xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E73UQH955_PL3VVW INGEST_TIME 2014-10-03T21:32:57Z PACKAGE UFE0046526_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
1 T HE R OLES OF S OCIAL AND A CADEMIC E NGAGEMENT AND P ERCEPTIONS OF B ELONGING IN A M ODEL OF S TUDENT P ERSISTENCE By JEAN ELIZABETH STAROBIN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF EDUCATION UNIVERSITY OF FLORIDA 2014
2 2014 Jean Elizabeth Starobin
3 To my children, Sonia, Max, and Charlie your belief in me inspired me. To my always encouraging parents, Sidney and Evelyn you taught me the importance of education. And to my most loving and supportive husband, Kirk this dissertation is dedicated to all of you.
4 ACKNOWLEDGMENTS I would first l ike to express my utmost gratitude to Dr. David Miller for not only serving as my dissertation committee co chair, but also as my teacher and mentor in approaching what felt like an insurmountable task. Your attitude t hat no question was too small or too insignificant to ask, helped me tremendously. I would also like to thank my other co chair, Dr. Arthur Sandeen. Your calm demeanor, interest, and breadth of knowledge were instrumental in my success as a student and res earcher. Your candor and critique of my work only improved the end result. In addition, I would like to recognize and show my sincere appreciation to Dr. Dale Campbell and Dr. Scot Smith for serving as members of my committee. Although I had the profession al support of my official committee members, I would like to thank an unofficial committee member who also supported me through this process: Dr. Jeanna Mastrodicasa. Her encouragement to just finish her advice during the process, and most importantly h er assistance in accessing my data was invaluable. A special thank you for the camaraderie and support of my fellow Student Affairs professionals and doctoral students: Maureen Miller, Mary C. Jordan, Cliff Haynes, and Tim Wilson. Thank you for all the c hats, lunches, and coffee that so helped to encourage me to keep working and not give up. I would also like to recognize my work family for their encouragement in finishing, and providing me with an environment that promoted my professional growth and deve lopment. Finally, I cherish the opportunity to take this moment to express my love and appreciation to my family: my parents Evelyn and Sidney, my children Sonia, Max, and Charlie, and most especially my husband Kirk. You each acted to inspire and en courage me; your unconditional love, unwavering support, and innumerable sacrifices
5 enabled me to pursue and achieve this goal. Thank you for being my best cheering squad; I could not have done this without you.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIS T OF FIGURES ................................ ................................ ................................ .......... 9 LIST OF DEFINITIONS ................................ ................................ ................................ 10 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 14 Statement of the Problem ................................ ................................ ....................... 22 Purpose of the Study ................................ ................................ .............................. 23 Research Questions ................................ ................................ ............................... 25 Significance of Study ................................ ................................ .............................. 26 Limitations of the Study ................................ ................................ ........................... 28 2 LITERATURE REVIEW ................................ ................................ .......................... 29 Historical Overview of Attrition Research ................................ ................................ 29 Models of Student Departure ................................ ................................ .................. 31 ................... 33 Impact of Involvement ................................ ................................ ............................. 37 Involvement Assessments ................................ ................................ ...................... 39 The Student Experience in the Research University (SERU) ................................ 39 The National Survey of Student Engagement (NSSE) ................................ ............ 40 Other Instruments ................................ ................................ ................................ ... 41 Conceptual Model ................................ ................................ ................................ ... 41 Student Characteristics ................................ ................................ ........................... 42 Minority Students ................................ ................................ .............................. 42 Generational Status ................................ ................................ .......................... 43 First Year Students ................................ ................................ .......................... 44 Social and Academic Integration ................................ ................................ ...... 45 Perceptions of Belonging ................................ ................................ .................. 46 Chapter Summary ................................ ................................ ................................ ... 47 3 METHODS ................................ ................................ ................................ .............. 50 Research Methodology ................................ ................................ ........................... 50 Popul ation and Sample ................................ ................................ ........................... 50 Measures ................................ ................................ ................................ ................ 51
7 Dependent Variables ................................ ................................ ........................ 51 Independent Variables ................................ ................................ ..................... 51 Academic Integration ................................ ................................ ........................ 51 Social Integration ................................ ................................ .............................. 52 Perceptions ................................ ................................ ................................ ...... 53 Background Characteristics ................................ ................................ .............. 54 Analysis Methods ................................ ................................ ................................ .... 54 Logistic Regression ................................ ................................ .......................... 54 Reliability ................................ ................................ ................................ .......... 55 Validity ................................ ................................ ................................ .............. 55 Participants and Sampling Procedure ................................ .............................. 56 Study Limitations ................................ ................................ .............................. 57 4 RESULTS ................................ ................................ ................................ ............... 59 Descriptive Statistics ................................ ................................ ............................... 59 Inferential Statistics to Explo re Differences between Retained Students Versus Non Retained Students ................................ ................................ ....................... 61 Factor Analysis of the Make up of Survey Items for Academ ic Engagement and Perceptions of Belonging ................................ ................................ ..................... 63 Logistic Regression ................................ ................................ ................................ 66 Chapter Summary ................................ ................................ ................................ ... 68 5 SUMMARY, CONCLUSIONS, AND RECOMMENDAT IONS ................................ .. 70 Summary of the Findings ................................ ................................ ........................ 70 Discussion of the Findings ................................ ................................ ...................... 74 Recommendations for Future Research ................................ ................................ 76 Implications for Practitioners ................................ ................................ ................... 77 Conclusion ................................ ................................ ................................ .............. 78 LIST OF REFERENCES ................................ ................................ ............................... 80 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 90
8 LIST OF TABLES Table page 4 1 Percentage of students by race/ethnicity and retention status ........................... 60 4 2 Percentage of students by gender and by retention status ................................ 60 4 3 Percentage of students by living on campus and retention status ...................... 60 4 4 Parental education and retention status ................................ ............................. 60 4 5 Descriptive statistics of the continuous independent variables ........................... 61 4 6 Tests of Normality of the continuous ind ependent variables .............................. 61 4 7 Academic Engagement Factor Analysis ................................ ............................. 64 4 8 Perceptions of Belonging Factor Analysis ................................ .......................... 65 4 9 Summary of Full Model of Logistic Regression Analysis for Predicti ng Retention of Freshmen and Sophomores ( N = 852) ................................ ........... 67 4 10 Summary of Reduced Model of Logistic Regression Analysis for Predicting Retention of Freshmen and Sophomores ( N = 955) ................................ ........... 67
9 LIST OF FIGURES Figure page 2 1 Integrated Model of College Student Persistence. ................................ .............. 49 2 2 Conceptual Model: Model of Student Retention ................................ ................. 49
10 LIST OF DEFINITIONS Attrition R efers to the act of leaving or dropping out of the institution of higher education. Co curricular activities D efined as activities that complement the classroom but occur outside the classroom. Disengagement Refers to the opposite of engagement or student engagement. Studen (Hu and Kuh, 2002, p. 556). Engagement Also (Hu and Kuh, 2002, p. 555). Extra curricular activities D escribe s activities that occur outside the classroom First generationa l student Is defined as a student who is first in their family to attend college. First time in college Is defined as a student who has never attended college or any other postsecondary institution. First year experience Can describe a program designed t o address the academic and social needs of the first year student at a college. It also can refer to the culmination of experiences and outcomes of the first year student at a college. First year student Is defined as an undergraduate student enrolled in an institution of higher education with less than thirty hours. Florida Opportunity Scholars Program Is a program designed specifically for first generation students from low income backgrounds. Also known by the acronym FOS program. Four year institution of higher education Refers to four year colleges and universities that offer bachelor level degrees. Four year institutions also can confer master level, doctoral level and professional degrees.
11 Involvement As of physical and psychological 1999a). NCHEMS I s the acronym for the National Center for Higher Educational Management Systems NCES I s the acronym for the National Center for Educational Statistics Non first generation student Is defined as a student who is not the first in their family to attend college. NSSE Is the acronym for the National Survey of Student Engagement. Persistence rate Is the percentage of students who are enrolled from one year to the reenrollment in college, whether continuous from one term to the next or temporarily interrupted and then T erenzini, 2005). The definitions of persistence and retention overlap with regards to this point of time. Retention Describes when a student is retained at an institution of higher education. SERU Is the acronym used for the Student Experience in the Research University Survey. This survey was developed for the University of California System by the University of California Berkeley Center for Studies in Higher Education. Satisfaction Refers to the l evels of happiness and fulfillment the student experiences from their academic, co curricular and extra curricular experiences at the institute of higher education. Student learning outcomes Refers to the knowledge, abilities and skills that a student acq uires as a result of their participation in an educational program or experience. Student success Is a term that refers to a variety of student indicators such as improved GPA, student engagement, achievement of academic or personal goals in an institutio n of higher education (Pascarella & Terenzini, 2005).
12 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Education T HE R OLES OF S OCIAL AND A CADEMIC E NGAGEMENT AND P ERCEPTIONS OF B ELONGING IN A M ODEL OF S TUDENT P ERSISTENCE By Jean Elizabeth Starobin May 2014 Chair: David Miller Cochair: Carl A. Sandeen Major: Higher Education Administration The present quantitative study investigated the relationship between academic engagement, social engagement, and perceptions of belonging on retention among freshman and sophomore students at a major Florida university The research explored self reported levels of academic engag ement, extra curricular activities, and perceptions o f belonging as reported on the s pring 2011 Student Experience in the Research University survey (SERU) In addition, the study measured the predictive value of patterns of academic engagement, social eng agement, perceptions of belonging a nd enrollment in the f all 2012. An analysis of both past and extant literature concerning retention issues revealed that retention is s academically at the institution. This study employed demographic and enrollment and self reported data con cerning engagement behaviors, as well as perceptions of belonging at the institution. Retained students were identified through f all 2012
13 enrollment records. Inferential statistics were employed to create a portrait of retained and non retained students. L ogistic regression analysis was completed to generate a parsimonious model predicting college student retention. Results from this research indicated that social engagement is a far greater predictor of retention than academic engagement or perceptions of belonging. Additionally, data from this study indicated specific demographic characteristics influence retention: race/ethnicity, parental education, and socio economic status. The amount of financial aid as designated by loans acted as an indicator for so cio economic successful retention at this institution. Further investigation into the academic and social processes need to be examined as they relate to retention and sub p opulations within the landscape of large, public institutions.
14 CHAPTER 1 INTRODUCTION environment in which declining funds and increasing enrollments are driving new business operation college a priority, but also has made finishing college a top priority for administrators and faculty. In explained his plan The president will invest $55 million in a new First in the World competition, to support the public and private colleges and non profit organizations as they work to develop and test the next breakthrough strategy that will boost higher education attainm ent and student outcomes. (Obama, State of the Union A ddress January 24, 2012 ) Secondary education attainment, positive student outcomes, and subsequent degree completion are key components to that plan. In this dissertation, data will be analyzed that may suggest a new model for student retention. This analysis will also demonstrate how practitioners on campuses can utilize large scale surveys to better understand student success at their institutions. For years now, legislative pressures have driven research, innovation, and the development of various retention and student success initiatives. Researchers and administrators have attempted to identify the factors contributing to student success in college. While graduation rates have risen, they continue to be low. For those students who entered four degree, the National C enter for Education Statistics (n.d.) reported that only 37.9 % of % finished within
15 five years; and 58.3 % completed within six years. Further examination of the data by gender, reveals the startling result that only 32.9 % of the males graduated within four years compared to 42.1 % of the females. Tinto (2012) documented differences in completion rates of high income versus low income students with high income students three times as like ly to graduate compared to low income students. Strategic enrollment management by institutions has shifted to increase college access and enrollment of under represented groups ( Bean, 1990; Hossler, 1990; Hossler, Dundar, & Shapiro, 2013). College retenti on is now thought of as beginning with intentional recruitment (Hossler, Dundar, and Shapiro, 2013). Raising graduation rates has become the focus of academic institutions from u niversities and c olleges to s tate and n ational forums for the advancement of e ducation. Increased graduation rates are critical for the United States. During Obama election, he set as a key goal for the U.S. to lead the world in college graduation rates by 2020 This is a serious goal for any nation or state, for any higher ed ucation institution, and for families and any individual. While it is acknowledged that college graduates earn more than their counterparts, it is important to examine the financial cost of not graduating from college for stakeholders. Specifically, what i s the cost of attrition or dropping out of college? The cost of attrition can be measured at many levels: the economic consequences of low graduation rates to students and their families (from tuition and fees cost, time, and income loss), cost to taxpaye rs, cost to institutions via lost credits that do not lead to a degree, and cost to both state and federal governments. Schneider (2010) in
16 Four Year Colleges and Universities examined attrition of full time first year students and the cost that occurs at both the state and national level. Using data obtained from (IPEDS), Schneider was able to doc ument the large financial burden attrition is causing on the nation. It is acknowledged that some students who leave an institution after their first year will enroll in another institution and eventually graduate. The impact of transfers is deemed small; Schneider estimates nationally that less than 10 % of all students who drop out of college will eventually return to a higher education institution and graduate. Still, it is important to recognize the limitation of the figures reported as they are most lik ely overestimates due to their inability to account for transfers. Even with this acknowledged overestimate, the losses both at the state and federal level are staggering. Over the last five years, Schneider estimates state and federal taxpayers have lost over $9 billion on the education of first year students who will not persist to the next year. On average, states spend $1.3 billion per year just on students who drop out after their first year in college. The federal government spends $300 million per y ear on these students. These numbers do not take into account lost taxes for future increases in revenues from these students. This statement is based on data from the U.S. Census Bureau showing that college educated adults between the ages of 25 and 34 wo rking year round earn approximately 40 % more than the same age group with some college education and about 60 % more than the same age group with only a high school education. Lower salaries are translated into lower tax revenues on both the state and feder al level. When you tease out the contributions from state
17 versus the federal government, the facts speak for themselves. Attrition is costly to all government stakeholders. Additionally, the financial burden for attrition is shared at the levels of instit utions and families and by the individual student. Johnson (2012) compiled an extensive report depicting a clearer picture of attrition in The Institutional Costs of Student Attrition Using multiple data sources including the Beginning Postsecondary Stude nts Longitudinal Study (BPS: 04/09), Johnson noted that a staggering 35 % of new undergraduates left without completing a degree after six years. The BPS: 04/09 is a national data sample that follows a representative sample of undergraduates who entered hig her education institutions during the 2003 04 academic year. This study tracks first time to college students over a period of six academic years from 200 4 through 20 09. Using the BPS cohort, Johnson was able to estimate the financial burden born e by an institution for each student who dropped out. The mean cost for each student amounted to $18,125. This does not include recruitment costs for each student. The data were also segmented by institution type. The attrition contribution of total expend itures was estimated to be 33 % at public two year institutions, 13 % at public four year institutions, and 9% at private four year institutions. Schneider and Yin (2011) explored the costs of attrition on an individual basis. Both students and their famili es are negatively impacted by attrition. Not only will the students face the financial cost of paying back loans for a degree not completed, but according to the U.S. Census Bureau they will earn on the average 40 % less than if they had completed their deg ree. Additionally these annual differences will continue to
18 will surpass a high school graduate s earnings by a half million dollars. Johnson (2012) further examines the reasons for attrition. While low academic performance is often the first reason parents and other stakeholders look at to explain attrition, the data does not support this concept that dropouts are just not college material 40 % of the dropouts had GPAs in the A and B range. Students with a GPA below a C accounted for only 10 % of the total dropouts. Johnson (2012) examined the numbers closely and was able to break the students who dropped out into four distinct groups based on the time of attrition and t The groups were defined to correspond with two questions: first, did the student leave the institution early or late in their program, and second, was the student in good academic standing ? The largest group was found to be students who left early and were in good standing. This group comprised 48 % of the total student dropouts. These facts led administrators to examine more closely the reasons behind these dropouts and the characteristics of those students who were retained. A successful student is a student who graduates, but the more critical components of that statement involve specific inducements such as who, what, why, and where the elements are that are associated with student success. The financial cost for attrition is huge. This pushes governments and institutions towards a greater culture of accountability. A facet of this accountability is to understand more about those students who drop out and those students who succeed. e completion, and the financial costs to the nation, state, institutions, and individuals, it is to be expected that there would be a strong focus on retention, student success, and degree completion. What do we know
19 about student retention? There have bee n a number of notable studies investigating student success elements H owever two main reports have resonated and continue to ood Practice in Undergraduate Education have largely shaped practical applications and surveys towards meeting the legislative mandates for greater accountability in student success rate s. Both of these studies, along with their criticisms, will be briefly presented next. The red flag was raised in 1984, when the S tudy G roup on the C onditions of E xcellence in American Higher Education by the National Institute of Education (NIE Study Group, 1984) produced a report titled Involvement in Learning: Realizing the Potential of American Higher Education report documents what the researchers refer to as wa rning signs from institutions the tendency to be more concerned about the input variables or characteristics of their incoming students rather than the outcomes of their graduating students. In order to ensure students achieve successful outcomes they identified three critical areas for improving success for undergraduate education: student involvement, expectations of faculty, and assessment. Student involvement, careful selection of students, high levels of student engagement, and careful administrati ve policies and practices to ensure student involvement in learning were also depicted as being vital to student outcomes. While researchers and practitioners have elucidated many important components to student success, a clear avenue for student success has remained unclear and fragmented (Braxton et al., 201 3 ; Tinto, 2012). Thus, we are now seeing, after many years of research, pressure from
20 legislators to turn applications of this research into viable classroom and institution policies to ensure the suc cess of students in undergraduate education. While the Involvement in Learning report focused on institutional policies, the Chickering and Gamson (1987) research focused more on student engagement. According to the Chickering and Gamson ( 1987, 1999; p. 76 ) t he seven principles of good practice in undergraduate education include: Encourages student faculty contact, Encourages cooperation among students, Encourages active learning, G ives prompt feedback, Emphasizes time on task, Communicates high expectations, Respects diverse talents and ways of learning Other researchers have built upon these seven principles to develop surveys and assessments to measure student engagement. In return, the results are to be used to guide adm inistrators and faculty in enhancing student engagement and thus student success (Radcliff, J.L., et.al., 1996). The first tool developed, the College Student Experiences Questionnaire, was designed around several of the original seven principles to examin e student engagement and experiences specifically at higher education institutions. This survey, in turn, has spawned several other surveys including the National Survey of Student Engagement (NSSE), the Law School Survey of Student Engagement (LSSSE), th e Community College Survey of Student Engagement (CCSSE) and others. The NSSE brought researchers and administrators further insight into the effects and consequences of academic engagement (Kuh, 2003). The LSSSE provides information to law schools about w hat their students think, what works, and what does not work in regards to the student experience. The CCSSE offers Community
21 College administrators and researchers an insight into the student experience. Unlike the Involvement in Learning report, each of these surveys was designed from the student perspective. While these surveys served an honorable goal to understand student engagement and hopefully increase the likelihood of student success they were not without their critics. One criticism centers a Porter and Umbach (2006) documented the various response rates from different institutions for the NSSE survey. They found that students designated as high ability were more likely to respond than their counterparts low ability students. Additionally, women tended to respond at higher rates than men, and students of color tended to respond at a lower rate than Caucasian students. This information l ed to the next question : are our results influenced by the differing response rates? Researchers (Porter, 2011) have also called into question the validity of self reported data. Porter questions the ability of students to report on their own behavior and perceptions. The instruments such as the NSSE depend on self reported data. These criticisms need to be taken into account when discussing theories of student persistence based on self reported engagement and persistence. How effective students are at tr anslating their student engagement into student learning outcomes is important (Kuh and Hu, 2001). Other researchers have stressed the importance of the quality of these experiences and the level of engagement that is important in regards to student retent ion (Astin, 1993; Pascarella & Terenzini, 1991). Whether the focus is on administrative policies which lead to student success such as
22 those encouraged in the Involvement in Learning (1984) report or whether the focus is on student engagement the same e nd result is still desired. Student success paired with institutional accountability is now the mainstay, the new business model for education; however, there still remains considerable room for improvement. Statement of the Problem The report, Measuring Up, The National Report Card on Higher Education (2008), provides a picture of how the United States compares internationally and how each state compares nationally on important benchmarking data including: college enrollment, certificate and degree comple tion, and adult educational levels. The data are further broken down by racial and ethnic groups. The information presented is significant as it demonstrates persistent gaps in the education of the U.S. workforce. A salient take home message was that 34 % o f young adults (18 24 year olds) were enrolled in college in the US while the national certificate and degree completion rate is only 18 % The report also documents persistent disparities concerning enrollment and completion of college when comparing race /ethnicity and family income. The time to degree has also increased with many institutions reporting six year degree completion rates versus four year. Compared to completion rates internationally, the United States has fallen behind other countries. This has implications not only for educators but also for the workforce and economy as well. Accountability for higher education institutions has moved to the forefront while retention rates have become extremely important ( Perna and Kurban, 2013; Seidman, 201 2 ; Tinto, 2012 ). College rankings such as U.S. News & World Report rely on such data points as variables related to completion rates to construct national rankings. A recent report by the American College Testing Program (2012) explored retention rates of first year students at four year colleges over
23 the past 20 years. Both public and private institutions have dropped compared to the prior year. The average retention rate during 2012 for private four year institutions was 71.9 % ; for public four year ins titutions the retention rate to the second year was 72.2 % All of these facts contribute to the emphasis legislators, administrators, families, and students themselves put on higher graduation rates and, by default, why there is more emphasis on student r etention initiatives. Researchers have also focused their attention on building models and theories to better understand student retention. Practitioners have attempted to utilize these models to make strategy recommendations, remediation programs, and co untless process changes to improve ins titutional retention and ultimately graduation rates. I n his book Leaving College: Rethinking the Causes and Cures of Student Att rition Tinto (1993) institution. This integration includes interactions with the campus community peers and faculty. The more positive these interactions, the more likely the student will be to persist at the institution. Braxton, Sullivan, and Johnson (1997) termed the issue the Purpose of the Study The purpose of this study is tw o fold. First, it is to provide a deeper understanding of the relationship of student characteristics, academic engagement, social engagement, and perceptions of belonging in regards to retention within the context of a large, researc h one public universit y. Second to depict and contrast the portrait of retained students and non retained students. Tinto (2012) challenges
24 s and both the role and responsibility of each institution to understand their student patterns is a fundamental rationale and purpose of this study. Over the past two decades researchers and administrators have studied retention, student engagement, experiences, perceptions, and satisfaction at higher education institutions from either the student or institutional perspective. Researchers have postulat ed different models on the process of retention and the reverse attrition. Seidman (2005; 2012) states that student retention is one of the most extensively researched topics in higher education. Particular attention has been devoted to the development a nd testing of theoretical models of student retention. The wealth of literature concerning college student retention will be reviewed in Chapter 2 According to researchers, by understanding the theoretical models to help practitioners identify variables Astin 1993; Cabrera and Nora, 1996; Tinto 1975, 1993, 2012 success, and thus their retention rate can be improved. These scholars in particular noted how stu dent variables such as high school achievement, gender, socioeconomic status, social and academic involvement, commitment, and race/ethnicity can affect student persistence. More recently, student experiences were examined in a survey designed by the Uni versity of California called the Student Experience in the Research University (SERU). Unlike past assessments, this instrument gives both administrators and faculty a better understanding about use of the data for benchmarking and program
25 improvement of t he undergraduate experience The design of the survey was built around the intent to provide a deeper understanding of undergraduate engagement and experience within the research university environment. The survey includes questions in five main areas: ove rall satisfaction and time use, academic engagement, global knowledge, skills and awareness, civic engagement, and student life and development. Individual participating institutions could include wild card questions that relate to their specific instituti on. In 2008, the SERU survey was made available to other research institutions creating a consortium for benchmarking purposes. Individual institutions house their own survey responses that can be coupled with individual student data to provide researcher s and administrators a robust database. This database can be mined to understand the students and their experiences, used for modeling to better understand student transitions and growth, and to make policy recommendations. It is posited that further inv estigation of the data provided by the SERU survey can help administrators examine the relationship between student persistence and three main constructs: academic engagement, social engagement, and perception of belonging. The SERU survey responses will b e paired with student characteristics to allow the researcher to control for these factors. Finally the outcome variable, retention to the next year will be mapped to the survey responses. Research Questions The study will be guided by the following overa rching three research questions: 1. Is there a relationship between academic engagement and student retention? 2. Is there a relationship between social engagement and student retention? 3. Is there a relationship between perceptions of belonging and student retent ion? The theoretical framework guiding this study incorporates the belief that a student enters a higher education institution with precollege characteristics that may
26 s perception of belonging at the institution are the main elements contributing towards retention. This framework is based on the Integrated Model of College Student Persistence initially formulated by Milem and Berger (1997). This model will be discussed in detail in Chapter 2 Significance of Study that influence student persistence within the conceptual framework offers scholars a comprehensive conceptual map to identify forces 2009; p. 675). An examination of the literature has led researchers to assert that both academic and social engagement and perceptions of belonging are correlated with retention of students in higher education. Student invo lvement and student engagement have been directly linked to student learning and retention (Gellin, 2003; Pascarella & Terenzini, 2005; Carini, Kuh, and Klein, 2006; Hu, 2010). This study will examine the el of College Student Persistence at a large, research, public four year higher education institution. Their model incorporates both behavioral and perceptual constructs into the process of student persistence and is eractionalist Model of Student Departure and motivation as an explanation for persistence. In his theory, Astin defines involvement as he amount of physical and psychological energy that the student devotes to the engagement and involvement, but like Astin, Axelson, and Flick (2011) merge these two
27 concepts interested students appear in their learning and how connected they are to their classes, their institutions, and each other persistenc e to his concept of student involvement. His Theory of Student Involvement incorporates a gradient of involvement with the bottom of the gradient equivalent to the graduation. Tinto (1975, 1993) Interactionalist Model of Student Departure also encompasses the pivotal role student involvement has in a college. Tinto states, "There appears to be an important link between learning and persist ence that arises from the i n terplay of involvement and the qual i ty of student effo r t Involvemen t with one's peers and with the faculty, both inside and outside the classroom, is i t self positivel y related to the quality of stude n t effort a n d in turn to both l earning and persiste n ce" (Tinto, 1993, p. 71 blended the behavioral measurements of involvement that are utilized by Astin with Other studies have examined the utility of the Integrated Model of College Student Persistence (Berger and Milem, 1999). Continued research to describe the relationship between student retention, social engagement, academic engagement, and perception of belonging will build this knowled ge base further. Research gathered for the purposes of this study will be used to provide assistance to this specific large, public, research institution to further understand the factors involved with student
28 retention. The research results will also cont ribute to the knowledge base of student retention as a whole. Limitations of the Study The current study was limited to one large, public, r esearch u niversity in Florida. The study specifically encompassed a population from one institution. The survey mea sures self reported data and therefore it is an indirect measure of engagement. Researchers have posed concerns about the validity of college surveys (Porter, 2011). The basis of the data in the SERU survey and other large scale student surveys is from sel f reported information. The concern that students cannot accurately report on their behaviors and attitudes is a valid limitation. While this is an existing limitation for such a survey, the SERU survey includes multiple questions concerning each factor to increase the validity. The outcome variable examines only retention from the period of the s pring 2011 semester to the fall 201 2 semester.
29 CHAPTER 2 LITERATURE REVIEW The objective of this literature review is to place the current study in the proper context of seminal research in the area of student retention modeling. Discussed first is (1975, 1993) Interactionalist Theory of Student Departure. The specific context will be research pertinent to student engagement, persistence and retention at higher education institutions. Research on student characteristics as they relate to involvem ent and retention will also be reviewed. Instruments used to assess student involvement will be discussed. A conceptual model will be presented. A summary completes this chapter. Historical Overview of Attrition Research Early research in student attrition was based on theories of suicide, sociology, student development, and economics (Astin, 1975; Bean, 1983; Cabrera, Nora, and Model of Student Departure is regarded as one of the prominent models in student retention. It is founded on the concept that attrition is a longitudinal process influenced examined the problem of student attriti on in a fixed longitudinal, multi institutional study from 1968 to 1972. Students selected for the study were from a national sample of two year and four year higher education institutions. The institutions included in the sample were participants in the Cooperative Institutional Research Program. The final sample consisted of approximately 300 students randomly selected from each institution for
30 follow up in 1972. This follow up sample included black students and students from institutions with enrollment s lower than 300. Included in the analysis of the data from freshman and the follow up data were institutional provided data concerning student characteristics. The analysis of the extensive data enabled Astin to estimate quantitative measures that predict The data were also used to quantify environmental experiences at the institution chances of dropping out. The concept of a good fit between the institution and the Involvement. Astin (1984) argued that findings from the longitudinal study indicated the presence of specific factors that impact student retention at institutions of higher education and that these factors depend on student involvement at those institutions. Astin asserted that active student involvement was positively linked to student development and ultimately student retention. Student involvement includes both a quality and a quantity component. Astin argues that involvement requires investments of time, energy and commitment. The theory focuses on two c omponents : the motivation and the behavior of the student. Institutional practices, policies and programs designed important role in the overall development and success of s tudents at those institutions. Astin (1996) contends social involvement and academic integration with faculty and student peers are critical components of positive involvement. Conversely, the negative or the lack of involvement has a significant negative impact on student outcomes.
31 Models of Student Departure student retention. He hypothesized that students enter a university with a commitment to stay in college and complete his/h er degree. The integration into this system at or depart from college. His model (1975) dictates that to successfully persist in college a student must successfully separate from family and childhood friends. Tint o later revised his model (1993) to include outside nity processes and motivators that cause a student to drop out prior to graduation. The student enters an institution with academic, social, and family characteristics that include their commitment and intent towards degree aspiration. This commitment is reframed as the student interacts with the institution. The model ties together the notion that persistence academic and social makeup. The higher the level of student integration both socially and academically, the higher the level of commitment and subsequent probability that students will persist. Within the concept of involvement, Tinto (1993) disc usses the need to understand how student involvement relates and influences the factor of student persistence. Tinto (1993) stressed the importance of student interaction between peers and faculty. Tinto implicates a socialization process that occurs betwe en a student and
32 Pascarella and Terenzini (1980) designed a longitudinal study of undergraduate students from Syracuse University to test the predictive validity o the context of social and academic integration. The academic integration component included student academic performance and development. Social integration focused on student interactions with both peers and faculty. Pascarella and Terenzini subsequently created a third component that described student commitment to the institution and goals. To examine the academic, social, and commitment components contributions, Pascarella and Terenzine created a n instrument with a five item respo nse Likert scale Additional pre college characteristics were included in the study. The the importance of social and academic integration on predicting student dropouts with unique influences of student characteristics. The analysis elucidated significant interactions between gender and peer interactions as well as institutional and goal commitments. The importance of student faculty interactions was also evident in the analy sis. Other researchers have stressed the importance of the interaction between faculty s tudent interactions was important in predicting persistence as well as the abse nce of campus involvement. Subsequent work by Bean (1975) led him to create a model to explain the college student dropout syndrome. In this model, Bean identified three factors that impact the socialization and selection process: 1) academic 2) social/psy chological and 3) environmental. The socialization and selection process
33 includes college grades, institutional/student fit and the institution. Cabrera, Nora, and Castaneda (1993) created a baseline model that incorporated bot included both academic and social integration. Cabrera, Nora, and Castaneda (1993) were able to provide a more robust understanding of the interplay between the student, environment and the institution. By combining the models, they were able to include of College Student Persistence Milem and Berger (1997) employed longitudinal data from first year students to Student Departure. Their study examined direct effects of va riables upon the persistence process. Previous research (Pascarelli & Terenzini, 1980) did not address of student departure and Tinto emphasized that for a student to successfully navigate and be retained at an institution, th e student needs to become integrated both socially and academically at that institution. He further stressed that students move through stages; first a separation from old friends, communities, ties, and beliefs; and second they develop new beliefs, norms, brings forth success. Between these two stages are what Tinto describes as the transition period beginning with a separation from old norms and ending with
34 integration i nto new norms and behaviors. The successful navigation of this transition period includes student academic and social involvement into the community of the institution. Astin (1994) focused on involvement as the linchpin for student success at an instituti on. Astin asserts that students invest varying amounts of emotional and physical energy into the institution through involvement. Milem and Berger (1997) asserted that that a student enters an institution with unique background characteristics that influence hypothesized that this initial involvement leads to perceptions of support which in turn influences subsequent involvement and ultimately retention. They proposed a model that includes a behavior perception as prescribed by Astin. The research sought to further describe student transition and integration at an institution and provides additional insights into the students for m their perceptions through their involvement. Milem and Berger took place at a highly selective private residential university with a total enrollment of under 10,000 students. The undergraduate enrollment was approximately 6,000 with 1,500 being full time freshmen. They constructed a study to examine retention of first year students. Data were gathered at three points of time: freshman orientation with a Student Information Form, mid fall semester through an Early Collegiate Experiences Survey, and finally in mid Spring
35 semester through a Freshman Year Survey. Data from all three instruments were matched for each student. Due to data loss at each point, the final data set contained information for 718 individuals. This equaled 46.4% of the enteri ng freshmen class and represented the population of the freshmen at the institution. Further, the researchers theory. The mapped independent variables includ ed student b ackground characteristics, initial level of commitment, mid fall involvement measures, mid fall perceptual measures, mid spring involvement measures, and mid spring perceptual measures. The dependent variable for this research study was a measure of studen t retention. This study did not utilize a direct measure of retention to the next year ; rather it outlined an indirect measure of the likelihood of the student continuing at that institution in the following year. While there is research that supports usin g an indirect measure of persistence (Pascaralla, Duby, and Iverson, 1983) it is important to note this limitation. Results of their analysis supported the idea that early involvement influenc ed student perceptions, and the level and nature of their cont inued involvement. Milem and Berger ( 1997 ) further investigated involvement with organized activities, with peers, and with faculty. The types of involvement were also important predictors of subsequent involvement and perception. They found involvement i n organized activities alone could be used as a predictor for subsequent spring involvement. Their findings indicated a positive role for early faculty involvement and subsequent persistence. Milem and Berger investigated the role academic and social integ ration played in predicting institutional commitment and found surprising results. While social integration served as
36 a positive predictor for institutional commitment, academic integration did not. Rather academic integration not only failed to positivel y predict institutional commitment but also failed to positively predict intention to re enroll. Results of this study shed light on the importance of early involvement the role involvement has in student perceptions, and predicting retention. While this study by Milem and Berger is critical to further understand the student mentioned, the study utilized a non direct measure of student persistence. Another notable limitation to their study was the narrow range of students at the tested institution. The institution in question was a highly selective private four year college. Milem and Berger noted that nearly every student at the institution entered with a high level of commi variability in goal commitment in their sample of students; consequently, goal commitment was not included in their subsequent model. Berger and Milem (1999) further researched the impa ct of involvement on student integration. They differentiated between behavioral and perceptual measures. Their study examined first year retention with a comprehensive set of variables including student characteristics, initial commitment, fall involveme nt measures, student perceptual measures and spring involvement measures. Their modified persistence model Departure was tested through path analysis. The final analysis presente d patterns of direct, indirect and total effects on academic integration, social integration, commitment, and persistence by a variety of student characteristics. The researchers found unique
37 patterns by different student groups that should be explored in future research. They also concluded that early involvement plays an important role in institution commitment and subsequent persistence to the second year. Both peer and faculty involvements were significant predictors of persistence. An important conclu sion that supported ed positive institutional commitment. Impact of Involvement Work done by Terezini, Pascarella, and Blimling (1996) examined the impact of a variety of col lege experiences on college student development. These experiences included Greek affiliation, extracurricular clubs and activities, intercollegiate athletics and sports clubs. A review of research by Moore, Lovell, McGann, and Wyrick (1998) stressed the i mportance of student involvement. In a longitudinal study done by Foubert and Urbanski (2006), students at a midsized public southeastern university completed the Student Development Task and Lifestyle Inventory. Analysis of the data indicated that student s who were highly involved on campus reported greater development in their psycho social development. Hu (2010) examined the relationship between different types of student engagement and persistence in college and surveyed students on both their levels o f social engagement as well as academic engagement. Results from this study indicated a non linear relationship between the participation in academic activities and the probability of retention. Hu also converted variables for social engagement and academi c engagement from continuous variables to categorical variables. Results indicated that this conversion improved persistence modeling. Further, findings indicated that students with a high level of academic engagement have a lower probability of being retained than students with a mid level of academic engagement.
38 Furthermore, if the academic engagement was not paired with a high level of social engagement the probability of being retained w was that the persistence variable was a self reported measure and not actual retention. This study has strong implications on how academic and social engagement are regarded by researchers, students, parents, and a dministrators. Other researchers have connected both academic and social engagement to persistence. Using data from the National Survey of Student Engagement (NSSE) gathered from 18 different universities, Kuh et al. (2008) found that the probability of s tudent persistence increased as students engaged in educationally purposeful activities In their study, researchers also found that this type of engagement had a stronger effect on minority students as well as on students with lower academic levels. Thi s study focused entirely on first year students and persistence between the first and second year. Active student engagement played an important role in successful student persistence. Hu and McCormick (2012) also used data gathered from the NSSE to devel op a typology of student engagement. They found distinct groups ranging from those highly engaged to those disengaged. These distinct levels of engagement also demonstrated differences in academic performance and persistence levels. McCormick et al. (2012) suggested this typology could climates. Differences in types, levels, and impact of engagement on differing student populations have been documented (Kuh et al., 2008; Harper and Quaye, 2009). Additional research is needed to explore these differences further.
39 Involvement Assessments A variety of instruments have been developed to examine student engagement and the student experience. These instruments were developed to include questions regarding college ranking, accreditation, student experiences and student satisfaction garnered greater interest. The Student Experience in the Research University ( SERU ) The Student Experience in the Research University (SERU) was designed by the Center for Studies in Higher Education at the University of California for undergraduate students on their campuses. The original survey was developed in 1999 specifically to expl ore the academic and civic engagement of undergraduate students in the University of California system. In 2008, the SERU Project was expanded to include other campuses outside the University of California. The mission of the SERU project for participating institutions was three fold to obtain a greater knowledge of who the students are, to university, and to utilize the data analysis to make program and policy recommendations. The survey utilizes a modular design with each module encompassing a number of items around a specific topic. The survey includes a set of core items that each participant respond ed to that concern ed use, and satisfaction. The modules evalu ate d student experience as related to five themes: academic engagement, global skills and awareness, community engagement, civic engagement, and student development. Participating institutions were allowed to create a set of unique questions as a wild card Student institutional data was matched to their survey responses by the institutional research office at participating campuses.
40 The SERU project formed a consortium of participating institutions. This allow ed individual institutions to benchmark themsel ves against the overall average of other institutions. It did not allow for individual institutions to compare themselves against specific institutions. This study will utilize data that was gathered in the SERU survey from the spring semesters of 2009 an d 2011. The survey was given to the undergraduate population at the University of Florida. While participation in the survey was voluntary, participation was encouraged by including incentive prizes and eligibility for a football ticket lottery. It is impo rtant to note that undergraduate students were not eligible for the football ticket lottery unless they completed 75% of the survey. The National Survey of Student Engagement ( NSSE ) student engagement and the quality of education at four year higher education institutions and two year higher education institutions. The National Survey of Student Engagement (NSSE) was developed to gather data from students relating to engagement, learn ing and experiences at four year institutions. The Community College Survey of Student Engagement collects the same type of data from students who are attending two year S even Principles for Good Practice in Undergraduate Education (Chickering and Gamson, 1987) The aggregated data from these surveys are intended to provide institutions with a portrait of engagement at their institution as well as the capability to benchmar k their institution against other institutions.
41 Other I nstruments The College Student Experiences Questionnaire (CSEQ) is another assessment tool that has been used by institutions and researchers to explore the quality of effort students apply in utilizi instrument are twofold. First it provides information for i nstitutions concerning their programs and policies. Second it provides an opportunity for the student to reflect and evaluate their p rogress during the academic year. The Cooperative Institutional Research Program (CIRP) Freshman Survey is a tool specifically designed to provide information and a portrait of incoming students. ckground and behaviors, expectations about college, values, financial concerns, and peer and faculty interactions. This survey can be followed up with Your First College Year (YFCY and College Senior Survey (CSS) which are also produced by CIRP. Conceptual Model The conceptual model for this study is based upon the Integrated Model of College Student Persistence initially formulated by Milem and Berger (1997). Further research (Berger and Milem, 1999) examined specifically the constructs of involvement in A Berger and Milem (1999) further refined their original conceptual model. This model is depicted in Figure 2 1. The conceptual model for this study will utilize student bac kground characteristics, social integration behaviors, academic integration behaviors, and institution perceptions as key variables to predict retention. Data to be used in this study are housed by the Institutional Research Office for stude nt background characteristics, housing variable s retention variable s and relevant
42 program participation data. The academic integration, social integration, and perceptions data will be obtained from the SERU survey. Hence, the constructs from Berger and Integrated Model of College Student Persistence were selected for this study whereas, variables concerning initial fall commitment and spring commitment were not included. Student Characteristics Minority Students The ethnic make up of the United States has changed. According to United States Census Bureau estimates, the ethnic make up of the nation will shift by 2050 They predict that the White non Hispanic population will fall below 53 % compared to the 2011 report of 63.4 % Along with this decr ease in White non Hispanic population is a predicted increase in minority groups. Two sub groups, African American and Hispanic students have lower levels of participation in h igher education compared to White non Hispanic students. Researchers have expl ored the involvement of specific student sub groups on their campuses (Flowers 2004; Harper, Carini, Bridges, & Hayek 2004; Jones, 2013; Taylor and Howard Hamilton 1995). The impact of student involvement on educational outcomes for unique student groups Involvement on a specific sample: African American college students. In this study, Flowers used data collected from the College Student Experience Questionnaire (CSEQ). This survey uses student self reported responses on demographics, involvement experiences and gains in intellect and social areas. The findings demonstrate that student involvement for this student sub group positively impacts student gains in understanding art s and humanities, personal and social development,
43 understanding science, critical thinking and writing skills, and vocational preparation. Flowers also reported that the quality of the involvement also impacted the reported student gains. Historically Afr ican American women who attended historically black colleges and universities HBCUs reported lower engagement than their male counterparts (Allen 1986). In a study conducted by Harper, Carini, Bridges, & Hayek (2004) that collected NSSE data from undergrad uates who attended twelve HBCU s researchers found that the engagement gap between genders is now closing. Generational Status First generation college students (FGS) are another sub group that has been explored recently First generation students are def ined as students who are the first generation in their family to attend a higher educational institute. Evidence has indicated that this group compared to its peers are at a disadvantage in many ways including lack of basic knowledge about higher education, level of family income and support, academic background, and higher education degree aspirations. Research has indicated differences when comparing first generation students and their counterparts (Ishitani 2003; Pa scarella, Pierson, Wolniak, & Terenzini 2004 Perna and Kurban, 2013 ). Ishitani explored the enrollment of first generation students compared to non first generation students and found that first generation students are 71 % more likely to drop out compared to non first generation students. Research done by Pascarella, Pierson, Wolniak, & Terenzini (2004) also demonstrated a higher attrition rate for first generation students. In their study researchers compared college students who participated in the Nati onal Study of Student Learning (NSSL). The NSSL is a longitudinal study that examined college student experiences and outcomes. The study compared first generational students to two groups. The first group included those
44 students where both parents complet findings support the idea that first generational students lag behind their peers in terms of the types of institutions they attend and the types of experiences they have during college. Specifically first generation students are more likely to attend institutions that are less academically selective. These students also enrolled in fewer credits, work ed more hours per week, live d off campus, and had lower levels of extracurricular involvement. First Year Students Institutions have created a variety of programs and policies to remediate differences between sub groups. First year students have been determined to be another unique sub g roup. This group faces challenges that include the transition from home to college as well as testing of their beliefs, values, and identities. Data collected across a variety of institution al types including both 2 year and 4 year institutions as well a s public and private institutions document that nearly one third of all undergraduate students leave their institution by end of their first year (Horn and Carroll, 1998). This high attrition rate at the end of the first year was also documented by Barefo ot (2000) who saw that the majority of dropouts happen at the juncture between freshman and sophomore year. Consequently, time, focus and resources have been put year. As institutions struggled to decrease the attrition rate during the freshman sophomore juncture, administrators and practitioners created first year programs. These programs were embedded with the findings of Tinto, Bean, Astin and other scholars who soug ht to integrate first year students socially and academically into the institution.
45 These programs were designed to decrease the attrition rate and increase the academic performance of the students. Jamelske (2009) reported that 95% of U.S. 4 year institut ions have incorporated some type of first year program. The kind of program or programs at each institution is often reflective of the population it serves and the mission of the institution itself. Social and Academic Integration In his Model of Student D eparture, Tinto (1993) hypothesized that the degree to which a student integrates into both the academic and social sector of the university community affects the probability that the student will persist at the university. Braxton, Hirschy, and McClendon (2004) examined these constructs. Their evaluation of a number of studies found a strong link between the construct of social integration and retention. Getzlaf et al. (1984) explored the levels of academic and social integration of students who dropped ou t compared to those who were retained at Washington State University. Their findings indicated that those who dropped out had lower levels of both academic and social integration. Jones (2010) examined the question of whether gender plays a role in the i mpact of social integration on institutional commitment. The researcher found that gender did indeed play an important role in social integration. Notably, Jones found that social integration had a more positive impact on institutional commitment for fema les compared to males. Pascarella and Terenzina (1983) differentiated the roles of academic and social integration. Their research indicated that academic integration was related to both institutional commitment and commitment to complete their degree, wh ile social integration was solely related to institutional commitment. Research by Beil et al.
46 importance of academic and social integration in the retention process. Many resear chers have examined and supported the impact both academic and social integration have on student retention. Perceptions of Belonging Perception of belonging is also described by researchers as institutional fit This concept of fitting in is directly connection between social integration and institutional commitment and thus retention has been explored by many researchers. Fitting in can be tied to social integration. Bean (2005) describes fitting in by saying, Social connectedness leads to satisfaction, self The issue of fitting has been emphasized by researchers who have explored retention of minority students. Hurtado and Carter (1 and persistence for Latino college students Many researchers have combined the sense of belonging construct with other constructs including institutional fit and institutional commitment (Cabrera et al,. 1992, 1993; Nora and Cabrera, 1996; Strauss and Volkwein, 2004). Hausmann et al. (2007) examined the role of perceptions about belonging on student retention rates While controllin g for social and academic integration, and student commitment to the institution, researchers examined the effect contributed to a sense of belonging. The construct of academic integration appeared to be associated with perceptions of belonging. Hausmann et al ( 2007) demonstrated that levels of above average academic integration were tied to an increase in perception s of belonging while a below average level was associated with a decrease in perceptions of belonging. Significantly,
47 intention to persist at the beginning of the academi c year. In a study by Johnson et al. (2007), the researchers explored the sense of belonging among 2,967 first year students from a national sample. Their findings indicated that students identifying as African American, Hispanic/Latino, and Asian Pacifi c American students indicated a lower sense of belonging compared to their White/Caucasian counterparts. This research is based on the work by Hurtado and assertion that to suc cessfully integrate into the university community, a student must transition away from their family and high school experiences to affiliations within the university community. Joh nson et al. (2007) results found that the sense of belonging is affected by both the racial/ethnic background of the student and the college climate. Their study demonstrated the importance and strong contribution that living in a residence hall has on a student s sense of belonging. A substantial body of research has explored h Chapter Summary This literature review provided an insight and understanding of different frameworks for understanding the world of college student persistence. Various conceptual models look ed at the process of student departure and contribute d to an explanation about why some students persist with their college education and others do theory of student departure were reviewed and discussed within the context of the model of student persistence postulated by Milem and Berger (1997). The Model of
48 Student Persistence (Berger and Milem, 1999) was then dis cussed and used to build a conceptual model to investigate the retention rates of college students with emphasis on academic integration, social integration, and perception of belonging.
49 Fig ure 2 1. Integrated Model of College Student Persistence This illustration is an adaptation of a figure appearing in The Role of Student Involvement and Perceptions of Integration in a Causal Model of Student Persistence. Joseph B. Berger and Jeffrey F. Milem, Research in Higher Education, 40 (6), 1999, p. 645 Figure 2 2. Conceptual Model: Model of Student Retention
50 C HAPTER 3 METHODS Research Methodology In this study, research efforts will be conducted in two phases. The first phase is descriptive in that variables of interest will be described but not causal relationships. For example, the SERU survey contains data on undergraduate students at research universities. Descriptive statistics will be used to characterize the frequencies and patterns of the o bservations or the distributions of target variables, data, or observations. The second phase of this research entails the use of quantitative methods to explore questions or relationships pertinent to student retention some of which may be causal or predi ctive. More specifically, logistic regression will be used to explore the influence of various independent variables on the dependent variable retention. Population and Sample Data from the 2011 SERU survey will be used. Survey questions co ncerning academic integration, social integration, and perceptions regarding belonging will be explored. Student responses in these areas will be matched with student demographic and college data (gender, ethnicity, generational status, housing, Greek affi liation, transfer status, high school GPA, SAT/ACT test scores, University GPA, and college major). Survey questions concerning academic integration, social integration, and perceptions regarding belonging will be grouped for each category to form three co mposite variables. Thus there will be an academic integration variable academic integration. The same will be done for social integration, and perceptions. Table 1 contains the question make up for each composite variable. This data will then
51 be analyzed to create a frequency pattern for academic integration, social integration, and perceptions for all students as well as specific groups of students based on demographic and coll ege data. Measures Dependent Variables The dependent variable (first year retention) will be operationalized with a dichotomous variable for enrollment in fall semester of 2012. For this variable, each student will be identified as either enrolled in fall semester of 2012, or withdrawn from the institution. The enrollment status of each student will be determined by the u niversity Office of Institutional Planning and Research analysis of the student data file for fall 2012. Independent Variables The just ification for including academic integration, social integration, perceptions of belonging, and background/college characteristics is presented in detail in C hapter 2. A brief review is presented here. Academic Integration The academic integration variable will be created by forming a composite variable that includes involvement in academic activities. All items relating to academic integration will be analyzed via factor analysis to determine underlying structure of this group of items. The academic integr ation items summated to create the composite or composites will be determined by this analysis. The rationale for including academic integration in the model is based on research by Tinto (1987, 1993). Incoming academic ability is often measured by weighte d high school GPA and SAT scores ; both have been shown to have a positive relationship with persistence (Ishitani & Snider, 2006; Tinto,
52 1993). Horn and Carroll (1998) also found that , the GPA was positively tied to pe rsistence. Academic integration is often measured by indicators of academic performance such as the GPA, and involvement in academic activities. The academic activities often include faculty student interactions and relationships as well as academic in teractions between peers. These activities can be both formal and informal in nature. Pascarella & Terenzini (1976) found a positive relationship between the amount of informal experiences between first year students and faculty and the degree of the stude demonstrated the positive effects that peer and faculty involvement can have in decreasing the attrition rate of African American students. Astin (1996) specifies that increased involvement with faculty and peers forms the strongest positive involvement. Social I ntegration I will operationalize the social integration variable by creating a composite variable that includes involvement in extra curricular activities. Research in retention has contributed to the rationale for including social integration in this study. For inst ance, Cabrera, Nora, and Castaneda (1992) found social integration at the institution was a positive predictor for degree attainment. Astin (1984) postulated that a student needed to be actively engage d in their institution. Tinto (1975, 1993) contends tha t social integration is a major contributor to student retention. Research by Kuh et al (1991) demonstrated the role social involvement has in a in to the academic and social culture of the institution. Astin (1996) confirmed that noninvolvement on campus has a strong negative impact on student retention.
53 Perceptions The perceptions of belonging variable will be created by forming a composite variable that includes survey items relating to this topic. Factor analysis will dete rmine underlying the structure of this group of items. The rationale to include perceptions of belonging items is grounded in previous retention research ( Hurtado & Carter, 1997; Hurtado & Ponjuan, 2005; Hoffman, Richmond, Morrow, & Salomone, 2002; Lee & D avis, 2000 ; Rosenberg and McCull o ugh, 1981; Tinto, 1993 ). The perceptions of their campus culture and their sense of belonging on that campus play a key role in how they become engaged, the quantity of that engagement, and the quality of th at engagement (Torres & Hernandez, 2007). Many researchers believe that for institution (Hurtado & Carter, 1997; Hurtado & Ponjuan, 2005; Hoffman, Richmond, Morrow, & Salomone, 2002; Lee & Davis, 2000). Milem and Berger (1997) modified perceptions. Milem and Berger suggested that students form a perception of their fit based on their i nteractions between the social and academic structures of the institution. They further academic and social systems affect their subsequent decision to be retained at the institution. Berger a nd Milem (1999) further explored the relationship between involvement actions, perceptions of belonging, and retention. Their findings supported the inclusion of both behavioral constructs of involvement as well as perceptual constructs in predicting stude nt retention.
54 Background Characteristics extensive research on student retention in higher education. Tinto (1975, 1993) contends that students enter college with their unique ba ckground characteristics and to persist at the institution. Researchers have documented that gender, race/ethnicity, family education, high school GPA, test scores on standardized exams, and family income can have effects on predicting retention and ultimately on degree attainment (Astin, 1993a, 1993b, 1996; Pascarella and Terenzini, 2005; Tinto, 1993). Students family income and education have also been examined by researchers and have been shown to have a positive effect on retention ( Horn and Carroll, 1998). Analysis Methods Logistic Regression Logistic Regression is used in studies to model a relationship between a categorical outcome variable a nd a set of independent variables. Dey and Astin (1993) explored different statistical modeling techniques to predict student retention. Researchers compared results from logistic regression, probit analysis, and linear regression. All three types of analy ses showed little practical differences. The outcome variable for this study, that is retention, is a categorical variable T hus logistic regression has been chosen to model the relationship between student retention and predictor variables. Logistic regr ession assumes the following: the observed data is from a random sample of the population, and multi collinearity is absent; meaning the predictor variables are not highly correlated with each other. To diagnose the presence of multi collinearity the varia nce inflation factor (VIF) will be calculated for each independent
55 variable. Stepwise logistic regression will be used to create the most parsimonious model that predicts student retention. The fit of the model will be tested after each variable is introdu ced. In an effort to determine if the calculated model provides the best fit for the population, a Likelihood Ratio test will be performed. Logit(P) = a + bX (3 1) w here : X = a vector of independent variable; b = a vector of parameters a = the intercept P = retention Reliability The constructs included in the proposed model are abstract concepts in nature and thus rely on students to self report. The measures are indirect in nature. This introduces unreliability into the results in the instrument and sub sequent measurement errors. For this study, the SERU survey is assumed to have a high degree of reliability and due to the non sensitive nature of the questions participants are reporting honestly and accurately (Porter, 2011). To improve reliability, com posite scores will be used for the specific constructs as opposed to relying on individual questions within the SERU (Chapman, 2011). Reliabilities of the composite will be calculated and reported. Validity Samuel Messick of the Educational Testing Servi ce (1989) defined validity integrated evaluative judgment of the degree to which empirical evidence and theoretical rationales support the adequacy and appropriateness of inferences and actions 13). Messick goes on
56 interpretability, relevance, and utility of scores, the import or value implications of scores as a basis for action, and the functional worth of scores in terms of social consequences 2013 ) emphasizes his concerns with both reliability and validity of large scale student surveys. McCormick and McClenney (2012) note that student engagement surveys such as NSSE, CCSSE, and possibly the SERU are not tests nor is the gathered information of high stakes in nature. McCormick and McClenney state d rather that these types of surveys are designed for consequential validity and supply information for practitioners. (p. 330) Parti cipants and Sampling Procedure The institution chosen for this study is a public, four year or above primarily residential university that maintains high research activity. The undergraduate instructional program is classified as balanced arts & sciences/ professions, with high graduate coexistence. The graduate instructional program is a comprehensive doctoral with medical/veterinary. The enrollment profile is majority undergraduate. The undergraduate profile is full time, four year, more selective and low er transfer in. The profile is based on the 2010 Carnegie Classification system. The student population is 50,691. The population in this study included nearly all of the undergraduate students enrolled at the institution during the spring semester of 2011. While all the undergraduate students were encouraged to participate in this study, approximately 63 % of the undergraduate population or 19,500 students completed the survey. Enrollment data at the u niversity is reported during the fall of each s emester. Fall 2010 undergraduate demographic statistics reflect 55% female and 45% male. The undergraduate population was 59.12% White, 16.37% Hispanic, 9.47% Black, 8.25%
57 Asian, 1.64% Non resident Aliens, 2.89% not reported, 1.71% two or more races, 0.39% American Indian or Alaska Native, and 0.15% Native Hawaiian or other Pacific Islander. retention and graduation rates. The retention rate for full time first time in college undergraduate students who began in fall 2010 and was retained the following fall 2011 was reported as 95.64 %. All new undergraduate students are required to attend the This inst itution also incorporates additional strategies to improve the first year experience. These include a first year reading program, and freshman co nvocation ceremony. Barefoot (2000) documents orientation programs, reading programs, co nvocation ceremonies, a nd first year seminar programs as important means to improve first year transition and ultimately improve retention. At this institution the first year seminar course is optional for most incoming undergraduate students. The course is mandatory for specifi c scholarship students. The four year graduation rate for full time first time in college undergraduate students who began in 2008 was reported as 67.05%. The six year graduation rate for full time first time in college undergraduate students who began in 2006 was reported as 84.9%. Study Limitations It is important to note the limitations of this study. The first limitation is the overall the culture and climate o f that particular institution. A second limitation was discussed in previous chapters and reflects the limitations of using self reported data.
58 It also should be noted that results from this study should not be generalized for other institutions and popul ations. The results should be used to inform practitioners at the institution where the study was performed.
59 CHAPTER 4 RESULTS The purpose of this study was to explore the relationship between academic engagement, social engagement, perceptions of belon ging, and retention. The student characteristics explored in this study included the following: race/ethnicity, gender, financial loans, and living on campus. This chapter presents the results of descriptive statistics, factor analysis, and logistic regres sion using the SPSS 22 statistical package The descriptive statistics from the data set include frequency distributions for all students, as well as for retained versus non retained students. Descriptive Statistics The analysis used in this research incl uded only freshmen (those students with less than thirty credit hours) and sophomores (students with more than thirty credit hours, but less than sixty credit hours). The data set included a total of 1146 students with 25.9% designated as freshmen and 74.1 % designated as sophomores. These students were randomly selected to receive the wildcard module concerning extra curricular activities. Gender, race/ethnicity, housing status, and retention during Fall f Institutional Planning and Research. For analysis purposes American Indian/ Alaska Native, Nonresident Alien, and not reported groups were excluded as each represented very small percentages of the group (0.4%, 0.4 % and 2.5% respectively). When examini ng the race/ethnicity of the retained students compared to the non retained students, an increase in black non retained students was observed (Table 4 1). Table 4 2 reflects the percentages of male and female students based on their retention status. The n on retained group of students was comprised of 67.2% females and 32.8% males. Table 4 3 reflects the housing
60 status of the respondents based on their retention status. The results showed that 47.5% of non retained students lived on campus compared to 61.9% of retained students. Table 4 4 reveals the differences between retained students versus non retained students based on parental education. Table 4 1. Percentage of students by race/ethnicity and retention status All Students Retained Students Non retained Students Race Percent Percent Percent White 64.14 65.08 47.46 Black/African American 8.04 7.25 22.03 Hispanic/Latino 17.98 17.75 22.03 Asian 9.85 9.92 8.47 Table 4 2. Percentage of students by gender and by retention status All Students Retained Students Non retained Students Gender Percent Percent Percent Male 40.2 40.2 32.8 Female 60.2 59.8 67.2 Table 4 3. Percentage of students by living on campus and retention status All Students Retained Students Non retained Students Living On Campus Percent Percent Percent Yes 62.2 61.9 47.5 No 38.8 38.1 52.5 Table 4 4. Parental education and retention status All Students Retained Students Non retained Students Parental Education Percent Percent Percent High School or below 20.2 19.6 30.0 Associates degree or above 79.8 80.4 70.0 Continuous independent variables were examined. Table 4 5 describes these variables. One particular variable involved the amount of accrued student loans. Notably
61 the range of the variable for the total loans for each individual student was $0 to $39,214. As this range is orders higher than the other continuous variables, a scaled version (scaled by one hundred) of this variable is created for use during the regress ion stage of analysis. The variables for extra curricular activities and total loans both exhibited high levels of skewness with levels of 2.630 and 3.831 respectively. This indicates that both variables are distributed to the left at the lower values. The normality of all the continuous variables was explored using the Kolmogorov Smirnov statistic. Table 4 6 explores these results. As the results were significant, p <.001, this suggests a violation of the assumptions of normality for these variables. Table 4 5. Descriptive statistics of the continuous independent variables Variable N Minimum Maximum Mean Std. Deviation Academic Engagement 1 1110 6 30 18.64 4.893 Academic Engagement 2 1075 11 60 31.76 9.163 Academic Engagement 3 1130 4 24 18.50 3.152 Perceptions of Belonging 1 1129 2 12 9.54 2.021 Perceptions of Belonging 2 1121 7 42 33.20 6.322 Extra Curricular Activities 1003 0 15 2.30 2.201 Total Loans 1125 0 39214 1901.416 4688.159 Table 4 6. Tests of Normality of the continuous independent variables Variable Kolmogorov Smirnov df Sig. Academic Engagement 1 .069 879 .000 Academic Engagement 2 .091 879 .000 Academic Engagement 3 .123 879 .000 Perceptions of Belonging 1 .207 879 .000 Perceptions of Belonging 2 .123 879 .000 Extra Curricular Activities .184 879 .000 Total Loans .419 879 .000 Inferential Statistics to Explore Differences between Retained Students Versus Non Retained Students Inferential statistics were conducted to explore differences between retained students and non retained students. A Chi square test of independence was performed to examine the relationship between ethnicity and retention. The relationship between
62 these va riables was significant, X 2 (3, N = 1107) = 18.574, p <.001. African American students were less likely to be retained compared to other ethnic groups. The Cramer V value wa s .130 which indicates a medium effect from an association between ethnicity and re tention. A Chi square test of independence was performed to examine the relationship between living on campus and retention. The relationship between these variables was significant, X 2 ( 1 N = 11 46 ) = 5.038 p <.0 5 This indicates there is a statistically significant relationship between living on campus and retention. Students who lived on campus were more likely to be retained compared to students who lived off campus. The Cramer V value wa s 066 which indicates a small effect from an association between living on campus and retention. A third Chi square test of independence was conducted to explore the relationship between parental education and retention. The relationship between the variables was statistically significant, X 2 (1, N = 1130) = 3.796, p =. 05. The results indicated that students who had at least one V value was .058 which indicates a minor effect from an association between parental education and retentio n. The Chi square test of independence was conducted for gender and retention. The relationship was not statistically significant, X 2 (1, N = 1146) = 1.319, p >.1. A Mann Whitney U test was performed to examine differences between retained students and non retained students in terms of extra curricular activity. The extra curricular activities variable violated normality assumptions as a parametric measure. The test revealed significant differences in extra curricular activity levels of non retained students ( Md = 2, n = 52) and retained students ( Md = 2, n=951), U = 20459.500, z =
63 2.137, p = .03, r = .07. The direction of the difference was higher for retained students, indicating retained students were more involved in extra curricular activities. A Mann Whitney U test was also performed to examine differences between retained students and non retained students in terms of financial aid loans. The loans variable violated normality assumptions as a parametric measure. The test revealed significant differenc es in the amount of financial loans of non retained students ( Md = 0, n = 43) and retained students ( Md = 0, n=1082), U = 18334.500, z = 3.154, p = .002, r = .09. The direction of the difference was that the retained students had lower amounts of loans. Factor Analysis of the Make up of Survey Items for Academic Engagement and Perceptions of Belonging To establish construct validity, a factor analysis on key survey items was conducted. This was performed for items that shared the topic concerning academi c engagement as well as items concerning perceptions of belonging. The factor analysis was performed to extract principal components and then followed by varimax rotation. This analysis extracted three factors for academic engagement with initial eigenval ues greater than 1.00 1 2 3 = 1.775) The analysis extracted two factors for perceptions of belonging with initial eigenvalues greater than 1.00 1 = 2 = 2.192) The make up of each academic engagement factor is listed in Table 4 7 The make up of each perceptions of belonging factor is listed in Table 4 8
64 Table 4 7. Academic Engagement Factor Analysis Factors Mean Std Dev Factor 1 Factor 2 Factor 3 Academic Engagement 1 Raised your standard for acceptable effort due to the high standards of a faculty member 3.51 1.233 0.533 Extensively revised a paper at least once before submitting it to be graded 3.88 1.457 0.618 Sought academic help from instructor or tutor when needed 3.49 1.447 0.722 Worked on class projects or studied as a group with other classmates outside of class 3.96 1.440 0.757 Helped a classmate better understand the course material when studying together 3.82 1.360 0.744 Academic Engagement 2 Took a small research orientated seminar with faculty 1.73 1.257 0.444 Communicated with a faculty member by e mail or in person 4.31 1.307 0.700 Talked with the instructor outside of class about issues and concepts derived from a course 3.30 1.502 0.756 Interacted with faculty during lecture class sessions 3.46 1.456 0.811 Worked with a faculty member on an activity other than coursework (e.g. student organization, campus committee, cultural activity) 2.22 1.520 0.556 Contributed to a class discussion 3.89 1.361 0.743 Brought up ideas or concepts from different courses during class discussions 3.37 1.391 0.765 Asked an insightful question in class 3.40 1.354 0.776 Found a course so interesting that you did more work than was required 3.16 1.331 0.648 Chose challenging courses, when possible, even though you might lower your GPA by doing so 3.75 1.422 0.452 Made a class presentation 3.50 1.545 0.626 Had a class in which the professor knew or learned your name 1.05 1.496 0.702 Academic Engagement 3 Turned in a course assignment late 5.35 0.888 0.623 Went to class unprepared 4.38 1.112 0.863 Skipped class 4.48 1.104 0.733 Went to class without completing assigned reading 3.92 1.330 0.795
65 Table 4 8. Perceptions of Belonging Factor Analysis Factors Mean Std Dev Factor 1 Factor 2 Perceptions of belonging 1 I feel free to express my political beliefs on campus 4.72 1.085 0.907 I feel free to express my religious beliefs on campus 4.76 1.095 0.895 Perceptions of belonging 2 Students are respected here regardless of their economic or social class 4.56 1.131 0.771 Students are respected here regardless of their gender 4.93 0.95 0.791 Students are respected here regardless of their race or ethnicity 4.72 1.055 0.868 Students are respected here regardless of their religious beliefs 4.64 1.082 0.776 Students are respected here regardless of their political beliefs 4.65 1.071 0.719 Students are respected here regardless of their sexual orientation 4.61 1.079 0.836 Students are respected here regardless of their disabilities 4.81 0.993 0.799
66 The relationships between the three academic engagement composites the two perceptions of belonging composites loans, and extracurricular activities were investigated using Pearson correlation coefficient. This analysis was performed to examine collinearity. Results indicated a strong positive correlation between academic engagement composite 1 and academic engagement 2, r = .558, n = 1045, p < .001 with high levels of academic engagement composite 1 associated with high levels of academic engagement composite 2. Results indicated a strong positive correlation between perception of belonging composite 1 and perception of belonging composite 2 2, r = .579, n = 1116, p < .001 with high levels of perception of belonging composite 1 associated with high levels of perception of belonging composite 2. The results also indicated a weak correlation between the extracurricular activities and both ac ademic engagement composite 1 and 2, r = 174 n = 978 p < .001 and r = 231 n = 942 p < .001 respectively. The remaining results from the correlation matrix indicated weak correlations between all other variables. Logistic Regression Logistic regressio n was utilized to create a model to predict student retention. The first logistic regression model utilized all the proposed independent variables. This full model is presented in Table 4 9. A second logistic regression was performed to create a reduced mo del and is presented in Table 4 10. The full model that contains all predictor variables was statistically significant X 2 (12, N = 852) = 25.010, p < .015. The model a s a whole explained between 2.9 % (Cox and Snell R Square) and 10.2% (Nagelkerke R Squared) of the variation in being retained in Fall 2012, and correctly classified 96.1% of cases. The chi square value for Hosmer and Lemeshow Test is 14.919 with a significance level of .061. This value is
67 larger than .05 thus demonstrating support for the model. The 2 Log Likelihood is 260.739. Table 4 9 indicates that two independent variables made a statistically significant contribution to the model. These were ethnicity/race and extra curricular act ivities each with a p <.05. Loans and living on campus were variables that also notably contributed to the model with p <.10. The strongest predictor was engaging in extra curricular activities with an odds ratio of 1.283. This indicated that for every unit of extra curricular activity that respondents participated in they were 1.283 times more likely to be retained. Table 4 9. Summary of Full Model of Logistic Regression Analysis for Predicting Retention of Freshm e n and Sophomores ( N = 852) Independent Vari able B S.E. Wald df p Odds Ratio Gender .40 .39 1.05 1 .31 1.49 Ethnicity/Race 8.04 3 .04** Asian .302 .77 0.16 1 .69 .74 African American 1.57 .83 3.57 1 .06 .208 Hispanic .548 .83 0.43 1 .51 .578 Living off Campus .637 .37 3.03 1 .08* .529 Academic Engagement 1 .031 .04 0.48 1 .49 1.03 Academic Engagement 2 .02 .04 0.62 1 .43 .98 Academic Engagement 3 .06 .05 1.17 1 .28 1.06 Perceptions of Belonging 1 .003 .11 0.00 1 .98 .997 Perceptions of Belonging 2 .006 .03 .03 1 .87 .994 Extra curricular activities .249 .128 3.80 1 .05** 1.283 Loans Scaled .005 .000 3.50 1 .06* .995 Constant 2.681 1.647 2.649 1 .104 14.597 Table 4 10. Summary of Reduced Model of Logistic Regression Analysis for Predicting Retention of Freshm e n and Sophomores ( N = 955 ) Independent Variable B S.E. Wald df p Odds Ratio Ethnicity/Race 7.815 3 .05** Asian .113 .64 0.031 1 .86 .893 African American 1.264 .46 7.656 1 .01 .282 Hispanic .365 .44 0.031 1 .86 .694 Living off Campus .643 .347 3.435 1 .06* .526 Extra curricular activities .223 .114 3.80 1 .05** 1.249 Loans Scaled .005 .002 5.035 1 .03** .995 Constant 3.482 .365 90.803 1 .00 32.539 The reduced model that contains a subset of the predictor variables was statistically significant X 2 (12, N = 955 ) = 22.320 p =.001 ). The model as a whole
68 explained between 2. 3 % (Cox and Snell R Square) and 8.3 % (Nagelkerke R Squared) of the variation in being retained in Fall 2012, and correctly classified 96.1% of cases. The chi square value for the Hosmer and Lemeshow Test is 8.907 with a significance level of .350. This value is larger than .05, thus demonstrating support for the model. The 2 Log Likelihood is 290.786. Table 4 9 indicates that three independent variable s made a statistically significant contribution to the model: ethnicity/race loans and extra curricular activities with a p <.05. L iving on campus also notably contributed to the model with p <.10. The strongest predictor was extra curricular activities wit h an odds ratio of 1.2 49 This indicated that for every unit of extra curricular activity that respondents participated in they were more 1.2 49 times more likely to be retained. To determine if the full model containing gender, the academic engagement var iables, and the perception of belonging variables significantly improved the model, the difference between the 2Log Likelihood statistics for the full model were compared to the 2Log Likelihood statistics of the reduced model. The result indicate that by adding these variables the model is significantly improved, X 2 ( 6, N = 852) = 30.047, p < .001. It is important to note that models with additional variables (GPA and parental education) were explored. The full model presented in Table 4 9 was the most pa rsimonious model that accomplished a significant prediction level with as few predictors as possible given the dataset available. Chapter Summary The primary purpose of this study was to investigate the landscape of retention by examining student character istics, academic engagement, and perceptions of belonging. The findings of the descriptive statistics, inferential statistics, and logistic regression presented in Chapter 4 generated answers to the research questions in this
69 study. In summary, there are differences between the make up of the retained student group and the make up of the non retained student group. Two predictive models were presented. The full model was chosen as it best predicts the outcome variable given the data available. Chapter 5 wi ll discuss in greater detail the results from this study and relate these results to previous research in this area.
70 CHAPTER 5 SUMMARY, CONCLUSIONS AND RECOMMENDATIONS This study sought to describe retained freshmen and sophomore students and the ir un retained counterparts at a large, public, four year, primarily residential research university using institutional data and data obtained by the Student Experience in the Research University survey (SERU) from spring 2011. This study used logistic re gression to examine the predictive value of various independent variables for retention. For example, student characteristics such as gender, race/ethnicity, and parental education were previously shown to be important in predicting college student retenti on. This study specifically explored three academic engagement composite variables, two composite variables relating to perceptions of belonging, level of extra curricular activities, and student characteristics including gender, and race/ethnicity, housin g status, parental education, and amount of financial loans. The study sought to validate the retention research of Astin (1975, 1999), Berger and Milem (1999), and Tinto (1993, 2012). This chapter includes a summary of the outcomes, a discussion of the findings, and implications for researchers and practitioners. Summary of the Findings This study employed descriptive statistics to create portraits of retained students and non retained students. Findings demonstrate that a number of student characteristi cs and involvement behaviors are relevant to college student retention. Student characteristics such as race/ethnicity, gender, and socio economic status influence persistence. When predicting retention, academic engagement, social engagement, and percepti ons of belonging among participating students differed in their
71 importance. Findings in regards to the three overarching research questions are summarized as follows: Research Question One: Is there a relationship between academic engagement and student re tention? The present study utilized factor analysis to create three academic engagement variables. The null hypothesis that academic engagement is not relevant to retention was accepted. The individual items that constructed each factor included interactio ns with faculty, academic interactions with peers, and academic behaviors. While the composite factors contributed to the overall model, each individual item exhibited a coefficient with a p value larger than 0.05. Research Question Two: Is there a relat ionship between social engagement and student retention? Social engagement, as defined by the available data, included levels of extra curricular activities. The null hypothesis that social engagement is not relevant to retention was rejected. Results indi cated that the strongest predictor of retention was indeed the amount of student engagement in extra curricular activities. Research Question Three: Is there a relationship between perceptions of belonging and student retention? Results indicated that the retention model improved significantly with the addition of factors relating to perceptions of belonging. The null hypothesis that perceptions of belonging are not relevant to retention was accepted. Similar to the academic engagement factors, the two per ceptions of belonging variables demonstrated coefficients with p values larger than 0.05. An important result was that by adding in the three academic engagement factors and the two perceptions of belonging factors, the model increases significantly in its power to predict retention, and more accurately fits the data when compared to the reduced model.
72 Additional analysis using logistic regression explored the influence of these variables as well as student characteristics on retention. Outcomes from the an alysis uncovered the fact that race/ethnicity played a role in the persistence puzzle at this university. Black/African American students were .208 times as likely to be retained compared to white students when controlling for all other factors in the full model. Additionally, Asian students were .74 times as likely to be retained compared to white students, whereas Hispanic students were .578 as likely to be retained. Socio economic status data was not available, but the loan amount was an indication of th financial need. For every unit of scaled student loan (loan amount divided by one hundred) acquired, the likelihood of retention decreased slightly (by .995 times), after controlling for other variables in the model. The influence of parental education on retention was also investigated. Parental education was found to have a statistically significant effect on retention, but in contrast did not significantly improve the prediction model. The fact that it did not play a significant role in the proposed retention model was contrary to previous research (Perna and Jones, 2013) and could be a result of the specific environment at this institution for this sub population. Extant research in regards to minority students, first generation students, an d low income students has indicated that many factors can influence retention for these sub populations, including: academic preparedness, social capital, involvement levels, and financial resources (Perna and Jones, 2013). Housing status also made notabl e contributions to the prediction model. Controlling for differences in gender, race/ethnicity, extra curricular activity, the amounts of loans, the levels of perceptions of belonging (both factors), and levels of academic
73 engagement for all three factors, a student who lived off campus was 0.529 times as likely to be retained compared to students who lived on campus. This is consistent with previous research (Pike, 1999) which demonstrated higher retention rates for students living in on campus housing. Th e evidence from this study suggests housing status as a possible factor to input in a monitoring system designed to identify high risk students. While this study explored the influence of academic engagement, social engagement, and perceptions of belongin g on college retention, it is limited by the nature of the instrument used to gather the information. Previous research by faculty interaction to be one of the most important variables c ontributing to college student persistence. The fact that this was not evident in this study could be an indication that the instrument did not adequately measure academic engagement. A similar limitation could be noted for the perceptions of belonging con structs. The instrument contained three wild card modules. The majority of items pertinent to perceptions of belonging were limited to one wildcard module. To include items relating to extra curricular activities, it was necessary to include responses fro m students who received a different wildcard module. Thus, items relating to perceptions of belonging were limited to those from the main survey. This is a potential limitation and should be explored with other instruments that can capture a robust variety of constructs. It is critical to note that this present study involved a single institution. Generalization across institutions is very difficult due to differences in the student demographics and the culture of the institutions themselves.
74 Discussion of the Findings The most striking result from this study suggests the variables of social versus academic engagement are not equivalent when it comes to predicting student retention. This is a relevant point that Tinto (2012a) emphasizes when he states that of the two systems are, however, not entirely symmetrical, and the degree of asymmetry academic and social systems of an institution. This is another w ay to describe the culture of the institution: its symmetry. This study investigated retention at a large, public, highly selective, residential institution, and as a result, the unique characteristics 2012a) recent work, the triad of social engagement, academic engagement, and perceptions of belonging was stressed. sense of belonging. Previous researchers stressed the importance of academic engagement (Astin, 1984, 1993; Pascarella and Terenzini, 1991, 2005; Terenzini and Pascarella, 1980). The engagement can come in the form of formal contact in the classroom and informal contact outside the classroom. Tinto (2012b) notes there is a connection between academic engagement and social engagement; this research, however, did not demonstrate any strong correlations between extra curricular activities and the academic engagement variables, nor the perceptions of belonging v ariables. There are many feasible explanations for these findings. They could be an artifact of large class sizes and the increase of online classes, they could be an indication of the culture of the institution, or they could be a reflection of the instr ument utilized.
75 In addition, the dimensions of engagement that are measured by the instrument could be restrictive. As discussed in Chapter Two, there is great variability in how researchers measure engagement the National Survey of Student Engagement (Ca rini, Kuh, & Klein, 2006; Carter, 2006), and College Student Experience Questionnaire (Hu and Kuh, 2002) are two frequently used instruments. In the study by Berger and Milem (1999), the research design included data gathered longitudinally from a number o f instruments over time. This enabled the researchers to examine a variety of types of involvement over the entire freshman year. One limitation of methods used in this study is the time factor in this analysis. The decision to include results from a singu lar instrument relates to the availability of data. To surmount this obstacle in future studies, one would have to initiate the use of multiple instruments strategically planned over a course of time and follow a cohort throughout their first year. As desc sample of U.S. college students. The lack of significance of the gender variable was surprising. Tinto (2012b) discusses differences between the persistence of women v ersus men. In research by Jones (2010), gender played a significant role in social integration. Pascarella and Terenzini (2005) noted that women were inclined to be retained and ultimately graduated at a higher rate than their male counterparts. While gend er did not act as a strong predictor of retention, it did play a role in the overall logistic model. Further research to explore and understand any underlying interplay could facilitate a remediation program.
76 This study takes a step towards understanding the patterns of retention that exist at one institution, and these patterns represent the first phase towards understanding the unique environment and challenges at this institution. The study facilitated the identification of sub populations within the s tudent body that should be explored with further enquiries. This study contributes to the study of retention by providing a portrait of retention at a large, public institution and a blueprint towards using available data to gain a deeper understanding of the unique landscape of this university. Recommendations for Future Research Retention research encompasses a large number of studies that examines a broad spectrum of influential factors, including student characteristics as well as institutional charact as a whole, none provide the complete picture or total explanation. This research contributes to the body of evidence by providing an analysis of data from a unique environment a large, public research university and an examination of the interactions between the academic and social systems of this institution. The research accomplished two things: one, it provided a vantage point for administrators, and two, it provided a starting point for more research. Porter (2013) has criticized surveys that utilized self reported student data. R ecommendations for future research are the following: 1. A multi institutional study could be carried out to examine predictive independent variables and patte rns across types of institutions. This type of research could postulate college retention models based on institution type. Strayhorn and DeVita (2010) provided an excellent example of a multi institutional study. Their research examined the student engage ment of African American males by institutional type. 2. A longitudinal study at this institution could be undertaken to examine different types of engagement variables and perceptions of belonging over time as they relate to retention. This research could si gnificantly help to understand if there are watershed
77 moments during the retention process at this institution. Longitudinal studies similar the processes of engagement on eac h campus. 3. Pair qualitative or what is often termed as authentic assessment with the quantitative survey data to enlighten researchers on the quality of different types of social engagement. This study utilized data gathered at one point of time rather th an studying a cohort. Furthermore, a longitudinal cohort study comprised of both quantitative and qualitative measures might provide insight and answers to the retention puzzle within the lenses of this institution. 4. One last recommendation would include a means to account for more complicated student attendance patterns such as student s transferring to other institutions or a student who drops out and returns later. Implications for Practitioners While social engagement may be an important factor in retent ion at this institution, further longitudinal studies will be needed before specific remediation programs can be recommended. The pressure for universities to create a healthy life cycle for students at their institution has increased. A healthy life cycle includes students entering, interacting, engaging, learning, and developing as students, and then completing their studies with successful graduation. It is the gaps in this cycle that practitioners are required to address. The implications of this resear ch for practitioners include: 1. The necessity to develop monitoring systems linked to predictive retention factors. Many researchers and practitioners are recommending monitoring systems to identify students at risk (Perna and Jones, 2013, Seidman, 2012; Tin to, 2012a). While monitoring systems can lend assistance, if the most potent factors are not identified, the monitoring system might not have the desired results of identifying students most at risk of attrition and in need of remediation and intervention. Assuming that college decision makers choose to treat or provide remediation programs for students who might be considered at high risk for attrition, I would suggest that they address distinctive populations separately. Alternative means to engage studen ts in extra curricular activities might need to be examined. Current research has highlighted the importance of academic support programs, orientation programs, and transition programs (Perna and Jones, 2013, Tinto, 2012a, 2012b).
78 2. Enrollment management s hould be carefully linked to retention strategies. This research demonstrated differences in the portrait of retention for different sub populations. It is important to note that as enrollment managers strive to recruit a more diverse pool of students, exp licit attention should simultaneously be given to retention strategies for these students (Hossler, Dundar, and Shapiro, 2013). 3. Practitioners will need to work closely with institutional research offices at their institutions to collect student data. From should arm themselves with data sets that collect pertinent variables which reflect the relevant student characteristics, watershed activities and processes. The climate of each institution needs to be taken int o account and the specific landscape of retention on their campus should be examined closely. As Tinto (2012b) succinctly states to be strategically matched with retention effor ts. The institutional context is a critical lynchpin to proposing improvements in regards to college access, retention, and completion. This institutional context, paired with strategic enrollment management efforts, greatly impact s the types of retention initiatives that should be recommended. Conclusion In conclusion, this study aimed to portray the retention landscape at a large, that involvement is critical to the pers istence process. While the most powerful factor for predicting retention was engagement in extra curricular activities rather than academic engagement, this study additionally found evidence that there are different patterns of demographics between retaine d and non retained students. For example, minority students and first generation students have lower rates of retention than their counterparts. Each institution is a unique community with its own academic and social culture paired with a unique population of students. Past retention models cannot act as a strict guide for all institutions. Rather each institution will need to create its own unique model comprised of the relevant retention factors that influence student success on their own campus. Due to t his disparity, understanding the departure puzzle and decreasing
79 the retention gap should continue to be a high priority for this institution and higher education in general.
80 LIST OF REFERENCES Allen, W.R. (1986). Gender and campus differences in black st udent academic performance, racial attitudes, and college satisfaction Atlanta, GA: Southern Education Foundation. Astin, A.W. (1970a). The methodology of research on college impact (I). Sociology of Education 43 223 254. Astin, A.W. (1970b). The meth odology of research on college impact (II). Sociology of Education 43 437 450. Astin, A.W. (1975). Preventing students from dropping out San Francisco: Jossey Bass. Astin, A.W. (1984). Student Involvement: A development theory for Higher Education. Jo urnal of College Student Development 25 (5), 297 308. Astin, A.W. (1993a). College retention rates are often misleading. The Chronicle of Higher Education 22 A48. Astin, A.W. (1993b). What matters in college: four critical years revisited San Francisco: Jossey Bass. Astin, A.W. (1996). Involvement in learning revisited: Lessons we have learned. Journal of College Student Development 37 (2), 123 134. Astin, A.W. (1999). Involvement in Learning Revisited: Lessons we have learned. Journal of Co llege Student Development 40 (5), 587 598. Axelson, R.D. & Flick, A. (2011). Defining Student Engagement. Change 43 (1), 38 43. Barefoot, B.O. (2000). The first year experience. About Campus 4 (6), 12 18. Barefoot, B.O. (2004). Higher education's revolv ing door: confronting the problem of student drop out in us colleges and universities. Open Learning 19 (1), 9 18. Barefoot, B.O. (2006) Foundations of Excellence: A New Model for First Year Assessment. Assessment Update 16 (2) 5 7. Barton, A., & Donah ue, C. (2009). Multiple assessments of a first year seminar pilot. The Journal of General Education 58 (4), 259 278. Bean, J. P. (1980.). Dropouts and turnover: The synthesis and test of a causal model of student attrition. Research in Higher Education 12 155 187. Bean, J. P. (1983). The application of a model of turnover in work organizations to the student attrition process. Review of Higher Education 12, 155 182.
81 Bean J. P. ( 1990 ). Strategic Planning and Enrollment Management. In J.M. Hossler and J .P. Bean (Eds.), The Strategic Management of College Enrollments (pp. 21 43). San Francisco, Cal.: Jossey Bass. Bean, L.P. (2005). Nine themes of college student retention. In A. Seidman (Ed.), College student retention: Formula for student success (pp. 2 15 243 ) Westport, CT: ACE/Praeger. Bean, J. P., & Eaton, S. B. (2000). A psychological model of college student retention. In John M. Braxton (Ed.), Rethinking the Departure Puzzle: New Theory and Research on College Student Retention. Nashville, TN: Vand erbilt University social integration, and first year persistence. Journal of College Student Development 38 441 452. Berger, J.B. (2000). Organizational behaviors at colleges an d student outcomes: a new perspective on college impact. Review of Higher Education 23 177 198. Berger, J.B., & Braxton, J. M. (1998). Departure through theory elaboration: Examining the role of organizational attributes in the persistence process. Research in Higher Education 39 (2), 103 119. Berger, J.B., & Milem, J.F. (1999). The Rol e of student involvement and perceptions of integration in a causal model of student persistence. Research in Higher Education 40 (6), 641 664. Braxton, J.M. (Ed.). (2000). Reworking the student departure puzzle Nashville, TN: Vanderbilt University Pres s. Braxton, J. M. (2004). Understanding and reducing college student departure San Francisco, Cal.: Jossey Bass. Braxton, J. M. Doyle, W.R., Hartley, A.S., Jones, W.A., & McLendon, M.K., (20 1 3 ). Leaving College: Rethinking the Causes and Cures of Student Attrition San Francisco, Cal.: Jossey Bass. Braxton, J. M., & Lien, L. (2000). The viability of academic integration as a central (Ed.), Reworking the departure puzzle (pp.1 1 28). Nashville, TN: Vanderbilt University Press Braxton, J. M., Sullivan, A., & Johnson, R. (1997). student departure. Higher education: Handbook of theory and research 12 107 158. Braxton, J. M., Vesper, N., & Hoss ler, D. (1995). Expectations for college and student persistence. Research in Higher Education 36 (4), 595 612.
82 Byrd, K.L., & MacDonald, G. (2008) Defining College Readiness from the Inside Out: First Generation College Perspectives Community College Rev iew 33 (1) 22 37. Cabrera, A.F. (1994). Logistic regression analysis in higher education: an applied perspective. Higher Education: Handbook of theory and research 10 225 256. Cabrera, A., Castanada, M., Nora, A., & Hengstler, D. (1992). The convergence between two theories of college persistence. Journal of Higher Education 63 (2), 143 164. Cabrera, A., & Nora, A. (1996). The role of perception and prejudice on the adjustment of minority students to college. Journal of Higher Education 67 (2), 119 148. Cabrera, A., Nora, A., & Castanada, M.B. (1992). The role of finances in the persistence process: A structural model. Research in Higher Education 33 571 593. Cabrera, A., Nora, A., & Castanada, M.B (1993). College persistence: Structural equations modeling test of an integrated model of retention. Journal of Higher Education 62 123 139. Carini, R.M., Kuh, G.D., & Klein, S.P. (2006). Student Engagement and Student Learning: Testing the Linkages. Research in Higher Education 47 (1), 1 32. Center for the Study of College Student Retention. (2008). Learning about college student retention Retrieved from www.cscsr.org/retention_issues_reten tion.htm Chickering, A.W., & Gamson, Z.F., (1987). Seven Principles for Good Practice in Undergraduate Education. AAHE Bulleting 39 (7), 3 7. Chickering, A.W., & Gamson, Z.F., (1999). Development and Adaptations of the Seven Principles for Good Practice in Undergraduate Education. New Directions for Teaching & Learning 80 75 81. Clark, M.H. & Cundiff, N.L. (2011). Assessing the Effectiveness of a College Freshman Seminar using propensity score adjustments. Research in Higher Education 52 616 639. College Student Retention (n.d.) Retrieved from on February 19, 2012 http://education.stateuniversity.com/pages/1863/College Student Retention.html Community College Trends and Statistics (n.d.). Retrieved from on February 19, 2012 http://www.aacc.nche.edu /AboutCC/Trends/Pages/default.aspx Dey, E.L & Astin, A.W. (1993). Statistical alternatives for studying college student retention: a comparative analysis of logit, probit, and linear regression. Research in Higher Education, 34(5), 569 581.
83 Dooris, M.J, & Blood, I.M. (2001). Implementing and assessing first year seminars Assessment Update, 13 (4), 1. Falcanetti, A. M. (2009). 2+2 Statewide articulation policy, student persistence, and success in Florida universities Community College Journal of Research and Practice 33 238 255. Fischer, M.J. (2007). Settling into campus life: differences by race/ethnicity in college involvement and outcomes Journal of Higher Education 78 (2), 125 161. Flowers, L.A. (2004). Examining the effects of student involvement on African American college student development Journal of College Student Development 45 633 654. Flowers, L.A., & Pascarella, E.T. (2003). Cognitive effects of college: differences between African American and Caucasian students Research in Higher Ed ucation 44 (1), 21 49. Foubert, J.D., & Urbanski, L.A. (2006). Effects of Involvement in Clubs and Organizations on the Psychosocial Development of First Year and Senior College Students Journal of Student Affairs Research and Practice 43 (1), 166 182. Francis, T. A. & Miller, M. T. (2008) Communication Apprehension: Levels of First Generation College Students at 2 year Institutions Community College Journal of Research and Practice 32 38 55. Gellin, A. (2003). The effect of undergraduate student involvement on critical thinking: a meta analysis of the literature 1991 2000 Journal of College Student Development 44 (6), 746 762. Getzlaf, S., Sedlacek, G., Kearney, K., & Blackwell, J. (1984). Two types of voluntary undergraduate attrition: Applica Research in Higher Education, 20 (3), 257 268. Gibson, A.M. & Slate, J.R. (2010). Student Engagement at Two Year Institutions: Age and Generational Status Differences. Community College Journal and Practice, 34, 371 385. Gilardi, S. & Guglielmetti, C. (2011) University Life of Non Traditional Students: Engagement Styles and Impact on Attrition. Journal of Higher Education 82 (1) 33 53. Goodman, K., & Pascarella, E. T. (2006). First year seminars increase persistence and retention: a summary of the evidence from how college affects students Peer Review 8 (3), 26 28.
84 Hagedorn, L.S., Moon, H.S., Cypers, S., Maxwell, W.E., & Lester, J. (2006) Transfer Between Community Colleges and 4 year Colleges: The American Game. Community College J ournal of Research and Practice, 30 223 242. Harper, S.R., Carini, R.M. Bridges, B.K., & Havek, J.C. (2004) Gender Differences in Student Engagement Among African American Undergraduates at Historically Black Colleges and Universities. Journal of College Student Development, 45 (3), 271 284. Harper, S.R., & Quaye, S.J. (Eds.). (2009). Student engagement in higher education: Theoretical perspectives and practical approaches for diverse populations, New York: Routledge. year college students. Journal of College Student Retention, 4 227 256. Horn, L.J., & Carroll, C.D. (1998) Stopouts or Stayouts? Undergraduates Who Le ave College in Their First Year, (National Center for Education Statistics Statistical Analysis Report No. NCES 1999 087), US Department of Education Office of Educational Research and Improvement, Washington, DC. Hossler J. M ( 1990 ). Enrollment Manageme nt in Academic Settings. In J.M. Hossler and J.P. Bean (Eds.), The Strategic Management of College Enrollments (pp. 1 20) San Francisco, Cal.: Jossey Bass. Hossler, D., Dundar, A., & Shapiro, D.T. (2013). Longitudinal Pathways to College Persistence and Completion: Student, Institutional, and Public Policy Perspectives. In L.W. Perna and A.P. Jones (Eds.), The State of College Access and Completion (pp. 140 165). New York, NY: Taylor & Francis. Hosmer, D.W., & Lemeshow, S. (2000) Model building strategies and methods for logistic regression. In Applied Logistic Regression, second edition (pp. 91 142). New York, NY: John Wiley & Sons, Inc. Hu, S. (2011) Reconsidering the relationship between student engagement a nd persistence in college. Innovative Higher Education, 36 97 106. Hu, S., & McCormick, A. (2012) An engagement based student typology and its relationship to college outcomes. Research Higher Education, 53 738 754. Hunter, M.S. (2006). Fostering student learning and success through first year programs. Peer Review 8 (3), 4 7. Hurtado, S., & Carter, D. F. (1997). Effects of college transition and perceptions of the elonging. Sociology of Education 70 324 345.
85 Integrated Postsecondary Education Data System (n.d.). Retrieved from on February 20, 2012 http://nces.ed.gov/ipeds/ Ishitani, T. T. (2003). A longitudinal approach t o assessing attrition behavior among first generation students: Time varying effects of pre college characteristics. Research in Higher Education, 44 (4), 433 449. Ishitani, T. T. & McKitrick, S.A. (2010). After Transfer: The Engagement of Community Colle ge Students at a Four Year Collegiate Institution. Community College Journal and Practice, 34, 576 594 Jamelske, E. (2008). Measuring the impact of a university first year experience program on student GPA and retention. Higher Education 57 373 391. Johnson, N. (2012 ). Delta Cost Project. The Institutional Costs of Student Attrition. Retrieved May 15, 2013, from Johnson, Nate. "The Institutional Costs of Student Attrition." The Delta Cost Project on Postsecondary Education Costs, Productivity, and Acc ountability. American Institutes for Research, Sep 2012. Web. 15 May 2013. . Johnson, D.R., Soldner, M., Leonard, J.B., Alvarez, P., Inkelas, K.K., Rown Kenyon, H.T., & Longerbeam, S.D. (2007). Examining Sense of Belonging Among First year Undergraduates from Different Racial/Ethnic Groups. Journal of College Student Development 48 (5), 525 542. Jones, P.A. (2013). Introduction: Improving Postsecondary Access, Persistenc e, and Completion in the United States: Setting the Stage. In L.W. Perna and A.P. Jones (Eds.), The State of College Access and Completion (pp. 1 9). New York, NY: Taylor & Francis. Keeley, E.J., & Davis, J.D. (1993, May). Transfer shock revisited: a long itudinal study of transfer academic performance Paper presented at the meeting of the Association for Institutional Research, Chicago, IL. Change 35 (2) 24 32. Kuh, G. D. & Hu, S. (2001 ) The effects of student faculty interaction in the 1990s. The Review of Higher Education, 24 309 332 Lee, R.M., & Davis, C.III (2000). Cultural orientation, past multicultural experience and a sense of belonging on campus for Asian college stude nts. Journal of College Student Development 41 (1), 110 115. Lester, J. (2006). Who will we serve in the future? The new student in transition. New Directions for Student Services, 114 47 61.
86 Mayhew, M.J., Vanderlinden, K., & Kim, E.K. (2010) A multi level assessment of the impact of orientation programs on student learning. Research in Higher Education 51 320 345. Marti, C. N. (2009) Dimensions of Student Engagement in American Community Colleges: Using the Community Colleg e Student Report in Research and Practice. Community College Journal of Research and Practice, 33 1 24. McCormick, A.C., Kinzie, J., & Gonyea, R.M. (2013) Student engagement: Bridging research and practice to improve the quality of undergraduate education In Higher Education: Handbook of Theory and Research. 47 92. McCormick, A.C., & McClenney, K. (201 2 ) Will these trees ever bear fruit? A response to the special issue on student engagement Review of Higher Education, 35(2) 307 333 .Milem, J.F., & Berger, J.B. (1997) A modified model of college student Journal of College Student Development 38 (4), 387 400. Moore, J., Loyell, C.D., McGann T., & Wyrick, J. (1998) Why involvement matters: A review of research on student involvement in the college setting. College Student Affairs Journal 17 (2), 4 17 National Center for Education Statistics (2011). Retrieved from on March 20, 2012 http://nces.ed.gov/programs/coe/tables/table pgr 1.asp National Institute of Education Study Group on the Conditions of Excellence in American Higher Education (NIE Study Group). 1984. Involvem ent in Learning: Realizing the Potential of American Higher Education. Washington, DC: National Institute of Education. Neumann, V., & Finaly and attrition: A quality of learning experience ap proach. Journal of Experimental Education 57 129 140. Nora, A. (2004). The role of habitus and cultural capital in choosing a college, transitioning from high school to higher education, and persisting in college among minority and nonminority students. J ournal of Higher Education, 3 180 208. Nora, A., & Cabrera, A. F. (1993). The construct validity of institutional commitment: A confirmatory factor analysis. Research in Higher Education 34(2) 243 262. Oseguera, L., & Rhee, B.S. (2009). The influence of institutional retention climates on student persistence to degree completion: a multilevel approach. Research in Higher Education 50 546 569.
87 Pancer, S. M., Pratt, M., Hunsberger, B., & Alisat, S. (2004). Bridging troubled waters: helping students make the transition from high school to university Guidance & Counseling 19 (4), 184 190. Pascarella, E.T. (1980). Student faculty informal contact and college outcomes. Review of Educational Research 50 545 595. Pascarella, E.T., Duby, P.B., & Iverson, B.K., (1983). A test and reconceptualization of a theoretical model of college withdrawal in a commuter institution setting. Sociology of Education 56 88 100. Pascarella, E.T., Pierson, C.T., Wolniak, G.C., & Terenzini, P.T. (2004). First generation college students. Journal of Higher Education 75 (3), 249 284. Pascarella, E.T., Seifert, T.A., & Blaich, C. (2010). How effective are the NSSE in predicting important educational outcomes? Change 42 (1), 16 22. Pascarella, E.T., & Terenzini, P.T. (1976). Informal interaction with faculty and freshman ratings of academic and non academic experience of college. Journal of Educational Research 70 (1), 35 41. Pascarella, E.T., & Terenzini, P.T. (1977) Patterns of student faculty informal interaction beyond the classroom and voluntary freshman attrition. Journal of Higher Education 48 (4), 540 552. Pascarella, E.T., & Terenzini, P.T. (1978). Student faculty informal relationships and freshman year educ ational outcomes. Journal of Educational Research 71 (4). Pascarella, E.T., & Terenzini, P.T. (1980). Predicting freshman persistence and voluntary dropout decisions from a theoretical model. Journal of Higher Education 51 (1) 60 75. Pascarella, E.T., & Terenzini, P.T. (2005). How College Affects Students: A Third Decade of Research. San Francisco: Jossey Bass. Perna, L.W., & Kurban, E.R. (2013). Improving College Access and Choice. In L.W. Perna and A.P. Jones (Eds.), The State of College Access and Completion (pp. 10 33). New York, NY: Taylor & Francis. Porter, S. R. (2011). Do college student surveys have any validity? Review in Higher Education, 35 (1), 45 76. Porter, S. R. (201 3 ). Self Reported Learning Gains: A Theory and Test of College Student Survey Response. Re search in Higher Education, 54 201 226 Porter, S. R., & Swing, R. L. (2006). Understanding how first year seminars affect persistence. Research in Higher Education, 47 (1), 89 109.
88 Porter, S. R., & Umbach, P.D. (2006). Student Survey R esponse Rates Across Institutions: Why do they vary? Research in Higher Education, 47 (2), 229 247. Radcliffe, P., Huesman, R., & Kellogg, J. (2010). Redefining student success: applying different multinomial regression techniques for the study of student graduation across institutions of higher education. Research in Higher Education 51 (2) 154 174. Radford, A.W., Berkner, L., Wheeless, S.C. & Shepherd, B. (2010). Persistence and attainment of 2003 04 beginning postsecondary students: After 6 years. Washington, DC: National Center for Education Statistics. Rosenberg, M., & McCullugh, B.C. ( 1981 ). Mattering: Inferred significance and mental health. Research in Community and Mental Health 2 163 182 Sadler, W.E., Cohen, F.L., & Kockesen, L. (1997). Fa ctors affection retention behavior: a model to predict at risk students. AIR 1997 Annual Forum Paper Strauss, L.C., & Volkwein, J.F. (2004). Predictors of student commitment at two year and four year institutions. Journal of Higher Education 75 (2) 203 2 27. Terenzini, P.T., Pascarella, E.T., & Blimling, G.S. (1996). of class experiences and their influence on learning and cognitive development: a literature review. Journal of College Student Development 37 (2) 149 162. Tierney, W.G. (1999) Models of minority college going and retention: cultural integrity versus cultural suicide. Journal of Negro Education. 68 (1), 80 91. Tinto, V. (1975). Dropout from higher education: A theoretical synthesis of recent research. Review of Educational Research. 45 89 125 Tinto, V. (1993). Leaving College: Rethinking the Causes and Cures of Student Attrition, 2nd edition. Chicago: University of Chicago Press. Tinto, V. ( 2012 ). Completing College : Rethinking Institutional Action Chicago: University of Chicago Press. Tinto, V., & Pusser, B. (2006). Moving from theory to action: Building a model of institutional action for student Success Washington, DC: National Postsecondary Education Cooperative. Retrieved from on May 20, 2013 http://nces.ed.gov/npec/pdf/Tinto_Pusser_Report.pdf Titus, M.A. (2004). An examination of the influence of institutional context on student persistence at 4 yer colleges and universities: a multilevel approach. R esearch in Higher Education, 45 (7), 673 699.
89 Titus, M.A. (2006a). Understanding the influence of the financial context of institutions on student persistence at four year colleges and universities. Journal of Higher Education, 77 (2), 353 375. Titus, M.A. (2006b). Understanding college degree completion of students with low socioeconomic status: The influence of the institutional financial context. Research in Higher Education, 47 (4), 371 398. Townsend, B.K., & Wilson, K. (2006). A hand held for a little bit: factors facilitation the success of community college transfer students to a large research university. Journal of College Student Development, 47 (4) 439 456. Weidman, J. (1989). Undergraduate socialization: A conceptual approach. Higher ed ucation: handbook of theory and research, 5 289 322. VanderShee, B.A. (2007). A freshman seminar success story: increasing student engagement, retention, and collaboration at the university of Pittsburgh Bradford. Zie, M.C., Reisen, C.A., Beil, C., & Cap lan, R.D. (1997). Predicting intention to remain in college among ethnic minority and nonminority students. The Journal of Social Psychology 137 (2), 149 160.
90 BIOGRAPHICAL SKETCH Jean received her Bachelor of Science degree in biology from McGill Univer sity in Montreal, Canada in 1983. In 1987, she received her Master of Science in botany from the University of Massachusetts in Amherst, Massachusetts. In 2001, she received her Master of Science in d ecision and i nformation Sciences from the University of Florida in Gainesville, Florida. Since then she has worked as a Coordinator of Statistical Research and the Director of Institutional Research for the College of Education at the University of Florida. She is currently the Associate Director for Administra tive Services at the J. Wayne Reitz Union at the University of Florida. Jean completed her doctorate in h igher e ducation a dministration and p olicy in May of 2014.