-interest fe ature had an early adva ntage that wore off over the course of the semester. It could be reasonable to say that students respond well to this method for a while, but it could be more effective when used in combination with other methods. An interesting finding in this study is the differences between groupsÂ’ class attendance. In the experimental group, wher e partners worked together for a grade, attendance was significantly better than in the control group as demons trated in a one-tail t-test: t (82)=1.935, p =.028 (see Figure 3). This may indi cate that the students who had a
26 personal responsibility to another student (the experimental class) were more likely to come to class and exert an effort in the group work. Figure 3: Absences Based on Group Placement But again, the vested-interest feature, did not significantly increase the final grades in either course t (83)=.518, p=.606. After considering these findings, multiple regression analyses were conducted to determine how much each variable impacted the learning outcomes. These analyses, presented in Table 5, test the initial questio n: Â“under which circumstances would vestedinterest learning be effective?Â” The multiple regression analyses show that in the first half of the course, vested-interest impact ed studentsÂ’ grades (p=.007). Even when controlling for other variables (sex, race, coll ege, previous number of sociology courses), vested-interest show a significant (although sl ightly less significan t) relationship to student success (p=.004). In other words, t hose in the vested-inter est learning classroom showed a greater increase in scores between the pretest and the midterm compared to the control classroom, independent of the studentsÂ’ previous sociology classes, sex, race, or control experimental 0 5 10 15# of absences 0 5 10 15 Count 0 5 10 15 # of absences
27 college. Additionally, the studentsÂ’ pairing a nd degree of partner liking did not influence the significance of vested-int erest learning (p=.088; p=.233 respectively), but number of absences did show a significant effect on student learning outcomes (p=.009). When controlling for all of these f actors together, number of ab sences (p=.011) and vestedinterest learning (p=.028) are the biggest indicators of student success (see Table 5). Table 5: Net Relationships Between Vested-Interest Learning (Model I), Demographic Variables (Model II), Conditi on Variables (Model III), and Learning Outcomes Attained During Fi rst Half of the Semester Model I Model II Model III Block One Vested-Interest .302.220 .238** Block Two Female -.025 -.050 Non-White -.187 -.128 Arts & Sciences -.073 -.071 Previous Sociology Courses -.187 .189 Block Three Homogeneously Paired .172 Degree of Partner Liking .097 Number of Absences -.276* r2 values .091.172 .282 p=.05, **p=.01, ***p=.001 In the second half of the semester, the e ffect of vested-interest learning seemed to have waned, as the ANOVA and multiple regres sion analysis presente d in Table 6 shows (p=.406). This could be because students re alized that vested-interest learning was not influencing their grades as much as they initia lly expected it to, or they may have felt that if their partner was affecting their grade, they were not able to control that. These are
28 only speculations based on the da ta, but other information from the second half of the course is evident. For example, student attendance remains a significant predictor of success (p=.019). Even when taking all va riables into account (grouping, sex, race, college, previous number of sociology cla sses taken, pairing, a nd degree of partner liking) number of absences still accounts for a large portion of success as determined by student learning outcome measur es (p=.037) (see Table 6). Table 6: Net Relationships Between Vested-Interest Learning (Model I), Demographic Variables (Model II), Conditi on Variables (Model III), and Learning Outcomes Attained During th e Second Half of the Semester Model I Model II Model III Block One -.094-.073 -.122 Vested-Interest Block Two Female .231 .206 Non-White -.079 -.035 Arts & Sciences -.084 -.068 Previous Sociology Courses -.004 -.020 Block Three Homogeneously Paired .051 Degree of Partner Liking -.003 Number of Absences -.242* r2 values .009.066 .124 p=.05, **p=.01, ***p=.001 As shown in a multivariate regression analysis presented in Table 7, in overall learning throughout the semester (s um of first half of course and second half of course), vested-interest was not a statistically si gnificant way of increas ing student learning outcomes (p=.246). However, number of abse nces is highly significant in determining student learning outcomes (p=.001). When taki ng all other variable s into consideration,
29 attendance is still the highest predictor of success (p=.002) (see Table 7). These findings are consistent with previous studies on the positive impact that regular attendance has on student learning outcomes (Durden and Ellis 1995; Devadoss and Foltz 1996; Marburger 2001; Dolton, Marcenaro and Navarro 2003; Kirby and McElroy 2003). This may indicate that vested-interest learning is more effective when students are attending classes regularly. It can also be in ferred from this data that ve sted-interest may be a way for instructors to increase their class attendance. Table 7: Net Relationships Between Vested-Interest Learning (Model I), Demographic Variables (Model II), Conditi on Variables (Model III), and Learning Outcomes Attained During the Full Semester Model I Model II Model III Block One Vested-Interest .131.159 .069 Block Two Female .149 .115 Non-White -.179 -.109 Arts & Sciences -.108 -.096 Previous Sociology Courses -.126 -.139 Block Three Homogeneously Paired .150 Degree of Partner Liking .062 Number of Absences -.359** r2 values .017.091 .237 p=.05, **p=.01, ***p=.001 Students expressed varying degrees of concern regarding their partners throughout the course (see qualita tive findings). It was intere sting that when asked to rate how well they lik ed their partners, students in th e experimental group rated their partners higher (M=4.40 on a scale from 1-5) than the control group (M=3.67), t (83)=-
30 2.873, p =.003. When comparing how well the each student liked his or her partner against pairing (homogenous or heteroge nous) no significant difference was found whatsoever between homogenous pairs (M=4.1 5 on a scale from 1-5) and heterogeneous pairs (M=4.15) t (83)=.015, p=.988. Students also expressed di fferent views of the instructor based on their group placement (control/experimental). In the final course evaluations, the instructor was rated significantly higher by the control class when asked to rate the instructorÂ’s ability to communicate ideas and information (M=4.97 cont rol, M=4.72 experimental on a scale of 1-5) t (69)=2.127, p<.001. When asked to rate the instructorÂ’s ability to facilitate learning, however, students re ported no significant differe nce (M=4.91 control, M=4.72 experimental on a scale of 1-5) t (69)=1.444, p=.153. The overall rating that the students gave to the instructor were proven to be significantly di fferent (M=5.0 control, M=4.82 experimental on a scale of 1-5) t (69)=1.824, p=.036. Qualitative Findings While these data cannot be backed with numbers or statistical findings, it may be of interest to add that some qualitative differe nces between the classes were noted. At the end of the class, a satisfaction survey was ad ministered in which st udents were given the option of writing additional co mments. Some examples of comments from the control class include: Â“The partners were a good way to make everyone feel comfortable about asking questions.Â” Â“I would have liked to pick my own partner, not have pre-selected ones.Â”
31 I liked working with my partner because Â“surprisi ngly, if I did not understand something in class my partner or the teacher helped me a lot to understand the materials.Â” Some examples of comments from the experimental class include: Â“We should have 2 partners instead of 1.Â” Â“We should be able to switch partners if we want to since some people donÂ’t work well with partners.Â” Â“I donÂ’t think my grade should be based on my partnerÂ’s. I liked my partner and we worked well together, but what if I wasnÂ’t that lucky and got someone who never did [work] or come to class.Â” Â“I would like my partnerÂ’s success not to influence mine because it hurt my grade.Â” Â“I liked working with a partner becaus e it forced me to review my notes.Â” Â“I donÂ’t like my partnerÂ’s grade giving me points because I donÂ’t think that I should be punished for their mistakes.Â” Interestingly, no student actually suffered in terms of his or her final grades based on partner performance. Additionally, there wa s no significant difference between the mean final course grades between the control and experimental classes t (83)=.518, p=.606. However, in the experimental class, students felt that a portion of their grade was outside of their control, which seemed to create a sense of hostility and anxiety toward the group work assignment. This finding can be reflec ted in the quantitative findings discussed previously. For instance, in the final course evaluations the instructor was rated significantly higher by the control class when asked to rate the instructorÂ’s ability to
32 communicate ideas and information (M=4.97 cont rol, M=4.72 experimental on a scale of 1-5) t (69)=2.127, p<.001. The overall rating that the students gave to th e instructor did yield a significant difference (M=5.0 contro l, M=4.82 experimental on a scale of 1-5) t (69)=1.824, p=.036. These findings imply that students in the control class were more comfortable with the class format (instr uctorÂ’s ability to communicate ideas and information) and were more comfortable with th e instructor when they were in a situation where they did not feel that th e instructor was putting a part of their grade outside of their control. It can also be noted that since the attendance was significantly better in the experimental class than in the control class, a portion of the course evaluations might be reflecting dissatisfaction in that respect. Th is postulation, however, is speculation based on the data which has no gauge for measurement in this study.
33 DISCUSSION AND CONCLUSION This study has found that vested-interest may be a valid coll aborative classroom technique when used in conjunction with various implementations of collaborative assignments. However, this study revealed that vested-interest alone is not enough to significantly raise studentsÂ’ l earning outcomes. More importa ntly, this study reveals that there are specific conditions when vested-int erest learning is most effective, and that vested-interest may create cert ain classroom conditions that could potentially improve or hinder class performance. Attendance, based on the data in this study, has the greatest impact on studentsÂ’ grades. Both vested-i nterest learning and attendance (which was greater in the vested-interest class) show signi ficant net relationships to student learning outcomes. Therefore, based on these data, ve sted-interest learning does have an initial positive impact on student learning outcomes and it can be inferred that students will perform the best (at least initially) if they are in a vested-interes t learning classroom and they are attending class regularly. Even when the effect of vested-interest dissipates, attendance still influences st udent learning outcomes. This study makes a significant co ntribution to the literature on cooperative/collaborative learning by applying social psychological principles of group interactions to already establ ished theoretical frameworks in education. These findings may have additional implications for group project s. In situations where all students in a group are assigned a single group grade, it ma y be the case that classroom dynamics change, as students feel that other students impact their grades. The qualitative findings
34 in this study should be considered by a nyone using groups in their classrooms or researchers looking at group dynamics. This study calls for more research applying social psychology to collaborative learni ng environments to help give a better understanding of what conditions or circumst ances allow vested-interest, or group work with a shared grade, to be considered effective. One of the limitations of this study is that there was no gr oup that was taught using no collaborative exercises. Future research may wish to address this in an effort to compare the effects of collaborative learni ng with non-collaborative learning in addition to the vested-interest feature, or non-vested-i nterest being present in class. Additionally, after determining ability levels of students, this study used a random selection criteria to place students in pairs. Future research may wish to allow students to self-select their partners as a way to possibly decrease the overall class anxiety of working with an unknown person. This study also has broad implications for future research. For instance, this study looked at individual data in order to come to conclusions about the effects of vested-interes t learning, but future research coul d explore pairs as units of data analysis in an effort to determine whether ther e is a net effect of vested-interest that could help explain its results. Another possibl e implication for research could be the differences between actual learning outcomes and perceived learning outcomes related to vested-interest. As stated earlier, even t hough there were no significan t differences in the studentsÂ’ final grades, student s in the vested-interest group seemed more concerned and anxious about their grades, and some felt that partners hurt their indi vidual performance. These areas remained largely unexplored in th is study, but future research may determine interesting effects of vested-interest by investigating these proposed guidelines.
35 REFERENCES Campbell, Donald T., and Julian C. Stanley. (1963). Experimental and QuasiExperimental Designs for Research. Chicago: Rand McNally Co. Cohen, Elizabeth. (1994). Â“Restructuring the Classroom: Conditions for Productive Small Groups.Â” Review of Educational Research 64 (1): 1-35. Cumming, Geoff. (1983) Â“The Introductory Statistics Course: Mixed Student Groups Preferred to Streamed.Â” Teaching of Psychology 57 (1): 34-37. Deutsch, M. Â“Cooperation and Trust: Some Th eoretical Notes.Â” In M.R. Jones (Ed.), Nebraska Symposium on Motivation (pp. 275-319). Lincoln, NE: University of Nebraska Press. Devadoss, S., and J. Foltz. (1996). Â“Evaluat ion of factors influencing student class attendance and performance.Â” American Journal of Agriculture Economics 78 (3): 499-507. Dolton, P., O. D. Marcenaro, and L. Navarro. (2 003). Â“The effective use of student time: A stochastic frontier produc tion function case study.Â” Economics of Education Review 22 (6): 547-60. Durden, G. C., and L. V. Ellis. (1995). Â“The effects of attendance on student learning in principles of economics.Â” American Economic Review 85 (2): 343-46. Felder, Richard M., and Rebecca Brent. (2001) "Effective Strategies for Cooperative Learning." Journal of Cooperation and Coll aboration in College Teaching 10 (2): 69-75. Felder, Richard M. (1996). "Navigating the Bumpy Road to Student-Centered Instruction." College Teaching 44: 43-47. Habron, Geoffrey. (2005). "Infusing Constructi vist Learning in Fisheries Education." Fisheries 30 (4): 21-26. . Johnson, D.W. and R.T. Johnson. (1989). Cooperation and Competition: Theory and Research. Edina, MN: Interaction Book Company.
36 Johnson, David W., and Roger T. Johnson. ( 1998). "Cooperative Learning and Social Interdependence Theory." Social Psychological Applica tions to Social Issues. Karau, S. J. & Williams, K. D. (1993). Social loafing: A meta-analytic review and theoretical integration. Journal of Personality and Social Psychology 65: 681706. Kirby, A., and B. McElroy. (2003). Â“The effect of attendance on grade for first year economics students in University College Cork. The Economic and Social Review 34 (3): 311-26. Mayberry, Maralee. (1998). "R eproductive and Resistant Pe dagogies: The Comparative Roles of Collaborative Learning and Femi nist Pedagogy in Science Education." Journal of Research in Science teaching 35 (4): 443-459. Marburger, D. R. (2001). Â“Absenteeism and undergraduate exam performance.Â” Journal of Economic Education 32 (Spring): 99-110. McKeachie, W. J. (1974). Â“I nstructional Psychology.Â” Annual Review of Psychology 25: 161-193. Michaelsen, Larry K., Arletta Bauman Knight, and L. Dee Fink, eds. (2002). Team Based Learning: A Transformative Use of Small Groups. Westport: Praeger. Milbrandt, Melody K., Janet Felts, and Neda Abghari. (2004). "Teaching-to-Learn: A Constructivist Approach to Shared Responsibility." Art Education ser. 57 (5): 1933. Scott, Lewis E., and Lewis E. Jennifer. ( 2005). "Departing from Lect ures: An Evaluation of a Peer-Led Guided Inquiry Alternative." Journal of Chemical Education 82 (1): 135-139. Sharan, Yael, and Shlomo Sharan. (1992). Expanding Cooperative Learning Through Group Investigation Teachers College Press. Slavin, Robert E. (1982). Cooperative Learning: Student Teams Second ed. Washington, D.C.: National Education Association. Slavin, Robert E. (1983). Student Team Learning: A Prac tical Guide to Cooperative Learning Washington, D.C.: Nati onal Education Association. Smith, Barbara Leigh, and Jean T. MacG regor. (1992). "What is Collaborative Learning?" Collaborative Learning: A Sourcebook for Higher Education Whitman, Neal A. (1988). Peer Teaching: To Teach is to Learn Twice ERIC-ASHE.
38 Appendix A: Coded Values for Va riables Used in SPSS Analysis Independent Variables Variable Name Actual Coded Values Grouping (Experimental; vested interest or Control; non-vested interest) grouping 1=experimental 2=control Pairing (homogeneously or heterogeneously paired) pairing 1=homogeneous 2=heterogenous Partner Liking (How well each student rated liking his/her partner) partlike 1= Â“nothing/not at allÂ” 2= Â“not very muchÂ” 3= Â“somewhatÂ” 4= Â“quite a bitÂ” 5= Â“very muchÂ”
39 Appendix A (Continued) Dependent Variables Variable Name Actual Coded Values Pretest pretest Coded as actual score Pretest Part 1 (material that would be revisited in the first half of the course) ptpt1 Coded as actual number answered correctly for this portion Midterm Exam midterm Coded as actual score Pretest to Midterm Difference (change in scores from pretest part 1 to midterm) ptmiddif Coded as actual point difference Pretest Part 2 (material that would be revisited in the second half of the course) ptpt2 Coded as actual number answered correctly for this portion Final Exam final Coded as actual score Pretest to Final Difference (change in scores from pretest part 2 to final) ptfindif Coded as actual point difference Final Grade fin_grad Coded as actual final grade
40 Appendix A (Continued) Control Variables Variable Name Actual Coded Values Sex sex 1=male 2=female Race (As answered on a voluntary survey, using the same categories as the US census) race 1=white 2=nonwhite (coded this way due to the relatively small numbers of each minority group represented) College college 1= other 2= Arts & Sciences Previous Number of Sociology Classes Taken previous Coded as actual number of classes taken Number of Absences absences Coded as actual number of classes missed 1. name = Student's name (removed from data set for privacy purposes) 2. sex = Student's sex 3. race = As answered on a voluntary survey, using the same categories as the US census (recoded as white/nonwhite due to small percentages of each minority group represented) 4. class = Progress through academic program 5. college = A student of the Colle ge of Arts & Sciences or other 6. previous = Number of pr evious sociology classes taken 7. pairing = Paired with a similar ability partner (homogenous) or a different ability partner (heterogeneous). Some students have missing information here because they failed to take the pretest. 8. pretest = Overall score on pretest, out of 37 possible
41 Appendix A (Continued) 9. ptpt1 = Since the final was not cumulative, I divided the pretest into 2 partsthe first part has questions that rea ppear on the midterm, the second part has questions that reappear on the final. This is "P re-Test part 1, out of 18 possible" 10. midterm = number of questions (from vari able #9) that the student actually got correct. Also out of 18 possible. 11. ptmiddif = Difference from Pre-Test to Mi dterm. (Variable #10 minus variable #9.) This should show improvement from Pre-test to Midterm. 12. ptpt2 = Second half of the pre-test from wh ich questions reappeared on the Final. Out of 15 possible. 13. final = Number of questions (from variable #12) that the student actually got correct. Also out of 15 possible. 14. ptfindif = Difference from PreTest to Final. (Variable #13 minus variable #10.) This should show improvement fr om Pre-test to Final. 15. absences = Number of classes missed by each student. 16. = experimental group or cont rol group. Course section #005 is experimental, #007 is control. 17. partlike = the rating that each student gave their partner on the question "How much do you like your partner?" from 1-5 on a sta ndard Lickert scale with 1 being very little and 5 being very much. 18. fin_grad = the final course grade that th is student received (numerical value).
42 Appendix B: Pretest Exam Used to Establish Pair Placement Pre-test Multiple Choice Identify the letter of the choice that best co mpletes the statement or answers the question. ____ 1. Which of the following statements is TRUE? a. Culture and social structure are synonymous. b. Culture and society are synonymous. c. Culture is limited to the arts, music, and literature. d. Culture is what makes humans unique in the animal kingdom. ____ 2. Socially shared ideas about what is right are called: a. ideas b. ideologies c. norms d. values ____ 3. The tendency to judge other cultures as inferior in terms of one's own norms and values is termed: a. cultural imperialism b. cultural relativity c. cultural stereotyping d. ethnocentrism ____ 4. A socially-defined position in a group is a: a. social boundary b. social marker c. status d. structural location ____ 5. The way in which society defines how an individual is to behave in a particular status is a: a. normative obligation b. role c. sanction d. status set ____ 6. A position or rank that is assigned to a person at birth and cannot be changed is: a. a closed status b. a fixed status c. an achieved status d. an ascribed status ____ 7. The shaping of behavior throug h reward or punishment is called: a. conditioning b. identity reinforcement c. modeling d. symbolic representation
43 Appendix B (Continued) ____ 8. In the opinion of many, the most controversial agent of socialization in American society is the: a. community b. mass media c. peer group d. School ____ 9. Which of the following is the BEST example of a primary group? a. a classroom b. a family c. an office d. people stranded in an airport ____ 10. A group that consists of two members is a: a. dyad b. primary group c. secondary group d. triad ____ 11. The ways in which a society prevents deviance and punishes deviants are known as: a. law enforcement agencies b. moral entrepreneurs c. normative systems d. social control ____ 12. An act or omission of an act for which the state can apply sanctions is called: a. a crime b. anomie c. deviance d. stigma ____ 13. The probability that a person who h as served a jail term will commit additional crimes and be jailed again is called: a. backsliding b. recidivism c. recrimination d. retrogression ____ 14. The division of the members of a soci ety into layers based on such attributes as wealth, power, and prestige is termed: a. homogenization b. social stratification c. status differentiation d. status sorting ____ 15. Material objects or behaviors that indicate social status or prestige are: a. deference patterns b. identity markers c. status indicators d. status symbols
44 Appendix B (Continued) ____ 16. Which of the following resources is most equally distributed throughout the U.S. population? a. education b. income c. power d. wealth ____ 17. The term referring to the biological differences between males and females is: a. anatomy b. destiny c. gender d. Sex ___ 18. When mothers teach their daughters to behave in "feminine" ways, they actually are teaching them: a. ancillary roles b. androgynous roles c. gender roles d. subordinate roles ____ 19. An ideology that justifies prejudice or discrimination based on gender is referred to as: a. ageism b. nativism c. racism d. sexism ____ 20. Race is essentially a: a. religious ideology b. geographic concept c. social concept d. mathematical principle ____ 21. The civil rights movement was a struggle to gain a. educational freedom b. equality of opportunity c. occupational opportunities d. right to own property ____ 22. The population of Native Americans in North America was reduced from over four million in the eighteenth century to below 600,000 in the early twentieth century as a result of: a. affirmative action b. amalgamation c. assimilation d. genocide ____ 23. In the modern world, economic r esources are increasingly controlled by: a. colonial powers b. government agencies c. great empires d. multinational corporation
45 Appendix B (Continued) ____ 24. The ability to control the behavior of others, even against their will, is termed: a. authority b. coercion c. influence d. power ____ 25. Power whose exercise is governed by the norms and statuses of institutions is referred to as: a. authority b. coercive power c. influence d. permissive power ____ 26. The time required for social institutions to adapt to major technological change is referred to as: a. cultural lag b. cultural regression c. structural disequilibrium d. technological dualism ____ 27. Initially, hospitals were: a. military establishments b. penal colonies c. religious centers d. workhouses ____ 28. A group of people related by blood, ma rriage, or adoption is referred to as a: a. consanguine group b. family c. kinship network d. sibling set ____ 29. Which of the following is the most frequently reported issue in American policing? a. murder b. burglary c. rape d. domestic violence ____ 30. Any set of coherent answers to the dilemmas of human existence that makes the world meaningful is called: a. a church b. a cognitive map c. an ideology d. a religion ____ 31. The term used to describe phenom ena that are not considered sacred is: a. holy b. mundane c. profane d. Secular
46 Appendix B (Continued) ____ 32. Tracking programs in educational institutions are thought to contribute to: a. educational success for all students. b. educational inequality. c. educational equality. d. none of these ____ 33. The increase in the number of pr ivate communities is an indication of a. community revisioning b. urban renewal c. gentrification d. fear of urban life ____ 34. Life expectancy is defined as a. the number of years one lives b. the number of years one plans to live c. the number of years one can expect to live d. the difference in life span and life years ____ 35. Which of the following refers to the effects of society on the natural environment? a. environmental stress b. pollution c. structural disequilibrium d. technological displacement ____ 36. An intentional effort by a group to cr eate new institutions or reform existing ones is a: a. protest movement b. revolution c. riot d. social movement ____ 37. Compared to today, war historically a. was more devastating b. was less devastating. c. used more sophisticated technology. d. used more women warriors.
47 Appendix B (Continued) Pre-test Answer Section MULTIPLE CHOICE 1. ANS: D 2. ANS: D 3. ANS: D 4. ANS: C 5. ANS: B 6. ANS: D 7. ANS: A 8. ANS: B 9. ANS: B 10. ANS: A 11. ANS: D 12. ANS: A 13. ANS: B 14. ANS: B 15. ANS: D 16. ANS: A 17. ANS: D 18. ANS: C 19. ANS: D 20. ANS: C 21. ANS: B 22. ANS: D 23. ANS: D 24. ANS: D 25. ANS: A 26. ANS: A 27. ANS: C 28. ANS: B 29. ANS: D 30. ANS: D 31. ANS: C 32. ANS: B 33. ANS: D 34. ANS: C 35. ANS: A 36. ANS: D 37. ANS: B
48 Appendix C: Voluntary Survey Questio nnaire Used to Gather Demographic Data on Students Name:_____________________________________ Student ID Number: _U________________________ Educational Data Class Standing at the BEGINNING of this semester: (check one) Freshman Sophomore Junior Senior Non-Degree Seeking Your Course Load: (check one) Full-time (12 credit hours or more) Part-time (Less than 12 credit hours) Your College: (check one) Architecture & Community Design Arts & Sciences Business Administration de la Parte Institute (FMHI) Education Engineering Health Sciences Honors College Marine Science Medicine Nursing Public Health Visual & Performing Arts Other: _____________________________________ Please list all of the previous SOCIOLOGY classes you have taken:
49 Appendix C (Continued) Your Major: (Circle One) Accounting Africana Studies American Studies and Humanities Anthropology Applied Sciences Art Studio and Art History Athletic Training/Sports Medicine Biology & Microbiology Biomedical Science Chemical Engineering Chemistry Civil and Environmental Engineering Classics Communication Communication Sciences and Disorders Computer Science and Engineering Criminology English Environmental Science and Policy Finance Foreign Languages General Business Administration Geography Geology Gerontology Government and International Affairs History Honors College Research Hotel and Restaurant Management Industrial & Management Systems Engineering Information Systems Information Technology Interdisc Classical Civilizations Interdisciplinary Natural Science Interdisciplinary Social Sciences International Business International Studies Liberal Studies Management Management Info. Systems Marketing Dance Dance Studies Early Childhood Education Economics Economics Electrical Engineering Elementary Education Mass Communications Mathematics Mechanical Engineering Medical Technology Microbiology and Biology Music Music Education Music Studies Nursing Philosophy Physical Education Physics Political Science Pre-Law Pre-Medical, Pre-Dental, & Pre-Veterinary Psychology Religious Studies Secondary Education Social Work Sociology Special Education Technical Education Theatre Women's Studies Other__________________
50 Appendix C (Continued) Personal Data: Your Date of Birth: ________/_______/19_______ Month Day Year Your sex: (check one) Male Female Are you Spanish/Hi spanic/Latino? Mark the "No" box if not Spanish/Hispanic/ Latino. No not Spanish/Hispanic/Latino Yes, Mexican, Mexican Am., Chicano Yes, Puerto Rican Yes, Cuban Yes, other Spanish/Hispanic/Latino Â— Print group. What is your race? Mark one or more races to indicate what you consider yourself to be. White Black, African Am., or Negro American Indian or Alaska Native Â— Print name of enrolled or principal tribe. __________________________________________________________ __________________________________________________________ Native Hawaiian Guamanian or Chamorro Samoan Other Pacific Islander Â— Print race. ___________________________________________________________ ___________________________________________________________ Asian Indian Chinese Filipino
51 Appendix C (Continued) Japanese Korean Vietnamese Continued on next pageÂ…Â… Other Asian Â— Print race. __________________________________________________________ __________________________________________________________ Some other race Â— Print race. __________________________________________________________________ __________________________________________________________________ What is your marital status? Now married Widowed Divorced Separated Never married What is your ancestry or ethnic origin? (For example: Italian, Jamaican, African Am., Cambodian, Cape Verdean, Norwegian, Dominican, French Canadian, Haitian, Korean, Lebanese, Polish, Nigerian, Mexican, Taiwanese, Ukrainian, and so on.) Do you speak a language other than English at home? Yes No b. What is this language? _______________________________________________ (For example: Korean, Italian, Spanish, Vietnamese)
52 Appendix C (Continued) c. How well do you speak English? Very well Well Not well Not at all Where were you born? In the United States Â— Print name of state. ___________________________________________ Outside the United States Â— Print name of foreign country, or Puerto Rico, Guam, etc. _____________________________________________________________________
53 Appendix D: Survey Questio nnaire, Including Item Wordings and Coded Values From Which Variables Used in Analysis are Taken Please answer the following questions as hones tly as possible. Your answers will not affect your gradeyou will receive credit for completion. Do not feel like you will be judged for your answers Â– I am seeking your honest input for several reasons. First, this will help me develop future courses. Second, I hope to find out if activities and information in this class ha ve been useful to you. Using a scale of 1-5, with 1 meaning Â“not hing/not at all,Â” 2 meaning Â“not very much,Â” 3 meaning Â“somewhat,Â” 4 meaning Â“quite a bit,Â” and 5 meaning Â“very much,Â” please answer the following questions: 1. How much do you feel you knew about Sociol ogy before the first day of class? (circle one) 1 2 3 4 5 2. How much do you think you know now? 1 2 3 4 5 3. How much or how often can you apply th ings from this class to your life? 1 2 3 4 5 4. How much did you like the vi deos and media clips? 1 2 3 4 5 5. How much did you like your partner? 1 2 3 4 5 Write comments in the spaces provided: 6. If I could change anything a bout the class, it would be_________________________ because______________________________________________________________ _____________________________________________________________________ _____________________________________________________________________
54 Appendix D (Continued) 7. If I could keep something about the class the same, it would be________________________because____________________________________ _____________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ 8. To improve the Clarification Question assignments, the best thing to do would be___________________________________________________________________ _____________________________________________________________________ _____________________________________________________________________ 9. Any other feedback youÂ’d like to provide?
55 Appendix E: Instructor Evaluation Ra tings Â– Including Wordings and Coded Values From Which Variables Used in Data Analysis are Taken Survey Question Variable Name Actual Coded Values Description of course objectives and assignments rating1 5=Excellent 4=Very good 3=Good 2=Fair 1=Poor Communication of ideas and information rating2 5=Excellent 4=Very good 3=Good 2=Fair 1=Poor Expression of expectation for performance rating3 5=Excellent 4=Very good 3=Good 2=Fair 1=Poor Availability to assist students outside of class rating4 5=Excellent 4=Very good 3=Good 2=Fair 1=Poor Respect and concern for students rating5 5=Excellent 4=Very good 3=Good 2=Fair 1=Poor Stimulation of interest in course rating6 5=Excellent 4=Very good 3=Good 2=Fair 1=Poor
56 Appendix E (Continued) Facilitation of learning rating7 5=Excellent 4=Very good 3=Good 2=Fair 1=Poor Overall rating of instructor rating8 5=Excellent 4=Very good 3=Good 2=Fair 1=Poor