‘-
33%;?

.v....a..t‘¢~.<$um
5:de .
Pl- . .I

3
)1

%,
fir

mm?»

‘ u"...

.110:

“gm.

. vi.“
1:... 3.2.
.1». 153.!

.1...» .v:

 

 

(.27 I- LIBRARY
' I t Michigan 5'0“ a
University

 

 

This is to certify that the
dissertation entitled

INSPIRING THE LIFE OF THE MIND: AN EXAMINATION OF THE
ROLES OF RESIDENTIAL COLLEGE ENVIRONMENTS AND
MOTIVATIONAL ATTRIBUTES IN PROMOTING
UNDERGRADUATE STUDENTS’ INCLINATION TO INQUIRE AND
CAPACITY FOR LIFELONG LEARNING

presented by

Jody E. Jessup-Anger

has been accepted towards fulfillment

of the requirements for the
Ph. D. degree in Higher, Adult, and Lifelong
Education

 

 

Modem 61.56:”

Major Professor’s Signature

film’- qr 071qu
Date

MSU is an Afiinnative Action/Equal Opportunity Employer

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K lProj/Acc8PrelelRC/Date0ue indd

INSPIRING THE LIFE OF THE MIND: AN EXAIVIINATION OF THE ROLES OF
RESIDENTIAL COLLEGE ENVIRONMENTS AND MOTIVATIONAL
ATTRIBUTES IN PROMOTING UNDERGRADUATE STUDENTS’ INCLINATION
TO INQUIRE AND CAPACITY FOR LIFELONG LEARNING

By

Jody E. Jessup-Anger

A DISSERTATION

Submitted to
Michigan State University
In partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Higher, Adult, and Lifelong Education

2009

 

 

 

 

 

ABSTRACT
INSPIRING THE LIFE OF THE MIND: AN EXAMINATION OF THE ROLES OF
RESIDENTIAL COLLEGE ENVIRONMENTS AND MOTIVATIONAL
ATTRIBUTES IN PROMOTING UNDERGRADUATE STUDENTS’ INCLINATION
TO INQUIRE AND CAPACITY FOR LIFELONG LEARNING
By

Jody E. J essup-Anger

 

As postsecondary education is promoted as necessary for participating in the 21St f_

century knowledge economy, academics, policymakers, and the public have voiced _.

 

concerns about the quality and coherence of undergraduate education (AAC & U, 2007;
Barr and Tagg, 1995; U. S. Department of Education, 2006). Critics point to the Size,
scope, and multiple missions of large, public research universities as contributing to
students’ feelings of anonymity, disengagement, anddisconnection from faculty (Astin,
1993; Boyer, 1987; Gaff, 1970; Gamson, 2000; Guskin, 1994; Hawkins, 1999).
University administrators increasingly turn to residential colleges and other types
of living-learning programs to address the size and scale conundrum facing large research
universities. By creating smaller enclaves of students living together initially, taking part
in a shared educational endeavor, and using resources within their environment that stress
academics (Inkelas, Zeller, Murphy, & Hummel, 2006), administrators and faculty
purport to create the atmosphere of a small liberal arts college while still offering students
the resources of a large university (Magolda, 1994; Schuman, 2005). Despite this claim,
virtually no attention has been paid to whether and how these environments promote
values associated with a liberal arts education, including whether they deepen students’

inclination to inquire and capacity for lifelong learning. Identified by the Center of

  

Inquiry in the Liberal Arts at Wabash College as one distinctive outcome of a liberal arts

  

" ”"4 L1 ‘ ’II‘SI‘J‘H‘ 9’ 35' ' "' I'
‘maybrunldkbrgr‘b“:3“. :;‘e‘,,ii'!u.-U In! I a” .. .

education, having a deep inclination to inquire suggests that a student has a strong desire
to learn and continues to pursue intellectual growth. Closely connected to students’
inclination to inquire, is their capacity for lifelong learning, defined by Hayek and Kuh
(1999) as students’ ability to “‘learn to learn’ and interact effectively with others in a
complex, information-based society” (p. 4).

In the current study, I used Moos’ social-ecological framework, which accounts
for environmental context and individual characteristics (Moos, 1976, 1979, 1986), to
examine how students’ attributes (including their motivation and other sociodemographic
characteristics) and residential college environments were associated with students’
inclination to inquire and capacity for lifelong learning. With data collected from over
1800 undergraduate students in 24 residential colleges at 10 research universities across
the United States, I used hierarchical linear modeling techniques to ascertain the
statistically significant individual, contextual, and cross-level associations of variables
with students’ inclination to inquire and capacity for lifelong learning. I found that a
statistically Significant amount of variation students’ inclination to inquire and capacity
for lifelong learning was attributable to students’ residential college environment and that
much of the variation was explained by the liberal arts emphasis of the residential college
environment. Specifically, an ethos marked by academic challenge and high expectations
was associated with a deepened inclination to inquire and an ethos marked by out-of-
class interactions with faculty was associated with a deepened capacity for lifelong
learning. Furthermore, students’ motivational attributes and desire to obtain more than a
bachelor’s degree were also associated both outcome variables. Implications for theory,

research, practice, and policy are discussed.

DEDICATION

I dedicate this dissertation to Eric and Olivia who bring me a constant source of joy.

iv

ACKNOWLEDGEMENTS

If ‘it takes a village to raise a child,’ it takes an entire city of support to birth a
dissertation and a child at the same time. I would like to express my gratitude to the
faculty, friends, and farmly who have provided cOuntless hours of advice and support as I
indulged in this academic undertaking. First, to Kris Renn, my advisor, thank you for
stepping me through a large portion of the writing process while on sabbatical. I
appreciate the countless drafts you read while traversing across the world or recovering
from jetlag. Thanks too for bringing stories of your adventures to our meetings. At times
they were much more exciting than my struggles with data collection. To Kim Maier,
thank you for your patience as you helped me to learn a new method of data analysis. I
appreciate your willingness to meet, to challenge mythinking, and to be candid about the
balance of family and academe. You are a wonderful inspiration. To Marilyn Amey and
Ann Austin, thank you for your time, thoughtful questions, and guidance in
conceptualizing the study. I aspire to teach as well as each of you.

Thank you to the other faculty in the Higher, Adult, and Lifelong Education
program for contributing to my knowledge of higher education. From deepening my
understanding of the rOle of postsecondary education in society, to bolstering my skills in
conceptualizing and designing a study, the education you have provided is first-rate.
Thank you to Matt Wawrzynski for taking me on as a research assistant four years ago
despite my limited knowledge of quantitative methods. Thank you to John Dirkx for
inviting me to participate on your research team examining the impact of study abroad.

And, thank you to Jim Fairweather for serving as my ‘dissertation psychologist’, meeting

with me throughout the proposal and writing process, and raising good questions when I
started to get stuck.

I am fortunate to have an amazing family and group of fiiends. To the rest of the
cohort, thank you for providing a wonderfully supportive learning environment. I learned
more because of each of you. To the yoga moms, thank you for candidly sharing your
struggles with motherhood and advice for overcoming the bumps along the way. To
Okemos Kids Club, thank you for showing such love for Olivia that I never hesitated to
leave her in your care.

To my parents, thank you for believing in me and raising me in a way that has
enabled me to believe in myself. Mom, I appreciate your willingness to fly across the
country on short notice to help with Olivia so that I could have more time to write. You
always did it with such little hesitation that I was comfortable asking again.

To Olivia, thank you for providing me with a deeper sense of purpose and focus.
When I think back on this time, I imagine that one of my most lasting memories will be
caring for you. I cannot tell you how much joy you have brought to our lives.

Finally, to Eric, my partner, fiiend, and colleague, thank you for journeying with
me. I have no doubt that I learned more, enjoyed the process more, and ate better because

of you. Thank you for deepening my joy in life. We did it!

vi

TABLE OF CONTENTS

LIST OF TABLES .............................................................................. xi
LIST OF FIGURES ............................................................................ xii
CHAPTER 1
INTRODUCTION .............................................................................. 1
Purpose and Research Questions .................................................... 7
Conceptual Framework .............................................................. 8
The Environmental System ................................................. 11
The Personal System ......................................................... 12
Mediating Factors ............................................................ l3
Coping and Adaptation ...................................................... 13
CHAPTER 2
LITERATURE REVIEW ..................................................................... 15
Historical and Contemporary Role of Residential Colleges ............ 15
Historical Overview and Modern Day Challenges ............. l6
Outcomes of Residential Colleges and Living-Learning
Communities ................................................ 18
Conditions that Promote Students’ Inclination to Inquire and
Capacity for Lifelong Learning ................................... 25
Examining Students’ Inclination to Inquire ..................... 25
Empirical Studies Related to Students’ Inclination to
Inquire ....................................................... 26
Historical Overview of Lifelong Learning ...................... 29
Recent Empirical Work on Lifelong Learning .................. 32
Relevant Motivation Theory ................................................ 37
Existing Studies Taking an Ecological Approach ........................ 39
Summary of Literature Review ..................................................... 41
CHAPTER 3
STUDY DESIGN .............................................................................. 42
Research Design ...................................................................... 43
Population, Sample, and Participants .............................................. 43
Data Collection Instruments, Variables, and Materials .......................... 52
Independent Variables ....................................................... 52
Pre-College Characteristics ....................................... 52
Motivation ............................................................ 53
Discipline ............................................................ 55
Liberal Arts Experiences .......................................... 56
Dependent Variables ......................................................... 62
Inclination to Inquire ............................................... 62
Capacity for Lifelong Learning ................................... 64

vii

Data Analysis Procedures ............................................................ 66

Limitations ............................................................................. 75
CHAPTER 4
RESULTS ....................................................................................... 77

Research Question One: Does students’ inclination to inquire or
capacity for lifelong learning vary across residential college
environments? ......................................................................... 77

Research Question Two: How are students’ sociodemographic
characteristics and motivation related to their inclination to

inquire and capacity for lifelong learning? ............................. 79
Sociodemographic Variables ....................................... 82
Pro-College Characteristics........................... ............. 84
Motivation, College Experiences, and Aspirations ............. 85
Liberal Arts Experiences ........................................... 86
Summary of Within-Residential College Model Findings ...................... 89

Research Question Three: Does the association between students’
sociodemographic attributes, pre-college characteristics,
and motivation and their inclination to inquire and capacity
for lifelong learning vary across residential colleges? ........... 90
Research Question Four: How is the environmental context of a
residential college related to students’ inclination to inquire

and capacity for lifelong learning? ......................................... 94

Environmental Liberal Arts Experience Variables ............. 95

Disciplinary Focus .................................................. 100
Summary of Between-Residential College Model Findings .................... 101

Research Question Five: Does the association between student-
level motivation and students’ inclination to inquire and
capacity for lifelong learning differ by residential college-

level liberal arts experience variables? ................................... 102
Summary of Results .................................................................. 104
CHAPTER 5
DISCUSSION AND IMPLICATIONS ...................................................... 107
Discussion of Findings ............................................................... 110
Null Model .................................................................... l l l
Within-Residential College Model ........................................ 113
Sociodemographic Attributes ..................................... 1 13
Pre-College Characteristics ............................... 1 17
Motivation, College Experiences, and Degree
Aspirations ......................................... 1 18
Individual Liberal Arts Experiences ..................... 120
Between-Residential College Model ...................................... 123
Inclination to Inquire ............................................... 124
Capacity for Lifelong Learning ................................... 127
Revisiting Moos’s Social Ecological Framework ........................ 129

viii

The Environmental System ......................................... 130

The Personal System ................................................. 132
Mediating Factors ...................................................... 132
Efforts at Adaptation ................................................. 133
Student Stability and Change ....................................... 134
Implications for Theory, Research, Practice, and Policy ........................ 135
Implications for Theory ..................................................... 135
Implications for Research ................................................... 136
Implications for Practice ..................................................... 139
Student Advising and Support ..................................... 140
Faculty and Student Interaction ................................... 142
Implications for Policy ....................................................... 144
Suggestions for Future Research .................................................... 146
Conclusion ............................................................................. 148
APPENDIX A: SURVEY ..................................................................... 150
APPENDIX B: VARIABLE INFORMATION ............................................ 166
APPENDIX C. FACTOR ANALYSIS OF MOTIVATION (GENERAL NEEDS
SATISFACTION) SCALE .......................................................... 173
APPENDIX D: FACTOR ANALYSIS OF LIBERAL ARTS EXPERIENCE
SCALE-ACADEMIC CHALLENGE AND HIGH EXPECTATIONS ...... 175
APPENDIX E: FACTOR ANALYSIS OF LIBERAL ARTS EXPERIENCE
SCALE-DIVERSITY EXPERIENCES ........................................... 178
APPENDD( F: FACTOR ANALYSIS OF LIBERAL ARTS EXPERIENCE
SCALE-INTERACTIONS WITH PEERS ....................................... 180
APPENDIX G: FACTOR ANALYSIS OF IN CLINATION TO INQUIRE
SCALE ................................................................................. 182
APPENDIX H: FACTOR ANALYSIS OF CAPACITY FOR LIFELONG
LEARNING SCALE ................................................................. 184
APPENDIX 1: BIVARIATE CORRELATIONS OF LEVEL-IVARIABLES ........ 186
APPENDIX J: BIVARIATE CORRELATIONS OF LEVEL-2 VARIABLES ...... 190
APPENDIX K: FULL MODEL OF INCLINATION TO INQUIRE .................. 192

ix

APPENDIX L: FULL MODEL OF CAPACITY FOR LIFELONG LEARNING... 199

REFERENCES ................................................................. _ ................ 2 06

LIST OF TABLES

Table 3.1: Sample and Distribution Information ............................................. 47
Table 3.2: Descriptive Information of Analytic Sample ..................................... 49

Table 3.3: New Scales for Good Teaching and High Quality Interactions with
Faculty ................................................................................... 59

Table 4.1: Null Models: Partitioning Variance in Inclination to Inquire and
Capacity for Lifelong Learning ....................................................... 78

Table 4.2: Association between Students’ Sociodemographic Characteristics,
Motivation, and College Experiences and their Inclination to Inquire and
Capacity for Lifelong Learning ...................................................... 81

Table 4.3: Deviance, Chi—Square Change, and Significance of Random Level-One
Slopes .................................................................................... 91

Table 4.4: Association between Environments Marked by Liberal Arts Experiences
and Students’ Inclination to Inquire ................................................. 97

Table 4.5: Association between Environments Marked by Liberal Arts Experiences
and Students’ Capacity for Lifelong Learning ..................................... 98

Table 4.6: Deviance, Chi-Square Change, and Significance of Cross-Level

Interactions between Motivation (Student) variables and Liberal Arts
Experience (Environment) variables ................................................ 103

xi

LIST OF FIGURES
Figure 1.1: Adaptation of Moos (1979) Model of the Relationship between
Environmental and Personal Variables and Student Stability and Change. . 11
Figure 5.1: Relationship between my Research Questions and Moos (1979) Model

of the Relationship between Environmental and Personal Variables and
Student Stability and Change ......................................................... 130

xii

CHAPTER ONE
Introduction

In the report Reinventing Undergraduate Education: A Blueprint for America ’s
Research Universities, the Boyer Commission on Educating Undergraduates in the
Research University (1998) called for research universities to redesign their
undergraduate curriculum systematically with an eye toward coherence and excellence.
The report was issued on the heels of growing concerns regarding fiagmentation of
undergraduate education in the United States, which many educators argued negatively
influenced motivation, persistence, and graduation rates of undergraduate students (Barr
& Tagg, 1995; Boyer Commission, 1998; Duderstadt, 2000). The call was especially
important because, next to community colleges, undergraduate students attending
nonprofit institutions are educated most often at research universities (National Center for
Educational Statistics, 2006). Often the sheer size of these institutions detracts fi'om
students’ ease in finding the sense of connection that they might at a smaller school
(Boyer, 1987; Gaff, 1970; Hawkins, 1999). Among the recommendations of the report
was a call for research universities to “cultivate a sense of place” by creating shared
rituals, nurturing community spirit within the residence halls, forming collaborative study
groups and project teams to customize the community, and using co-curricular activities
to bring students together (Boyer Commission, 1998, p. 34).

The development of living-learning communities is one promising practice that
addresses the size and coherence conundrum at research universities. These communities
consist of a cohort of participants who “(1) live together on campus, (2) take part in a

shared academic endeavor, (3) use resources in their residence environment designed

specifically for them, and (4) have structured social activities in their residential
environment that stress academics” (Inkelas, Zeller, Murphy, & Hummel, 2006, p. 11).
Residential colleges are a subset of living-learning communities that are comprehensive
in form, incorporating the community-oriented ethos, academic emphasis, and co-
curricular focus throughout undergraduate students’ entire educational experience, and
typically culminating in the attainment of a degree. In their ideal form, these communities
cohere and integrate students’ intellectual and personal lives (Smith, 1994). Smith
explained that “well developed residential colleges address critical higher education
issues: involvement, community building, diversity, retention, bridging the gap between .
ideal academic standards and actual student performance, fostering faculty and student
interaction” (p. 242). Robert O’Hara, an evolutionary biologist and residential college
advocate, explained his afi'mity for their structure, stating

I have spent many years living and working in residential college settings as a

graduate student and a faculty member, and 1 have also had the contrasting

experience of being an undergraduate in a big university without residential
colleges. I am convinced that small residential colleges within a large university
provide the best educational environment for students and faculty alike. (O’Hara,

n. d., 11 1)

Among the claims of many residential colleges at research universities is that they
emulate the atmosphere of a small liberal arts college with their intimate feel,
commitment to high quality undergraduate education, and holistic approach to learning,
but boast the resources of a large university, which include comprehensive research and

library facilities (Magolda, 1994; Schuman, 2005). The website of Lyman Briggs College

(LBC), a residential college at Michigan State University, for example, makes the
following claim, “All under one roof, LBC encompasses physics, chemistry, biology, and
computer laboratories; classrooms; faculty, administrative, and academic support staff
offices; student residences; a dining hall; and a convenience store. . .. LBC offers the
benefits of a small, liberal arts college with the resources of a great research university”
(Lyman Briggs College Website, n.d., 1] 1).

Implicit in the comparison of residential colleges to small liberal arts colleges is
an assumption that these small enclaves within research institutions are devoted to the
liberal arts ideal, which consists of a clearly defined mission promoting students’
intellectual development and values congruent with their mission, including a
commitment to holistic student development, to democratic ideals, and to the creation of
lifelong learners (Hawkins, 1999; Hirt, 2006; Kuh, Kinzie, Schuh & Whitt, 2005;
Michalak & Robert, 1981). The goals of the University of Michigan’s Residential
College, which include the desire to “foster a genuine appreciation and lifelong passion
for learning; not merely individual quests for knowledge, but preparation and
encouragement that lead to effective and responsible engagement in the real world,”
illustrate the presence of the liberal arts ideal within the community (University of
Michigan Residential College website, 11. d., 1] 2).

Also implied by the comparison of residential colleges to small liberal arts
colleges are assumptions regarding quality of instruction and importance placed on the
development of intellect. Pascarella, et a1. (2005) examined the environmental differences
in liberal arts institutions, regional institutions, and research universities and found that

liberal arts college environments in general generate greater student-faculty contact,

active learning/time on task, expectations, quality of teaching, and a supportive campus
environment. A recent working definition of liberal arts education proposed by Blaich,
Bost, Chan, and Lynch (2004) supports this finding, highlighting the conditions they
believe encapsulate liberal arts education, including,
1. An institutional ethos and tradition which places a greater value on
developing a set of intellectual arts, than professional or vocational Skills.
2. Curricular and environmental structures that work in combination to create
coherence and integrity in students’ intellectual experiences.
3. An institutional ethos and tradition which places a strong value on student-
student and student-faculty interactions both in and out of the classroom. (p.
2)
There is growing evidence demonstrating positive learning outcomes associated
with living-learning communities and residential colleges (Pascarella & Terenzini, 2005),
including greater academic achievement (Inkelas & Weisman, 2003; Pasque & Murphy,
2005), social integration and involvement (Pike, 1999; Stassen, 2003), and intellectual
engagement (Pasque & Murphy, 2005). Less attention has been paid to whether and how
these environments foster the transformative educational outcomes espoused (and often
achieved) by liberal arts colleges (Kuh, et al., 2005; Pascarella, et al., 2005), including
Whether they deepen students’ inclination to inquire and capacity for lifelong learning.
Identified by the Center of Inquiry in the Liberal Arts at Wabash College as one
distinctive outcome of a liberal arts education, the inclination to inquire
[S]uggests a strong desire to learn, to ask questions, and a willingness to consider

new ideas. Such learning involves taking initiative, not being satisfied with a .

quick answer, and intrinsic motivation for intellectual growth. These learning

traits lend themselves to a lifelong pursuit of knowledge and wisdom. (Center of

Inquiry in the Liberal Arts at Wabash College Website, n.d., 1i 3)

Closely connected to students’ inclination to inquire is their capacity for lifelong
learning, which is defined by Hayek and Kuh (1999) as students’ ability to “‘learn to
learn’ and interact effectively with others in a complex, information-based society” (p. 4).
Whereas a deepened inclination to inquire promotes a value for continuing to pursue
knowledge, a robust capacity for lifelong learning provides students the tools to act upon
their value for inquiry.

Examining students’ inclination to inquire and their capacity for lifelong learning
is more relevant (and interesting) as access to postsecondary education widens and
student demographics become more diverse. Whereas historically upper-class White men
pursued a bachelor’s degree to claim their place in civic life (Thelin, 2004), today access
has widened and myriad students pursue postsecondary education for a variety of
reasons. Many people believe that participation in postsecondary education is becoming
de facto compulsory because national and individual economic prosperity are
increasingly tied “to the attainment of high levels of knowledge and skill, and to the
ability to continue learning over a lifetime” (Association of American Colleges and
Universities (AAC & U), 2007, p. 7). Colleges and universities are being called upon to
provide students with the knowledge and skills necessary to successfully navigate an
increasingly complex and globally focused world, and also to promote an education-
focused mindset that will prepare students for a volatile and evolving workforce requiring

continuous learning (AAC & U, 2007). A survey by the Bureau of Labor Statistics (2006)

indicated that people born in the late baby boom years (1957 to 1965) changed jobs 10
times from the time they turned 18 until they turned 40. With the advent of the global
knowledge economy, it is probable that younger workers will not only change jobs as
often, but also require retraining within their positions.

Compounding the difficulty of the charge directed at colleges and universities to
develop students who possess an inclination to inquire and the capacity for lifelong
learning is the reality that Students entering postsecondary education are more disengaged
academically than ever before, with a greater number reporting frequently being bored in
class and studying less during their last year in high school (Astin, Oseguera, Sax, &
Korn, 2002). Furthermore, as access to postsecondary education widens, more students
view a college education merely as the next logical step in their educational trajectory
(AAC & U, 2007), necessitating that more attention he paid to understanding the
interaction between students’ motivation and their collegiate environment, and
specifically how to create environments that foster and develop students’ inclination to
inquire and capacity for lifelong learning.

The size and scope of many large research universities, with the accompanying
research pressures and emphasis on graduate study, limit the extent to which they focus
on undergraduate education (Boyer, 1987; Jerome, 2000). Large classes often curb
faculty and student interaction and constrain the range of assignments possible (Boyer,
1987; Guskin, 1994). Students also feel more anonymous in large environments, and
report less willingness to get involved than they might in a smaller environment (Astin,
1977; Gamson, 2000; Guskin, 1994). Despite the educational limitations of large research

universities, they remain a likely destination for many students to begin or complete their

baccalaureate education because of their relative affordability and diversity in educational
offerings. Consequently, it is vital that faculty and administrators at these large
institutions find ways to create environments that help students feel connected to their
curricular and co-curricular pursuits, thereby deepening their inclination to inquire and
capacity for lifelong learning. Although residential colleges offer one way of addressing
the scale conundrum, it remains unclear how these environments interact with the
students within them to influence their inclination to inquire and their capacity for
lifelong learning.
Purpose and Research Questions

The purpose of this study was to examine how students’ motivational attributes
and environmental contexts facilitated or impeded students’ inclination to inquire and
capacity for lifelong learning in residential college settings. Using data collected from
over 1800 students in 24 residential colleges, 1 investigated the following questions:

1. Does students’ inclination to inquire or capacity for lifelong learning vary .
across residential college environments?

2. How are students’ sociodemographic characteristics and motivation related to
their inclination to inquire and capacity for lifelong learning?

3. Do the associations between students’ sociodemographic attributes and
motivation and their inclination to inquire and capacity for lifelong learning
vary across residential colleges?

4. How is the enviromnental context of a residential college related to students’

inclination to inquire and capacity for lifelong learning? Specifically,

3. Environmental Liberal Arts Experience variables — (good teaching and
high quality interaction with faculty, academic challenge and high
expectations, diversity experiences, and peer interactions)

b. Disciplinary Focus

5. Does the association between student-level motivation and students’
inclination to inquire and capacity for lifelong differ by residential college-
level liberal arts experience variables (good teaching and high quality
interaction with faculty, academic challenge and high expectations, diversity
experiences, and peer interactions)?

Conceptual Framework

In their comprehensive review of research related to college impact, Pascarella

and Terenzini (2005) encouraged researchers to acknowledge the multitude of factors
affecting student change, and to adopt broader conceptual models which might “more
fully account for the multiple sources of influence,” instead of relying upon a single
disciplinary perspective or dimension of students’ experiences (p. 630). Their call echoes
that of human development researchers, who in the mid-1970s began to shift their
perspectives and research designs beyond traditional experimental models involving only
a subject and experimenter, instead expanding their gaze to include the social contexts in
which development occurs and those that influence development indirectly
(Bronfenbrenner 1979; Moos & Insel, 1974). Influenced by Kurt Lewin’s proposition that
behavior is a fimction of a person and an environment (1936), two complementary human
ecology models developed virtually simultaneously. The ecology of human development

arose out of developmental psychology, specifically from Urie Bronfenbrenner’s work

exploring infant and adolescent development. Bronfenbrenner (1979) emphasized the
importance of studying human development in the context of “the actual environments,
both immediate and remote, in which human beings live” (p. 12). His theory stressed the
importance of studying development within the context it occurs, and specifically how
biological factors — including physical characteristics and genetic propensities — interact
with the “immediate environment, and the way in which this relation is mediated by
forces emanating from more remote regions in the larger physical and social milieu”
(Bronfenbrenner, 1979, pp. 12-13, italics in the original).

Simultaneous to the emergence of the ecology of human development, social
ecology, “the multidisciplinary study of the impact that physical and social environments
have on human beings” arose out of Stanford University, and held three basic
assumptions:

(1) Human behavior cannot be understood apart from the environment in which it

finds its expression. Accurate predictions about behavior simply cannot be made

only from information about the individual; information about the environment is
essential. (2) Physical and social environments must be studied together since
neither can be fully understood without the other. . . [and] (3) Social ecology has
an explicit value orientation in that it attempts to provide knowledge relevant to

promoting maximally effective human functioning. (Moos & Insel, 1974, p. ix-x)

Whereas the ecology of human development emphasized the interaction of direct '
and indirect environmental effects on biologically determined development
(Bronfenbrenner, 1979), social ecology theory placed more emphasis on the immediate

physical environment as a mediator of development, and underscored the importance of

creating an environment that promotes effective human functioning (Moos & Insel,
1974). For the current study, I use the social ecological framework because of its explicit
acknowledgement of the importance of the immediate environment. However, I also
incorporate into my research design Bronfenbrenner’s notion of environments as
multiple, interacting, and influential even when not directly interacting with the student,
the environment as representing an ‘ethos,’ so to speak.

Rudolph Moos (1976, 1979, 1986) developed a social-ecological framework to
evaluate educational settings. His work stemmed from his observations that by focusing
only on personal traits (including students’ demographic characteristics, motivation, or
values) or environmental settings (including a residential college environment),
researchers could not adequately account for variations in behavior (Moos, 1979).
Instead, Moos believed it important to consider both the social-environmental and
physical-environmental variables together (Moos, 1976). Moos’s (1979) model, “notes
the existence of both environmental and personal systems, which influence each other
through selection factors...[and] mediating processes of cognitive appraisal and activation
or arousal (motivation)” (p. 4). These mediation processes typically arise when the
environment necessitates a response and result in efforts at adaptation and use of coping
skills. The initiation of adaptation efforts may change both the environmental and the
personal systems, and ultimately determines stability or change in student behavior. A
visual representation of the model is depicted in figure 1.1 and each variable in the model

is described below.

10

 

A

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Environment #
system ‘
7 Student
Stability
Cognitive Activation. gem at &
or Arousa optotion c
Appraisal :5 :3 and :9 hango
Coping
Personal
system

 

 

 

 

 

 

 

 

 

 

 

A A

 

Figure 1.1: Adaptation of Moos (1979) Model of the Relationship between
Environmental and Personal Variables and Student Stability and Change

The Environmental System. Moos (1979) described four major domains of
variables within the environmental system, including “the physical setting, organizational
factors, the human aggregate, and social climate” (p. 6), each of which can potentially
influence educational outcomes directly or indirectly thrOugh interaction with the other
environmental variables. The physical setting includes the physical design and
architecture of the environment. In a residential college setting, the physical setting may
include the building in which the college is housed, the presence or lack of study and
gathering spaces, and the amenities provided within the college. Organizational factors
include such dimensions as size of the residential college, faculty-student ratio, and
offerings provided to students in the way of co-curricular activities. The human aggregate
is comprised of the total characteristics of students in the setting, and may include “age,
ability level, socioeconomic background, and educational attainment” (p. 8). Faculty and

staff characteristics may also be part of the human aggregate. Moos found the human

11

aggregate pertinent to the environmental system because of the “notion that most of the
social and cultural environment is transmitted though other people,” and the implication
that “the character of an environment depends in part on the typical characteristics of its
members” (p. 8). Also included in the human aggregate would be the collective attitudes
of students, and their collective beliefs about the environment as promoting or thwarting
their educational pursuits. The fourth domain, social climate, is inferred by the
“continuity and consistency in otherwise discrete events” (p. 10). Within a residential
college, the social climate would be the integrating features of the environment, including
the overall attitude of students toward one another with regard to learning. In addition to
serving as a domain of the environmental setting, Moos viewed the social climate as a
mediator of the other environmental variables.

The Personal system. AS might be expected, individual characteristics that assist
in explaining students’ responses to an environmental context comprise the Personal
System. “Background and personal indexes include age, sex, ability level, interests and
values, ego strength and self-esteem, and preferences for such coping styles as active
engagement in the environment, tension reduction and exploration” (Moos, 1979, p. 11).
Other personal factors considered within the Personal System include attitudes,
expectations, and roles. Moos explained that “People who have more responsible
organizational roles (such as administrators, professors, and teachers, as compared with
high school and college students) tend to perceive educational settings more positively,”
and furthermore, “Expectations of new environments can influence both an individual’s

choice and later perception of an environment” (p. 11).

12

Mediating Factors. Moos (1979) identified two factors that mediate the
interaction between the Environmental System and the Personal System, namely 1)
Cognitive appraisal and 2) Activation or arousal. Cognitive appraisal is the process by
which an individual evaluates the environment as “being either potentially harmful,
beneficial, or irrelevant (primary appraisal) and his or her perception of the range of
available coping alternatives (secondary appraisal)” (p. 11). Activation or arousal occurs
when an individual appraises the environment as needing a response, which in turn
“prompts efforts at adaptation, or coping, which may change the environmental system
(students decide to use a recreation room as a library or study hall) or the personal system
(students seek and obtain information that changes their attitudes or expectations)” (p.
12).

Coping and Adaptation. Moos (1979) explained that although situations chosen to
study coping and adaptation usually involve major life changes including death, financial
disaster, and serious illness, more common transitions and everyday Situations including
the transition from adolescence to adulthood also demand coping responses. Moos also
noted coping and adaptation are not only mediators of outcomes, but also can become
outcomes, depending on the interest of the investigator.

The social-ecological framework provided a general model for examining the
process of person-environment interaction. It is relevant to the current study because it
considers the dialectic relationship between person and environment, with each
potentially transforming the other. In other words, the model allows for the possibility
that motivated students transform their environment into a setting in which a deepening

inclination to inquire and an expanding capacity for lifelong learning are the norm, and

13

also the possibility that converging environmental factors transform students toward a
deeper inclination to inquire and capacity for lifelong learning. By using an ecological
approach, not only was I able to examine how students’ motivation influenced students’
inclination to inquire and capacity for lifelong learning, I was also be able to examine the
interaction between students’ motivation and their environments, exploring how the
interactive effects evoke or inhibit students’ inclination to inquire and capacity for
lifelong learning. In chapter two, I situate the study in existing literature related to
residential colleges, the growth of students’ inclination to inquire and capacity for
lifelong learning, motivation theory, and ecological approaches to postsecondary
educational research. In chapter three, I explain the research design, detailing the
methods by which I carried out the study. In chapter four, 1 detail the findings of the
study. Finally, in chapter five, I discuss the findings of the study in light of existing
research, the implications of the findings for research and practice, and detail future

research.

14

CHAPTER TWO
Literature Review

The purpose of this chapter is to locate the current study within the context of
existing literature. The chapter is divided into four sections that inform and situate the
examination of how environmental contexts and motivational attributes facilitate or
impede students’ inclination to inquire and capacity for lifelong learning in residential
college settings. Section one situates the study in the context of residential college and
living-learning community literature, examining the historical and contemporary role of
residential colleges and living-learning communities in US postsecondary institutions;
section two examines existing literature and empirical research outlining the conditions
that promote students’ inclination to inquire and capacity for lifelong learning; section
' three reviews relevant motivation theory, outlining how students’ motivational attributes
may influence their inclination to inquire and capacity for lifelong learning; and section
four provides an overview of existing studies taking an ecological approach to
understanding students and environments within postsecondary education.

Historical and Contemporary Role of Residential Colleges in the United States

Although residential colleges have been a part of the higher education landscape
in the United States since Harvard was founded in 163 6, only relatively recently have
there been attempts to assess the outcomes of these environments on students. This
section situates the current study in the context of existing literature, detailing a historical
overview of residential colleges in the United States and providing information about
modern day challenges, while also reviewing recent empirical literature that illustrates

some of the outcomes of these communities.

15

Historical Overview and Modern Day Challenges

Much of the literature on residential colleges in the United States is descriptive,
detailing their historical role and evolution over time (Duke, 1991; Gaff, 1970; Ryan,
1992, 1993; Smith, 1994). Ryan (1992, 1993) traced the foundation of residential
colleges to the English College model of Oxford and Cambridge, where older fellows
instructed younger ones about subject matter and character, expanding the content
cOvered beyond the classroom and into students’ day-to-day lives. Although the structure
of the early colonial colleges such as Harvard and Yale reflected a commitment to
holistic education and character development with residential facilities and tutors to
oversee day-to-day life, the expansion of higher education and the incorporation of the
German model (with its nonresidential structure and intense focus on research and
graduate training) raised questions about the fundamental purpose of higher education
(O’Hara, 2001; Ryan, 1993), and the way it Should be structured.

The early 19003 brought about intense debate about the purpose of higher
education and a partial revival of the collegiate ideal, as college leaders including
Woodrow Wilson, who was then the president of Princeton, encouraged movement away
fi'om the laissez-faire elective system toward a more coherent curriculum (Ryan, 1992,
1993). At large universities, curricular innovation spurred initiatives that attempted to
carve the campus into smaller communities. The Experimental College at the University
of Wisconsin, developed in the early 19203 by Alexander Meiklejohn, was intended to
assist students in actively exploring the values of democracy through an integrated
curriculum that was designed to facilitate faculty—student interaction (Smith, 2001).

Although the experiment was short-lived, it provided a template of integrating small

16

residential communities within research universities, a model that continues to be
emulated today (Gabelnick, MacGregor, Matthews, & Smith, 1990; Matthews, Smith,
MacGregor, & Gabelnick, 1997; Ryan, 1992, 1993).

By the 19608, the rapid expansion of postsecondary institutions led many leaders
at research universities to invest resources in creating smaller communities for
undergraduates. Residential colleges gained momentum within a variety of different
institutions, including Wayne State University, Wesleyan University, University of the
Pacific, University of California San Diego, the University of Michigan, University of
Massachusetts-Amherst, Michigan State University, and F ordham University, among
others (Gaff, 1970). In addition, new institutions like University of California Santa Cruz
and the Claremont Colleges were created that partitioned the larger campus into small
enclaves, incorporating the residential college structure throughout the larger campus
(Duke, 1991). Duke (1991) explored the historical evolution of residential colleges
beginning with Harvard, Yale, and the University of Chicago and later colleges including
the Claremont Colleges and University of California Santa Cruz. He argued that the
English Oxbridge model, which provided a rationale and model for the development of
residential colleges in the United States, was often used by administrators symbolically to
gain support for these endeavors. However, the model was widely misunderstood by
leaders who were advocating for reform, and as a consequence has resulted in difficulty
maintaining the residential college structure at many institutions (Duke, 1991).

The difficulty creating and maintaining residential colleges and other types of
living-learning programs becomes evident when one examines the breadth of literature

describing how to implement and sustain these communities (Gabelnick et al., 1990; Hart

17

& Smith, 1993; Hohenbary et al., 1993; Lenning & Ebbers, 1999; Macy, 1993; O’Hara,
2001; Shapiro & Levine, 1999; Smith, 1994). Common difficulties cited in the literature
include recruiting faculty participation in the communities (Gabelnick, et al., 1990;
Lenning & Ebbers, 1999), planning for sustained development of the communities
(Lenning & Ebbers, 1999; Gabelnick et al., 1990), bridging the student affairs/academic
affairs divide (Gabelnick, et al., 1990; Lenning & Ebbers, 1999; O’Hara, 2001), finding
an administrative home for a community (Gabelnick, et al., 1990), and maintaining
facilities conducive to community building (O’Hara, 2001). Despite these modern day
challenges, recent empirical research details mostly positive outcomes associated with
these communities on undergraduate students’ learning and engagement.
Outcomes of Residential Colleges and Living-Learning Communities

Interest in exploring the outcomes of residential colleges and living-learning
communities has increased with the rise in the number of these initiatives on college and
university campuses. Much of this existing research has focused on the outcomes of
living-learning communities, which are often less comprehensive than residential
colleges because they may only accommodate a student for one or two years, may offer
less curricular integration across courses, or may focus solely on students’ extra-
curricular experiences (Inkelas et al., 2004). Because living-learning communities are
smaller in scope than residential colleges, positive gains found in these communities are
also likely to be found in residential colleges; thus, it is important to include the empirical
research on the outcomes of living-learning communities to see if it may elucidate the
potential influence of these communities on students’ inclination to inquire and capacity

for lifelong learning.

18

One approach researchers have taken in examining the influence of living
learning community environments on student outcomes is to group students from varying
living-learning communities (typically on a single campus) together and compare them to
peers in traditional residence halls (e.g., Pasque & Murphy, 2005; Pike, 1999; Pike,
Schroeder, & Berry, 1997). Pike, Schroeder, and Berry (1997) examined the impact of
one campus’s multiple living-learning communities on student persistence. They found
that residing in living-learning communities had a significant direct positive effect on
faculty-student interaction, social integration, and institutional commitment; however
there was no direct or indirect effect of the living-learning community environments on
students’ academic achievement, nor did living-learning community students’ reports of
greater institutional commitment and social integration contribute significantly to their
persistence.

In light of Pike, et al.’s (1997) findings, Pasque and Murphy (2005) conducted a
single institution study (at a different institution) examining whether participation in
living-learning communities predicted academic achievement and intellectual
engagement. They used linear regression analysis to ascertain whether there were
differences in intellectual engagement and self reported GPA between living-learning
community and traditional residence hall students, and found that living-learning
community participation was a small, but significant predictor of both intellectual
engagement and academic achievement. Several variables in the intellectual engagement
factor of Pasque and Murphy’s (2005) study, including ‘eager to try new experiences’
and ‘enjoys courses that are intellectually challenging,’ allude to the possibility that

living-learning community environments may promote students’ inclination to inquire,

l9

however, because there are other variables included in the factor (for example, “discusses
ideas or concepts outside of class” and “challenged by learning”) it is difficult to decipher
the overall meaning of the findings for students’ inclination to inquire.

In addition to exploring the role of living-learning communities in student
persistence, Pike (1999) examined the differences in experiences and learning reported by
students in living-learning communities versus their non-living-learning community
peers, exploring Specifically whether the relationships between experiences and
educational outcomes were the same for both groups. Using survey data from the College
Student Experiences Questionnaire (CSEQ) coupled with institutional data, Pike
developed ten scales, which measured student involvement in 1) the arts, 2) clubs and
organizations, and 3) the residence halls; interaction with 4) faculty, and 5) peers; 6)
topics of conversation; integration of 7) course information, 8) information in
conversations; and gains in 9) general educatiOn and 10) intellectual development. Using
a one-way analysis of variance, Pike found that students in living-learning communities
had significantly higher levels of involvement, interaction, integration and gains than
their traditional residence hall peers, with the effects of involvement and interaction
being greater than those of integration and gains. Based upon these findings, Pike
concluded that living-learning communities “exert a positive direct effect on day-to-day
behavioral aspects of students’ college experiences and indirect effects on the integration
of information and gains in student learning and intellectual development” (p. 280).

Although these studies highlight some of the potential benefits of living-learning
communities on student outcomes including greater faculty-student interaction, social

integration, involvement, and institutional commitment, the inconsistency in findings

20

regarding the role of living-learning communities in promoting academic achievement
(Pasque & Murphy, 2005; Pike et al., 1997) illuminates some of the methodological
difficulties encountered when attempting to isolate the role of these environments in
promoting student outcomes. One such difficulty occurs when different living-learning
community environments on a single campus are grouped together without taking their
varying organizational elements (such as quality of faculty/ student interaction, degree of
challenge, and peer environments) into account, potentially mitigating or overstating the
impact of the different environments. Another problem with using participant/non-
participant designs to ascertain the influence of a living-learning community or
residential college environment is the reality that most students self-select into these
environments, and this motivation, if not examined and possibly accounted for, may
confound the actual influence of these environments. A final difficulty with Pike,
Schroeder, and Berry (1997) and Pike (1999) studies in particular is the fact that students
were surveyed during the 10th week of their first semester on campus, which may have
been an insufficient timeframe for the environment to affect student persistence,
interactions, integration, and commitment.

Recent literature on outcomes of living-learning communities and residential
colleges has broadened to include multi-campus studies, including the National Survey of
Living-Learning Programs (Inkelas et al., 2004) and others (e. g., Snider & Venable,
2000). In addition, researchers have delved deeper into examining differences among
different communities within a Single institution (Inkelas & Weisman, 2003; Stassen,
2003). Furthermore, several recent studies have examined specific aspects of living-

leaming community environments, including the nature of faculty-student interaction

21

(Cox & Orehovec, 2007), variation in alcohol consumption (Brower, Golde, & Allen,
2003) and student engagement (Zhao & Kuh, 2004).

The National Survey of Living-Learning Programs, a multi-year, multi-
institutional study fimded by the Association of College and University Housing
Officers—Intemational (ACUHO-I) was initiated in 2002 to examine the impact of living-
leaming programs on various student outcomes (Inkelas et al., 2004). Preliminary results
from the project indicated that students in living-learning communities were more likely
than non-living learning community peers to perceive a positive residence hall
environment, have positive peer interactions, report a smooth transition to college, and
persist. In addition, students in living-learning communities had greater gains in
academic achievement and participation in civic engagement. On the other hand, there
were no statistically significant differences between living-learning community
participants and non-living learning community peers in appreciation of diversity, self-
confidence, and cognitive development.

Again, however, like Single institution studies that pool various communities and
compare them to traditional residence halls, caution should be taken when interpreting
the results of this multi-institution study because the tremendous variation in the
structures of living-learning communities that participated in the study were not
accounted for, and may result in the effectiveness of some communities being overstated
and others understated. For example, of the 34 living-learning communities included in
the study that were at research universities, 73% offered no courses for credit, 50%
required no co-curricular activities, and 33% had no faculty involvement (Inkelas, et al.,

2004). These variations in structure and offerings among living-learning communities

22

make it difficult to discern what aspects of the communities are making a difference for
students, a point that will be discussed in more depth in chapter three.

Variation in living-learning community outcomes becomes even more apparent
by examining studies of different communities conducted on one campus. Inkelas and
Weisman (2003) examined student outcomes among participant in three types of living-
learning programs (honors, transition, and curricular) at one university in the Midwest.
They sought to determine whether students from different living-learning communities
were more involved in college activities and perceived their environments more
positively. Specifically, they explored whether students in different environments
exhibited more positive outcomes in their transition to college, their preferences for and
enjoyment of challenging academic pursuits, and their interest and openness to learning
new and different perspectives. The researchers also examined whether students in each
living-learning community demonstrated greater gains in the domain of their particular
community (i.e., honors, transition, curricular) than the other students did.

They found that students in living-learning communities were significantly more
involved in their residence hall environments and perceived them more positively than
their non-living-learning community peers. There were variations in students’ perceptions
of their environments based on their affiliation with a specific living-learning community.
For example, students in the honors living-learning community were more likely to report
enjoyment of academic pursuits than either of the other two communities or the control
group. Likewise, students in the transition-focused community reported a smoother
transition to campus than either of the other two living-learning communities and the

traditional residence hall group. Students in the curriculum-based programs were more

23

similar to the traditional residence hall students than to the students in the other two
communities, as they were less often engaged in out of class activities than students in the
other two living learning communities. These findings illustrate that the mission and
conteXt of the living-learning communities matter, as students in living-learning
communities with a specific focus indicated satisfaction or gains in that particular
domain. However, the study did not account for the variation within each community,
rather it grouped the students from each community together and then compared across
communities, ultimately disregarding individual variation.

In summary, the existing literature on living-learning communities sheds light on
some of the potential benefits students in these derive from participation, including
greater interaction with faculty and peers, higher academic achievement, persistence, and
overall satisfaction with their residence hall experience. However, less is known about
how interactions between students and their environments may lead to these outcomes.
Furthermore, no research exists that specifically examines whether living-learning
communities are successful in promoting outcomes that are congruent with a liberal arts
education, which many of them purport to emulate. The current study will begin to
address this gap in the literature, but before doing so, it is important to understand the
literature and research detailing the conditions that promote students’ inclination to
inquire and capacity for lifelong learning in order to know what aspects of residential

. college environments may foster them.

24

Conditions that Promote Students ’ Inclination to Inquire and Capacity for Lifelong
Learning

The inclination to inquire and capacity for lifelong learning are different but
related concepts. The first centers on students’ intrinsic tendency to engage in and enjoy
thinking, whereas the second focuses on the skills necessary for a student to act upon
their value for inquiry. Both are outcomes associated with a liberal arts education (Center
of Inquiry in the Liberal Arts at Wabash College Website, n.d) in that they encourage
students to build a foundation for continuous learning throughout their lifetime. This
section situates them in existing theoretical and empirical literature.
Examining Students ’ Inclination to Inquire

Psychologists and higher education researchers continue to be interested in
understanding and measuring the development of students’ inclination to inquire deeply
into their life and academic pursuits (Biggs, 1993; Cacioppo & Petty, 1982; Cohen,
Stotland, & Wolfe, 1955; Marton & Sttle, 1976, 1984). Drawing on the work of social
psychologists Cohen, Stotland, and Wolfe (1955), Cacioppo and Petty (1982) “sought to
identify differences among individuals in their tendency to engage in and enjoy thinking”
(p. 116), which they labeled the ‘need for cognition.’ The researchers developed a
measure of individuals’ need for cognition using a sample of faculty and factory workers
from Iowa City. Subsequently, they tested the measure with different populations,
examining the relationship between the need for cognition scale and cognitive style, test
anxiety, intelligence, social desirability, dogma, and affective reactions to completing a
cognitive task. They found that the need for cognition scale assessed one primary factor

that was gender neutral, different from but related to cognitive style, related to

25

intelligence, unrelated to test anxiety and social desirability, and predictive of enjoyment
of cognitive tasks (Cacioppo & Petty, 1982).

Since the development of the original need for cognition scale, Cacioppo, Petty,
and Kao (1984) developed and tested a Shortened version. In addition, Cacioppo, Petty,
F einstein and Jarvis (1996) conducted a meta analysis of over 100 studies that used the
need for cognition scale to examine individual differences. The researchers found that an
individual’s need for cognition is somewhat stable in the short term, but not invariant
because it can be developed or changed over time. Furthermore, one’s need for cognition
is “derived from past experience, buttressed by accessible memories and behavioral
histories, manifest in current experience, and influential in the acquisition or processing
of information relevant to dilemmas or problems” (Cacioppo, et al., 1996). Because of
the scale’s robust psychometric properties, and consideration of both dispositional and
situational influences, it has been deemed as a “valuable tool for measuring the cognitive
goals of a liberal arts education” (Brown & Rogers, 2005, 1] l), as it can be used to
measure “the extent to which cognitive activities are desirable and important” to students
(11 2)-

Empirical Studies Related to Students ’ Inclination to Inquire

Several higher education researchers have used the need for cognition scale
specifically to examine students’ inclination to inquire. As part of their study examining
the impact of liberal arts experiences on liberal arts outcomes, Seifert, et al., (2008)
explored the extent that a liberal arts experience, (which they defined as “an institutional
ethos that values student-student and student-faculty interaction within a supportive

environment characterized by high expectations for developing the intellectual arts” [p.

26

108]) predicts liberal arts outcomes, including students’ ‘inclination to inquire and
lifelong learning’ among others. Using a sample of 909 students from 4 institutional types
(research university, regional institution, liberal arts college, and community college),
they conducted ordinary least squares regression using a liberal arts experience variable
as an independent measure of students’ experiences. They used liberal arts outcomes,
including students’ inclination to inquire and lifelong learning as dependent variables. ’
The researchers used the aforementioned ‘need for cognition’ scale (alpha = .89) and the
‘positive attitudes toward literacy’ scale (alpha =.65), which measures students positive
attitudes towards reading and writing, as measures of their inclination to inquire and
lifelong learning.

Among their findings relevant to the current study was that the liberal arts
experience variable significantly changed the amount of explained variation in the
inclination to inquire and lifelong learning, as the R2 change in need for cognition was
.038, (s=.235) and the R2 change for positive attitude for literacy was .052 (p=.258). Also
relevant to the current study was the finding that the liberal arts experience variable
significantly changed the amount of explained variation in the liberal arts outcomes’
measures, a finding that will be discussed in more detail in the methods section. Although
this study advances the notion that liberal arts experiences do indeed influence the
development of students’ inclination to inquire, questions remain about whether these
environments can be fostered intentionally at research universities and also what aspects
of the environment are the most influential. Furthermore, the researchers acknowledged
that their results may have been confounded by students’ pro-college tendencies toward

liberal arts outcomes. In addition, the study did not differentiate between students’

27

enjoyment of thinking, as measured through the need for cognition scale, and their
capacity for lifelong learning, leaving questions about whether students possess the skills
necessary to continue learning throughout their lifetime.

Mayhew, Wolrriak, and Pascarella (2008) also used the need for cognition scale to
examine how educational practices influence students’ development. Interestingly, they
used the need for cognition scale as a proxy for ‘lifelong learning orientation,’ arguing
that the scale provides a measure of students’ intrinsic cognitive motivation, which they
deemed is a prerequisite for lifelong learning. They examined how curricular conditions
and educational practices impacted the development of lifelong learning orientations in
undergraduate students, specifically exploring how provisions of opportunities for

reflection, active learning, and perspectiVe taking, influenced the students’ lifelong
learning orientations in five different courses (including a philosophy, psychology,
service-leaming, intergroup dialogue, and introduction to sociology course). Their
longitudinal, comparative, research design used the short version Cacioppo and Petty’s
(l 984) need for cognition scale to gauge changes in the need for cognition over the
Course of a semester.

The researchers found that the relationship between need for cognition and course
enrollment was marginal, with enrollment in the Introduction to Sociology course having
the most positive impact on students’ change in the need for cognition. Classroom
Practices, on the other hand, significantly contributed to a change in the need for
cOgnition, with negative interactions with diverse peers hampering growth in the need for

Cognition most significantly, and positive interactions with diverse peers and instruction-

based educational practices significantly promoting growth.

28

In summary, these studies illustrate the potential influence that environments and
individual courses can have on students’ inclination to inquire deeply into their life and
academic pursuits. However, more research is needed to determine how the interaction
between students’ motivation and their environment influences their inclination to
inquire. Furthermore, a finer empirical distinction should be made between students’
inclination to inquire and their capacity for lifelong learning, as evidenced by the
literature, to which I now turn.

Historical Overview of Lifelong Learning

For the past half century, educatOrs, policy makers, and researchers have begun to
focus more acutely on the concept of lifelong learning, examining its philosophical
underpinnings and varying definitions (Dave, 1976; T uijnman & Bostrorrr, 2002), its
practical application (OECD, 1996), and the roles of families, schools, and society in
promoting it (Abukari, 2005; Chapman & Aspin, 1997; Cropley, 1978; International
Labour Organization, 2000; Knapper & Cropley, 2000; McCombs, 1991; Smith &
Spurling, 2001). Tuijnman and Bostrom (2002) traced the conceptual evolution of
lifelong learning from its roots in the 196OS when it was called ‘lifelong education,’
through the 197 OS and 1980s when it was termed ‘recurrent education,’ to contemporary
examinations, in which the label ‘lifelong learning’ emerged. The authors highlighted
how the differing names and definitions of lifelong learning reflected societal trends, with
‘lifelong education’ reflecting the growing need and demand for firrther education, and
‘recurrent education’ focusing the scope of inquiry to the relationship between work and

education, reflecting the economic difficulties of the 19805 (Tuijnman & Bostrom, 2002).

29

The contemporary notion of lifelong learning emerged in the 1990s, and is
broader and more holistic, having been defined by the UNESCO Institute for Education
(UTE) as “a process of individual learning and development across the life-span, from
cradle to grave - from learning in early childhood to learning in retirement” (OECD,
1996). The UIE’s definition refers to education in formal settings (such as schools,
universities, and adult education settings), and also informal settings (such as at home, at
work, and in the community). Although educators, researchers, and policy makers have
generally accepted this broader definition of lifelong learning, several have emphasized
that the learning undertaken must be intended or planned (Smith & Spurling, 1999),
rather than the “spontaneous, day-to-day learning of everyday life” (Knapper & Cropley,
2000, p. 11). Furthermore, these researchers have argued that lifelong learning has
specific goals, and these goals are the reason that the learning is undertaken (Knapper &
Cropley, 2000).

In addition to examining varying definitions and notions of lifelong learning,
researchers, educators, and policy makers have also explored the skills necessary for
lifelong learning to occur. Cropley (1977) identified several specific skills of lifelong
learners, including the ability to adapt previous learning to new tasks, to apply learning
practically, to use media, to employ various learning strategies to different situations, and
to think critically. More recent efforts at identifying skills that promote lifelong learning
have revealed similar findings. Tuijnman, Kirsch, and Wagner (1997) argued that the
skills necessary for lifelong learning include those dealing with abstract and symbolic
information (both verbal and numerical), analytical skills, problem-solving skills,

adaptation skills, and technical skills. Although their terminology is somewhat different,

30

many of the skills identified as necessary are similar the earlier ones posited by Cropley
(1977).

Researchers have also identified several psychological characteristics upon which
lifelong learning is built, including awareness of the importance of lifelong learning,
motivation to carry on the process of lifelong learning, belief in one’s ability to continue
learning, and an ability to identify one’s own learning needs (Cropley, 1977; Knapper &
Cropley, 2000; Tuijnman, Kirsch, & Wagner, 1997).

Existing research has also explored the environmental conditions that promote
lifelong learning. Chapman and Aspin (1997) examined the roles of schools, family, and
the community in promoting lifelong learning. They emphasized that- schools should
focus on providing a foundation for lifelong learning and cultivate students’ motivation
for it. McCombs (1991) drew more specifically upon motivation and other psychological
theory and argued that in order to promote lifelong learning, schools should infuse into
the curriculum meaning and relevance, supportive climates that are conducive to
cultivating personal relationships, a sense of control and personal choice in the learning
process, and mechanisms for helping student to understand and work with their
psychological tendencies. McCombs (1991) stressed the importance of examining
lifelong learning not from a deficit perspective, where motivation for lifelong learning is
a problem to be fixed, but to focus on the “inherent capabilities and higher levels of
functioning within all persons that can be accessed by them if they are placed in
supportive environments” (p. 122). Smith and Spurling (2001) echoed the perspective
advanced by McCombs, arguing that the “general ethos of an institution is hugely

important” to promoting lifelong leaming and underscoring the control that a

31

postsecondary institution has on aspects such as students’ peer networks, through the way
it “structures groups of students, and the effort it makes to influence student culture
toward mutual support and altruistic activity - by encouraging group learning rather than
individualistic, competitive leaming” (p. 81).

The existing philosophical and theoretical literature illustrates the evolution of
lifelong learning to a broad and holistic concept rooted in formal educational settings as
well as informal setting including the home, work, and community. The literature’s
emphasis on the skills and psychological characteristics of lifelong learners, coupled with
the conditions that promote lifelong learning provides a rationale for examining students’
inclination to inquire (a psychological characteristic) together with their capacity for
. lifelong learning (a skill set), within the context of different residential college settings.
More substantiation of the importance of studying as separate, yet conceptually related
constructs is illustrated in recent empirical work.

Recent Empirical Work on Lifelong Learning

Perhaps as a result of the belief that postsecondary education increasingly plays a
vital role in fostering the skills necessary for competing in the 21”1 century knowledge
economy, there has been an increase in the number of empirical studies exploring
different facets of lifelong learning, including what individual characteristics predict
participation in lifelong learning (Gorard & Selwyn, 2005), how students’ capacity for
lifelong learning has changed in the face of the 21”t century knowledge economy (Hayek
& Kuh, 1998), and what environmental factors influence the development of varying

lifelong learning orientations (Hayek & Kuh, 1999).

32

Gorard and Selwyn (2005) examined adults’ patterns of participation in lifelong
learning in the United Kingdom. Using a survey designed to obtain adults’ varying forms
of participation in lifelong learning and demographic information (including age, sex,
occupation, nature of schooling, employment status, family background, and access to
technology) the researchers sampled 1001 adults living in several different electoral
wards throughout England and Wales. They found that many of the predictors of lifelong
learning are categorized as “birth related,” including age (with younger adults
participating more than older adults), educational level and occupation of parents, and
geographic mobility (with those living further from their birthplace participating more).
They also found that adults who undertook formal education in their adult years were also
those who reported using intemet and computer technology. This research sheds light on
some of the potential demographic characteristics that may influence one’s capacity for
lifelong learning. However, questions remain as to whether the findings are relevant to
students enrolled in postsecondary institutions in the United States.

Perhaps more relevant to the current study because of their focus on college
students, Hayek and Kuh (1998, 1999) conducted two studies exploring the capacity for
lifelong learning in college seniors. The first study compared two cohorts of seniors, one
from the 1980s (1984-1987) and the second fiom the 1990s (1994-1997), and examined
the extent to which they developed the capacity for lifelong learning throughout college,
whether the capacity for lifelong learning remained stable across the time periods even as
the need for workers to participate in the knowledge economy increased, and which types
of institutions better prepared students in developing the capacity for lifelong learning

(Hayek & Kuh, 1998). Using data from 26,629 completers of the College Student

33

Experience Questionnaire (CSEQ) from doctoral granting universities, comprehensive
colleges and universities, selective liberal arts colleges, and general liberal arts colleges,
the researchers examined students’ scores on the “Capacity for Life-Long Learning
Index, “a measure created by summing students’ responses to eleven Estimate of Gain
items,” which included variables measuring the extent to which students believed they
made progress in such areas as ‘general education,’ ‘synthesizing information,’
‘analyzing quantitative problems,’ ‘writing,’ ‘getting along with others,’ and ‘learning on
one’s own’ (Hayek & Kuh, 1999, p. 5) The researchers found that students’ capacity for
lifelong learning remained stable or increased in the 19903 largely as a result of seniors’
self-reported gains in their ability to use computers. After the ‘computers’ measure was
removed, students’ capacity for lifelong learning actually decreased from the 19803 to the
19903, with significant decreases occurring in seniors’ self-reported abilities in general
education, analyzing quantitative problerrrs, getting along with others, understanding
scientific and technological developments, synthesizing information, and leaming on
their own. Despite these decreases, seniors reported significant increases in their ability to
write and function as a team member. Hayek and Kuh also reported that students within
Selective Liberal Arts institutions reported the highest capacity for lifelong learning in
the 19803 and 19903, and with statistically significant increases occurring fi'om the 19803
to the 19903. The researchers acknowledged that their study failed to take into account
students’ motivations, which may have influenced their gain in their capacity for lifelong
learning. Furthermore, they argued that more research is needed in order to determine
how learning communities and other interventions promote the acquisition of lifelong

learning skills.

34

Hayek and Kuh’s second study (1999) examined how college activities and
environmental factors influenced undergraduate seniors’ development of lifelong learning
capacities. Using data from 17,541 CSEQ completers from 106 institutions (including
doctoral granting, comprehensive, selective liberal arts, and general liberal arts) the
researchers examined the areas that students with the greatest gains in the capacity for
lifelong learning devoted their time and energy to, and the characteristics of students and
institutions that are associated with above average gains in the skill areas comprising the
capacity for lifelong learning.

Using multiple regression and correlation analyses (with separate models run for
gender, race, and socioeconomic status), coupled with calculating effect sizes, the
researchers found that students’ major field resulted in medium to large effect sizes, with
majors including engineering, physical sciences, and biological science scoring high, and
arts, humanities, education, and foreign language scoring low. Furthermore, they found
that certain clusters of activities and environmental factors influence students’ capacity
for lifelong learning, including students’ overall satisfaction with college, the amount of
effort they devote to classroom activities, the amount of effort they devote to science and
technology, and an institutional environment that values critical, evaluative, and
analytical performance. Surprisingly, the researchers found that students’ participation in
formal extra-curricular activities and student-faculty interaction outside of class had low
effect to no effect on students’ capacity for lifelong learning .

These two studies are relevant to the currently study because they illuminate some
environmental factors that influence students’ capacity for lifelong learning. Specifically,
the finding that environments characterized as valuing critical, evaluative, and analytical

35

performance are most efi‘ective in promoting students’ capacity for lifelong learning
provides credence to the notion that a residential college, with its emphasis on providing
a small, liberal arts education with the resources of a research university, may be
efl‘ective in promoting students’ capacity for lifelong learning. However, questions
remainastowhetherresearchuniversifiescanbesuccessfidinemulafingthis small
liberal arts atmosphere. Furthermore, more research is needed to understand the
intersection of students and these environments, and specifically whether these outcomes
are a result of the environment itself, or if they are successful in attracting a student
already motivated toward developing the capacity for lifelong learning.

In summary, the inclination to inquire and capacity for lifelong learning are
diflemnt but related concepts that convey students’ willingness and ability to engage in
learning throughout their lifetime. The development of these capabilities is dependent on
the interaction of individuals and their environments. More research is needed to
understand how postsecondary institutions can promote conditions that encourage the
development of students’ inclination to inquire and capacity for lifelong learning,
especially at large research universities, which face the dual challenge of size and scale
with a large undergraduate student body and many (often competing) missions. Of
additional importance is research that accounts for students’ self-selection into these
environments, examining how their motivational attributes interact with their
environments and whether these environments are effective in influencing students with
varying motivation. Consequently, it is important to examine relevant motivation theory
to get a sense of how students’ motivational attributes might influence students’

inclination to inquire and capacity for lifelong learning.

36

Relevant Motivation Theory

Among the shortcomings of existing literature examining the environmental
influences affecting development of students’ inclination to inquire and capacity for
lifelong learning (e. g. Hayek & Kuh, 1998; Mayhew et. al., 2008; Seifert, et al., 2008) is
its failure to account for the varying motivations students possess in these environments.
This oversight has resulted in enduring questions as to whether the positive outcomes of
these contexts (which include selective liberal arts colleges, courses that promote active
learning, etc.) are attributable to their design, or rather are a result of the type of student
attractedtothe context. lnordertoclarifythe influenceofindividual studentsandtheir
environments, it is important to take students’ motivation into account.

The body of research on individual motivation is vast. Many motivational
theorists conceptualize individual motivation as a product of expectancy and value
reasoning within the broader social milieu (Brophy, 2004; Svinicki, 2004; Wigfield &
Bccles, 2000). The expectancy component of the motivation equation explores students’
beliefs about their ability to perform a task and the rewards for doing so; whereas the
value reasoning component explores students’ beliefs regarding the overall worth of the
process and the reward (Brophy, 2004). The social milieu is the context in which the
motivation occurs, which might be in a residential college, classroom, or work setting.

Perhaps most relevant to the current study, self—determination theory (SDT) is a
motivation theory incorporating aspects of both the value and expectancy portions of the
equation and planting them squarely in an environmental context. With roots in the
Aristotelian view of human development, which posited that people have an “active

tendency toward psychological growth and integration,” SDT embraces the assumption

37

that people have innate, natural, and constructive tendencies to develop a unified and
elaborated sense of self (Deci & Ryan, 2002, p. 3). Despite this integrative tendency,
however, SDT also acknowledges “that there are clear and specifiable social-contextual
factors that support this innate tendency, and that there are other specifiable factors that
thwart or hinder this fundamental process of human nature” (p. 5). As a result, SDT
predicts differing developmental outcomes based upon an individual’s perceptions of
social-environmental conditions.

Self-determination theory hypothesizes that among the social-environmental ‘
conditions affecting developmental outcomes are three basic or ftmdamental
psychological needs that must be met in order for development to occur. These needs
include autonomy, described as a sense of choice or control over one’s actions;
competence, described as a positive feedback mechanism that signifies efficacy and
improvement; and relatedness, described as “a secure relational base [which] provides a
needed backdrop” for the growth of people’s personalities and cognitive structm‘es (Deci
& Ryan, 2000, p. 235). A sense of well being results from a person’s basic needs for
autonomy, competence, and relatedness being met. Moreover, these basic needs serve as
a fomdafion for supporting an individual’s internalimtion of motivation (Deci & Ryan,
2002). As a consequence, the more fulfilled students feel in having these needs met in a
specific context, the more intrinsically motivated they will be. Self-determination theory
is relevant to the cmrent study because it provides a mechanism for assessing students’

motivation within a residential college setting, which existing studies have failed to do.

38

Existing Studies T ala'ng an Ecological Approach

The final aspect of literattu'c important to situating the current study in existing
scholarship is an examination of studies that take an ecological approach to
tmderstanding student development. As I discussed in the theoretical fi'amework,
environmental considerations continue to play an ancillary role in much of the college
student development literature, as many researchers continue to isolate outcome variables
instead of acknowledging the role of environments and students interactions within their
environments (Herzog, 2007; Kuh, 1995; Renn, 2004; Rem & Arnold, 2003). As
illustrated bythe firstthree sections ofthe literaturereview, inordertoexaminetherole
of residential college environments in promoting liberal arts outcomes like students’
inclination to inquire and capacity for lifelong learning, it is essential to take students’
sociodemographic and motivation attributes into account in addition to their
environmental contexts. Existing literatme approaching student development from an
ecological perspective accounts for the confluence of factors that ultimately affect
development.

Early attempts to incorporate enviromnental considerations into student
development typically centered on students’ interactions with their physical environments
(e.g., Banning, 1989), or were focused on assessing the appropriate fit between students
and a single environmental context like a residence hall or specific resource omce (e.g.,
Aulepp & Delworth, 1976; Schuh, 1990; Schuh & Veltman, 1991). Strange and Banning
(2001) provided a comprehensive summary of the amalgamation of early environmental
approaches to student development in their book, Educating by Design: Creating Campus

Learning Environments that Work. Written to provide a framework for administrators and

39

student affairs professionals to build supportive learning communities, the book drew
heavily from Moos’ theoretical perspective and examined campus environments,
including the physical setting, demographic and psychosocial fit for students,
organizatioml design, and social construction.

Recent empirical literature has used Urie Bronfenbrenner’s human ecology model
to examine the influence of peer and institutional culture in student development (c. g.
Arnold, 1995; Guardia & Evans, 2008; Johnson, 2005; Renn 2003, 2004). Specifically, in
the realm of race, ethnicity, and collegiate experiences, researchers have examined the
roles of fratemity membership on ethnic identity development of Latino men (Guardia &
Evans, 2008), the racial consciousness of White men attending a historically Black
University (Peterson, 2006), and the experiences of low-income Latina women as they
sought financial aid for college (V enegas, 2005). These studies found the Bronfenbrenner
model useful in accounting for the myriad influences on college student development,
which included familial and cultural norms in addition to institutional and personal
factors.

Drawing upon the studies each conducted earlier, Renn and Arnold (2003)
advocated for researchers to consider taking an ecological perspective to understanding
peer cultm'e, because

While much is said about “peer pressure” and the influence of peer attitudes on a

number of undesirable behaviors (binge drinking, sexual harassment and assault,

incivility, cheating, etc.), there is surprisingly little recent research linking peer

culture with college student behavior and outcomes. (p. 263)

40

The authors explained that the ecology model “takes into account the specificity of
individual life history, the campus milieu, and the larger societal and historical context of
development” and thus is “particularly well suited to understanding how individuals and
their proximal environments interact to shape identities of individuals and of groups” (p.
273).
Summary of Literature Review

Existing literature on residential colleges, learning outcomes, motivation, and
campus ecology illustrates both the potential of residential college environments in
fostering students’ inclination to inquire and capacity for lifelong learning and the need
for additional research on their effectiveness in doing so. With the number of residential
colleges at large research universities increasing as administrators work to integrate and
cohere undergraduate education, questions about the whether these communities promote
the outcomes associated with the liberal arts education they try to emulate have barely
been examined. Furthermore, existing approaches to researching outcomes of these
communities continue to fall short because often researchers aggregate the data they
collect to the environmental level, disregarding the individual differences of students in
these environments and ignoring the varying motivations students bring to these
environments. By following an ecological approach that better accounts for
environmental contexts and motivational attributes, the current study examined whether
residential colleges do indeed foster‘the development of students’ inclination to inquire
and capacity for lifelong learning. In chapter three, I explain the research design,

detailingthemethodslusedtocarryoutthestudy.

41

CHAPTER THREE
Study Design

This chapter describes how I conducted the study. After reintroducing the
research questions, I discuss my research design, sampling, and instrumentation. Then, I
explain my analytic approach and detail how I addressed each of the research questions. I
conclude by detailing the limitations of my data collection and analytic procedures.

The goal of the study was to examine how students’ motivation and
environmental contexts facilitate or impede students’ inclination to inquire and capacity
for lifelong learning in residential college settings. Using data collected from over 1800
students in 24 residential colleges, I investigated the following questions:

1. Does students’ inclination to inquire or capacity for lifelong learning vary

across residential college environments?

2. How are students’ sociodemographic characteristics and motivation related to
their inclination to inquire and capacity for lifelong learning?

3. Do the associations between students’ sociodemographic attributes and
motivation and their inclination to inquire and capacity for lifelong learning
vary across residential colleges?

4. How is the environmental context of a residential college related to students’
inclination to inquire and capacity for lifelong learning? Specifically,

a. Environmental Liberal Arts Experience variables - (good teaching and
high qudity interaction with faculty, academic challenge and high
expectations, diversity experiences, and peer interactions)

b. Disciplinary Focus

42

5. Does the association between student-level motivation and students’
inclination to inquire and capacity for lifelong differ by residential college-
level liberal arts experience variables (good teaching and high quality
interaction with faculty, academic challenge and high expectations, diversity
experiences, and peer interactions)?

Research Design

I adopted a quantitative, cross~sectional survey design to conduct the study. A
quantitative design is appropriate when “examining the relationships between and among
variables is central to answering questions” (Creswell, 2003, p. 153). The survey
approach enabled me to identify students’ inclination to inquire and their capacity for
lifelong learning in a numerical format (Creswell, 2003). The design was cross-sectional
because data were gathered from students across different class years (Creswell, 2003).

By using an ecological perspective for data collection and analysis, I examined
the influence of both personal and environmental characteristics on students’ inclination
to inquire and capacity for lifelong learning. The data I collected were hierarchical in
nature, with lower-level observations (i.e., students) nested within environments (i.e.,
residential college settings) (Kreft & De Leeuw, 1998). Because the study explored the
influence of both individual (level one) and organizational (level two) characteristics on
individual-level outcomes, I employed hierarchical linear modeling (HLM) to analyze the
data I collected.
Population, Sample, and Partich

The population of students to whom the study was intended to generalize included

all students in residential colleges located within large public research universities.

43

According to the ‘Basic Classification Description’ of the Carnegie Classification of
Institutions of Higher Education, at the time of the study there were 136 public
institutions classified as “Research Universities (very high or high research activity)” or
“Doctoral Research Universities” that were also classified as “Large (including primarily
nonresidential, primarily residential, and highly residential),” defined as those with Full-
Time Equivalent enrollment of at least 10,000 degree-seeking students. By examining the
websites of these 136 universities, and typing, ‘learning community,’ “living-learning
community” and “residential college” separately into the search engine and then
examining the websites to ascertain whether they met my definition of a residential
college, I determined there were 32 degree-granting (major or minor) residential colleges
located within 11 of the large, public research universities in my sampling fiame.
Although all of the residential colleges in my sampling frame were degree granting and
offered students a shared academic experience, variations remain in terms of their
admission requirements and organizational structures (e.g., faculty-to—student ratio, how
long students are required to live with the residential college, and whether the faculty had
their tenure homes in the college).

After identifying the 32 residential colleges appropriate for inclusion in the study,
I contacted the deans of 31 of them and invited them to encourage their students to
participate in the study. (Because I conducted a pilot of my survey at one residential
college which had been established only 2 years prior, I excluded it from my sampling
fiame.) Of the 31 residential colleges (located within 11 universities) contacted, 30
expressed initial interest in participating in the study. One university (with six residential

colleges embedded in it) dropped out of the study before sending information out to

students. In many ways, the university that dropped out of the study was similar to other
universities in the sample, as it was a large, public research university. However, it was
the only university in that region of the United States that ofi‘ered multiple residential
colleges on a single campus, and therefore may have attracted a different type of student
than the other residential colleges fi'om that region.

I worked in partnership with deans and directors of the remaining 24 residential
colleges at 10 universities to distribute information to students in a way that they deemed
most appropriate. Most of the administrators with whom I partnered believed that the best
way to distribute information about the survey was for a representative of the residential
college to send the information about the study and a link to the survey out via email or
listserv. Others preferred to send the information out in an on-line, weekly newsletter sent
to students. Three institutions provided me with contact information for students and
asked that I send information to them directly; of those three, two gave me contact
information for all the students, and one selected students whom they described as
‘student leaders.’ The different modes of contacting students led to some variation in
response rate, with the information being sent via newsletter leading to a smaller than
average response rate, and the information and follow-up being sent by me leading to
larger than average response rates. These alternative modes of contacting students did not
lead to any anecdotal differences in the representation of the samples.

As an incentive for completing the survey, I offered students the opportunity to
receive one of five $100 gift cards to Amazon.com. The gift card information was in no

way tied to students’ responses, as upon completion of the survey, a link took them to a

45

separate website where they provided their contact information for entry into the
drawing.

Administrators’ willingness to send follow—up reminders to students urging them
to take the survey varied by residential college. Some sent one or two follow up emails to
students, whereas others expressed concern that they not burden the email in-boxes of
their students, or cause ‘survey fatigue,’ a word used by one administrator to describe the
constant pleas they sent to students asking them to complete various institutional and
national surveys. Also important to note because of its potential influence on survey
response rate was that the IRB of one institution asked that the gift card incentive for
participation not be used as it could be construed as a lottery and therefore was against
state law and university policy. Given these constraints, I followed Umbach’s (2004)
recommendations for improving the response rate of a web-based survey to the greatest
extent possible, as I or an administrator sent an initial email inviting students to take the
survey followed by several reminders, kept the survey relatively short, included a
deadline for response to the survey and an estimated time to allot for completion, and
ofl‘ered an incentive for completion. Information about the sample sizes and procedures is
included in table 3.1.

There were 2,202 students who responded to the survey with some information.
Of these students, 391 did not complete at least 20% of the survey, did not indicate their
affiliation to a particular residential college, or did not answer all the questions to at least
one of the outcome variables and were subsequently excluded from the analysis. The
final analytic sample was comprised of 1811 students affiliated with 24 residential

colleges at 10 large, public research universities throughout the United States

 

 

cousmsoe e83

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

a; E: a a 3:3: as, a as 8:853 seam emcee as:
cogmga

so \ a... 3 NS 3:2 2 E as Sensuous straw emcee $52
83338?

as \ an a: E 838% 3 as 8:853 355 seem Ens:
toefiasas

as \ a» «S me 2:35 B as usages stem cease seem SEE
negates

.xa \ as a: on. «3:81 3 ans 85:85 3:5 emu—.8 omens
assesses

so \ as we 8 2982 3 as segues seam emcee uses: spa.
835583

a: 3.3 a am 8:02 3 use usages seam amaze Bananas Essa
gasses

an \ so :. 3 «ENE B as Suwanee 3:5 09:8 .520
easiest

an \ so 5 a 8382 E as manages seam emcee €38
scanning;

s: 3:: on 3 «RES 3 as nonsense seam amaze 3:an a 28
83538?

as. \ a.” 8 a: $252 .3 as confines 53 a: 03:8
aegis“; ox sensuous as

as \ an em 3 ES: 3 as gaseous seam 8:8538 a 09:8
assesses

an \ so me 8 335: B as guesses spam 2:2 09:8
sausages

.xa \ as 2 a m: 3232 3 as Senses 555 29m 09:8
canines. ease...

as: 3.8 3. on @3an B as assumes spam sea—Ewes: ESE

use—~58 mailer. anal—.8 E con—E:— in

as: 38 £3238 :3 .3». cause: .59: 335:8

as. 8:85. E .32.. a as: .38. 2.2.5:: sz 8...»... assets... .3th _ «mo—.8 .6538.—

 

 

aoeaaeea 85.3% as 038$ ”2 23.

47

5.88... .5 ass... as. u... a Bass swag. Base... 22.. a... as... Swans... 355...
o: .23 689% 8 been 2: we flame 05 8883 damage? mum—9:8 o5 unaccep— 3 amazon. igua— aoeome can: in.— ofi 9.: 8 gee _ 3.553..

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

an \ so . :2 menu aamogm

.25 woman—e MEX anemone
$8 \ sum 2 a. amen :85 as sauce. a. 883:8 . Sac... 03:8 Ease
.x... . tan. 8 S 23% =2... «.5 saves a. 303:8 . 352.3. esnwawhwocw
s. w .38. m. e. .832 as 38.5. ass 38...”... .uses 55.31
s... \ a... m: .2 2.2.2.: 3 as aofiwaaM%m suzoo 822%
.xa. \ an S. on. .2 Q8. m E as aofihuofimmaem 03.30%
.x... :a. E am $58: .3 as aofihcommfinfi .330 2.33.
:2 E... 8 ab. 8.33 .3 as 8%Nfi o: 2P
so \ so 3 mm. 3.232 .3 as cogfifi owe—.8 .85..
so \ .x... 8 8. E25: .3 as éfifi 09:8 go

 

ease 3 one.

48

The response rate for the final analytic sample was 5%.

The analytic sample approximates the national sample of Hispanic or Latino
students, American Indian or Alaskan Native students, Native Hawaiian students, and
multiracial students attending large public research miversities (NCES, 2004). However,
Black or Afiican American students are underrepresented by about 8%, White students
are underrepresented by 10%, and Asian students are overrepresented by about 13% in
comparison to national averages of students attending large public research institutions
(NCES, 2004).In addition, women are overrepresented in the analytic sample by about
20%. I provide complete descriptive information, including mean, standard deviation,
range of responses, and the number of responses for each particular question or scale in
the analytic sample in table 3.2. Appendix B provides additional information on all the
variables in the analysis not provided in table 3.2.

Table 3.2: Descriptive Information of Analytic Sample

RESIDENTIAL COLLEGE STUDENT CHARACTERISTICS (Level 1)

 

VARIABLE NAME MEAN SD MIN. MAX N
Sociodemographic variables
Afiican-American 0.02 0.13 0 1 181 l
Asian-American 0.21 0.41 0 1 181 1
Hispanic/Latino 0.07 0.26 0 1 181 1
White (reference group) 0.56 0.50 0 l 1811
International 0.02 0.14 0 l 1811
Native Am. or Alaskan Native 0 0.05 0 l 1811
Native Hawaiian or other Pac. Is. 0.01 0.08 0 1 1811
Multiracial 0.1 0.3 0 1 181 1
No response forrace 0.01 0.1 0 1 1811.
Male 0.33 0.47 0 l 1809
Family Income 350-1 10]: (reference
grow) 0.43 0.49 o 1 1740
Family income Below $50K 0.26 0.44 0 1 1740
Family Income Above $110K 0.31 0.46 0 1 1740
HS GPA 3.63 0.32 1.7 4 1749
Combined ACT Score 27.72 3.98 11 36 710
Combined SAT Score 1258.28 155.14 650 1600 1120
First generation 0.24 0.42 0 l 1713

49

Table 3.2 (cont’d).

 

 

VARIABLE NAME MEAN SD MIN. MAX N
College Experience and
Motivation Variables
Years lived in the Res. College 1.45 0.73 0 4 1774
Degree Asp. - Less than bach. 0.01 0.12 0 l 1802
Degree Asp. — bach. (reference
group) 0.15 0.36 0 1 1802
Degree Asp-More than bach , 0.83 0.37 0 1802
Motivation 89.87 13.45 42 1 19 1737
Individual Liberal Arts Variables
(centered and entered at level 1)
Good Teaching and High Quality
Interactions with faculty
- Classroom Practices ' 42.42 6.42 11 55 1669
- Out of Class Interactions 17.03 4.03 5 25 1689
Acad. Challenge & Expectations 72.29 12.57 32 105 1531
Diversity Experiences 25.24 6.16 8 40 1606
Interaction with Peers 22.78 5.38 6 30 1628
Student Outcome Variables
Capacity for Lifelong Learning 29.71 6.58 l l 44 1769
Inclination to Inquire 54.45 8.27 18 75 1737

 

RESIDENTIAL COLLEGE ENVIRONMENT CHARACTERISTICS (Level 2)
Environmental Variables

(group mean entered at level 2)
Good Teaching & Quality
Interactions with Faculty
- Classroom Practices 42.45 1.70 40.16 46.93 24
- Out of Class Interactions 17.25 1.48 15.23 21.53 24
Acad. Challenge & Expectations 72.74 6.63 64.06 91.76 24
Diversity Experiences 25.29 2.00 21.53 31.18 24
Interactions with Peers 22.90 1.53 19.18 27.12 24
Discipline - Soc. Sci. 0.04 0.2 0 l 24
Discipline - Nat. Sci. 0.08 0.28 0 l 24
Discipline -
Interdiscp. (reference grorp) 0.83 0.38 0 1 24
Discipline - Humanities 0.04 0.2 0 l 24

 

De Leeuw, Hox, and Dillman (2008) outlined four cornerstones of data collection,
which include sampling (selecting a subset of the population), coverage (creating a

sampling frame that is representative of the population), response (obtaining responses to

50

questions from all participants in a sample), and measurement (ensuring that the
questionnaire measures what it intends to measure). The authors explained that three of
the four cornerstones (including sampling, coverage, and response) rest on the notion that
surveys collect data from a sample, or a fi'action of the population of interest. They
stressed the importance of minimizing sampling errors by “sampling enough randomly
selected units to achieve the precision that is needed,” coverage errors by ensuring that
“every member of the population has a known and nonzero chance of being selected into
the survey,” and nonresponse errors by ensuring that “everyone responds or if the
respondents are just like nonrespondents in terms of the things we are trying to measure”
(de Leeuw, Hox, and Dillman, 2008, p. 7).

The primary data collection approach for the current study was census, as in most
cases every member of the residential college was asked to participate in the study. This
approach reduced the possibility of sampling and coverage errors. However, because not
all deans and directors of residential colleges granted access to their students, there is still
some coverage error, as 7 of the 32 residential colleges located in large, public, research
universities in the United States are not represented in the sample. Furthermore, one dean
limited the sample of students included to a handful of student leaders.

Admittedly the largest source of data collection error in the current study is as a
result of nonresponse. Because of the census approach taken, and the limited
opportunities to send follow-up reminders encouraging students to complete the survey,
response rates are low. However, the descriptive statistics of the sample indicate that
there is a normal distribution of data. Furthermore, because the chance to win a gift card

was used as an incentive to encourage participation, it is possible that some students who

51

optedtoparticipate inthe study didsotogainentry intothedrawing, while others wished
to provide feedback about their residential college environment. Despite these facts, the
possibility remains that there is bias in the sample.

Certainly, selecting a random sample of students and diligently following up with
them to ensure a high response rate would have been the most ideal data collection
method. Practically, this method proved impossible because of privacy concerns. Several
deans and directors of the residential colleges were unwilling or felt unable to share
contact information of the students in their respective residential college. As a result,
some caution should be used when generalizing the results of this study to the general
population of residential college students at large research 1miversities.

Data Collection Instruments, Variables, and Materials

The survey instrument for the study was developed using a compilation of several
existing surveys along with questions aimed at identifying demographic information of
students in the sample. Survey items included were meant to operationalize the individual
motivational attributes and environmental contexts that influence students’ inclination to
inquire and capacity for lifelong learning (see Appendix A for complete survey).

Independent variables. Predictor variables for the study included: (1) students’
pre-college characteristics (e. g., sociodemographic traits and academic preparation); (2)
students’ motivation; (3) academic discipline; and (4) liberal arts experiences (see
Appendix B for complete description of variables).

Pre-college chmtm “es. I collected sociodemographic information to
determine whether my sample was representative and also control for potentially

conformding information. Specifically, I collected information on students’ sex,

52

race/ethnicity, parental education and family socioeconomic status; class year, high
school GPA, pre-college test scores (ACT/ SAT), number of years living in the residential
college, and degree aspirations.

Mom 'on. Self-determination theory (Deci & Ryan, 1985, 2000, 2002), which
posits that when individuals satisfy their basic needs for autonomy, competence, and
relatedness they are more intrinsically motivated, served as a foundation for measuring
students’ motivation. Specifically, I used the General Need Satisfaction scale (Gagne,
2003), which was adapted from a measure of needs satisfaction at work (Ilardi, Leone,
Kasser, & Ryan, 1993), and measures the degree to which individuals feel that their
choices and activities are selfidetermined (autonomy), the degree to which individuals
feel a sense of eficacy in their activities (competence), and the degree to which
individuals feel a sense of connectedness to others (relatedness). Gagné (2003), who used
the scale to predict the extent to which satisfaction of basic needs will predict
engagement in prosocial behavior with a sample of 121 college students, reported the
following Cronbach alpha values for the subscales: autonomy (7 items) = .69;
competence (6 items) = .71; and relatedness (8 items) = .86, and an overall alpha of .89.
The correlations among the subscales were between .61 and .66 (Gagne, 2003). Questions
on the General Need Satisfaction scale included item such as, “I generally feel free to
express my ideas and opinions,” “Most days I feel a sense of accomplishment from what
I do,” and “People in my life care about me.”

To ensure the General Need Satisfaction scale measured specific constructs that
were consistent with Self-Determination Theory (Deci & Ryan, 1985, 2000, 2002) for my

sample, 1 factor analyzed the variables associated with each subscale. I expected one

53

construct to emerge from the analysis and so used quartimax rotation in order to examine
how highly the variables from the scale loaded on to that dominant factor.

I began by conducting a principle components factor analysis of the 7-item
Autonomy subscale of the General Need Satisfaction scale. I found that one factor
explained 38.07% of the variance. However, two of the factor loadings on the rotated
matrix were low (-.05 for “I feel pressured by life” [recoded] and .042 for “In my daily
life, I generally do what I am told” [recoded]) and the reliability analysis indicated that
the reliability would improve (from .706 to .721) or stay virtually the same (.706 to .697)
with each of the items removed. I deleted these in turn and formd that for the S-item
subscale, 50.72% of the variance was explained by the dominant factor. Factor loadings
for the 5-item scale ranged from .818 to .541. Reliability of the 5-item scale was a = .75,
compared to a = .70 for the 7-item scale. As a result of the improvement in construct
validity, I opted to delete the two variables, and keep the 5-item scale. The scale’s
constituent items are found in appendix C.

I then turned my attention to the Competence subscale of the General Need
Satisfaction scale. Again I conducted a principle components factor analysis with
quartimax rotation of the 6-item scale. I found tint one factor explained 46.63% of the
variance. However, two of the factor loadings on'the rotated matrix were low (.087 for “I
have been able to learn interesting new skills recently,” and .129 for “People I know tell
me I am good at what I do”). After deleting the first of the two low items, the dominant
factor explained 49.69% of the variance and factor loadings improved, ranging from .832

to .536 with a reliability of the 5-item scale of a = .745 in comparison to a = .767 of the

54

6-item scale. Because of the improved factor loadings, I opted to use the S-item scale.
The scale’s constituent items are found in appendix C.

Finally, I examined the Relatedness subscale of the General Need Satisfaction
scale, factor analyzing the 8-item scale using quartimax rotation. I found that one factor
explained 50.45% of the variance, Two of the factor loadings on the rotated matrix were
low (.294 for “I pretty much keep to myself and don’t have a lot of social contacts
(recoded)” and .304 for “There are not many people I am close to”). After deleting the
first of the two low items, the dominant factor explained 52.98% of the variance and
factor loadings improved, ranging fi'om .800 to .569. The reliability of the 7-item scale
was a = .828 in comparison to 0. == .842 for the 8-item scale. Because of the improved
construct validity, I opted to use the 7-item scale. The scale’s constituent items are found
in appendix C.

Because I was most interested in ascertaining how different aspects ofthe
environment influence students’ inclination to inquire and capacity for lifelong learning, I
used the overall motivation scale (which was comprised of the 3 subscales) in my
analysis. The reliability of the overall scale was a =.884.

In addition to individual independent variables (level one), I included several
environmental variables (level two) as independent variables.

Q'gipline. I included the disciplinary focus of the residential college in my
analysis because past studies (Hayek & Kuh, 1999) indicate that it has a bearing on
students’ capacity for lifelong learning, and specifically that students’ in science-oriented
environments report a higher capacity for lifelong learning. I dummy coded variables

representing the disciplinary focus of the residential college, including natural sciences,

55

social sciences, humanities, and other/interdisciplinary (reference). I chose to place this
variable at level two of the analysis because I was interested in examining the influence
of the cultural implications of an academic discipline within a residential college setting
as opposed to the individual, psychological reasons that one might choose one discipline
over another.

Liberal m Experiences= On the survey, I included several subscales of a liberal
arts experiences scale originally developed by Pascarella et al. (2005) and adopted by
Seifert et al. (2008) and Blaich and Wise (2008) that was designed to measme the
institutional practices and conditions that comprise the environment of effective liberal
arts colleges. These scales were used in the model at levels one and two, with the level-
one responses representing each individual students’ estimation of the liberal arts
experiences, and level two representing the mean of all students in the residential college
(so as to capture the overall ethos of the residential college). The liberal arts experience
scale included aspects of the environment that are

most closely aligned with the Center of Inquiry’s definitions of a liberal arts

education (Blaich et al. 2004): positive and influential student-faculty contact;

faculty interest in teaching and student development; instructional clarity,
organization, and preparation; academic effort and challenge; degree to which the
institution is supportive; positive influence of interactions and relationships with
peers; integration of ideas through class activities and assignments; challenging
classroom environment characterized by high expectations; instructor feedback to
students; emphasis on higher-order examinations and assignments; frequency of

engaging in cooperative learning activities; frequency of faculty contact;

56

frequency of student afl'airs contact; overall diversity experiences and

interactions; academically meaningful out-of-class experiences; involvement with

active learning; diversity courses; and out-of—class research with faculty member.

(Seifert et al., 2008)

Seifert et al. (2008) examined the extent to which the liberal arts experience variable
predicted liberal arts outcomes, including moral reasoning, efl‘ective reasoning and
problem solving, intercultrn‘al effectiveness, inclination to inquire and lifelong learning,
well-being, and leadership. They found that for four of the six outcomes (representing 18
out of 20 separate measures), including intercultural effectiveness, inclination to inquire
and lifelong learning, well-being, and leadership had a positive effect on the outcome
variables. Although the liberal arts experiences variable failed to predict moral reasoning
or effective reasoning and problem solving outcomes, for the purpose of the current
study, the nonsignificant finding for these two dimensions of the liberal arts experience is
irrelevant.

In their longitudinal study of liberal arts education, Blaich and Wise (2008)
reported that foursubscales of the liberal arts experiences scale were significant in
predicting changes in students’ need for cognition at liberal arts colleges. The specific
subscales they found to have a positive influence include,

Good Teaching and High Quality Interactions with Faculty. (alpha = .92) A
measure created by summing students’ answers to 23 questions about their experiences
with “faculty interest in teaching and student development, prompt feedback, quality of
nonclassroom interactions with faculty, [and] teaching clarity and organization” (Wabash

National Study of Liberal Arts Education, 2008, p. l).

57

Academic Challenge and High Expectations. (alpha = .88) A measure created by
summing students’ answers to 31 questions about their experiences with “academic
challenge and effort, frequency of higher-order exams and assignments, challenging
classes and high faculty expectations, [and]integrating ideas, information, and
experiences” (Wabash Natiorml Study of Liberal Arts Education, 2008, p. 1).

Diversity Experiences. (alpha = .80) A measure created by summing students’
answers to 9 questions about their “diversity experiences [and] meaningful discussions
with diverse peers (Wabash National Study of Liberal Arts Education, 2008, p. 1).

Interaction with Peers. (alpha == .85) A measure created by summing students’
answers to 9 questions about their extracurricular activities and the extent of positive
interactions with peers (Blaich, personal communication, September 4, 2008).

To ensure that the four subscales of the liberal arts experiences scale were strong
and measured the constructs I intended them to measure, I factor analyzed (using
principle components analysis) the variables associated with the scales and using these
results as a guide, revised the subscales to increase construct validity. Because I expected
one construct to emerge from each subscale I analyzed, I used quartimax rotation in each
analysis in order to examine how highly the variables fiom the scale loaded on to a
dominant factor.

I began by examining the results of the factor analysis of the 23-item ‘Good
Teaching and High Quality Interactions with Faculty’ scale from the Wabash National
Study of Liberal Arts Education. Instead of finding one dominant factor, I found two
factors explaining 40.19% and 11.63% of the variance respectively. In examining the

rotated component matrix, I noted that three variables, including, “frequency that you

58

received prompt oral or written from faculty on your academic performance,” “frequency
that faculty informed you of your performance in a timely manner,” and “frequency that
faculty checked to see if you learned the material well before going on to new material”
did not load well on either of the two factors. I deleted them and reran the analysis. Two
dominant factors emerged again, explaining 42.26 % and 13.30 % of the variance. I noted
that four variables loaded moderately well on both factors, including for example, “Most
faculty with whom I have contact are willing to spend time outside of class to discuss
issues of interest and importance to students,” “Most faculty with whom I have had
contact are interested in helping students grow in more than just academic areas,” and
“Most faculty with whom I have had contact are outstanding teachers.” I deleted these
variables and reran the analysis. In the new model, which had 16 items, two factors
explained 42.36% and 13.82% of the variance respectively. I examined the questions
associated with the factors and noted that the first factor dealt with classroom practices,
and the second dealt with out of class interactions with faculty. I conducted reliability
analyses for each of the emergent factors and found them to be reliable (a=.91 l and .844
respectively). I present the new constructed independent variables related to teaching and
their constituent items in table 3.3.

Table 3.3: New Scales for Good Teaching and High Quality Interactions with Faculty

 

Residential college students’ self-reports of ‘good Reliability Factor
teaching and high quality interactions widrfaculty’ Loading
Teaching - Classroom Practices 0 = .911
1. Frequency that faculty were well prepared for
class .823
2. Frequency that the presentation of material was
well organized. .783
3. Frequency that class time was used effectively. .781
4. Frequency that faculty had a good command of
what they were teaching. .781

59

Table 3.3 (cont’d).
Reliability Factor

Loading
5. Frequency that faculty made good use of ‘
examples and illustrations to explain difficult
points .741
6. Frequency that faculty gave clear explanations. .741
7. Frequency that course goals and requirements
were clearly explained. .730
8. Frequency that faculty interpreted abstract ideas
and theories clearly. .704
9. Frequency that faculty gave assignments that
helped in learning the course material. .660
10. Frequency that faculty effectively reviewed and
summarized the material. .655
l 1. Most faculty with whom I have had contact are
genuinely interested in teaching. .535

Teaching - Out of Class Interactions with Students on = .844

1. My non-classroom interactions with faculty have

had a positive influence on my intellectual

growth and interest in ideas. .839
2. My non-classroom interactions with faculty have

had a positive influence on my personal growth,

values, and attitudes. .839
3. My non-classroom interactions with faculty have

had a positive influence on my career goals and

aspirations. .806
4. Since entering the residential college, I have

developed a close, personal relationship with at

least one faculty member. .712
5. I am satisfied with the opportunities to meet and
interact informally with faculty. .640

I then turned my attention to the factor analysis results of the 30-item ‘Academic
Challenge and High Expectations’ scale. As expected, one dominant factor emerged that
explained 29.34% of the variance. However, upon examining the factor loadings on the
rotated matrix, I noted that two factor loadings were particularly low, with “In a typical
week, how many problem sets take you more than an hour to complete” loading at .045

and “During the current school year, about how often have you come to class about

60

completing readings or assignments (recoded)” loading at .015. The reliability analysis
also revealed that reliability of the construct would improve with the items deleted (from
.905 to .909 and .910) respectively. I deleted both and re-ran the analysis. Again one
dominant factor explained most of the variance, but several factor loadings remained low
and the reliability analysis indicated marginal improvement were items deleted. I deleted
variables in turn until I was satisfied with the factor loadings of the variables and
believed they were representative of ‘Academic Challenge and High Expectations.’ The
final scale contained 21 items, with one dominant factor that marked students’
assessment of their experience as challenging explaining 36.91% of the variance. The
factor loadings ranged from .718 to .387, with 20 of the 21 factor loadings above .4.
Reliability ofthe 21-item scale was a = .912, in comparison to a = .950 for the 30—item
scale. The scale’s constituent items are found in appendix D.

Next, I factor analyzed the 9-item ‘Diversity Experiences scale.’ Again, I opted to
use quartimax rotation because of my desire to ascertain how strongly the variables
loaded onto one dominant factor. The loadings of the variable, “How often do you attend
a debate or lecture on a cm'rent political/social issue?” were low (.177) and the reliability
analysis indicated only a marginal change in reliability were the item deleted (from .841
to .838) so I deleted the variable and reran the analysis. The result was one dominant
factor which explained 47.95% of the variance with factor loadings ranging from .824 to
.302. Six of the eight items loaded on the factor above .4. Two factor loadings were low
(.331 and .302 respectively), however, their associated variables dealt unequivocally with
diversity experiences, and the reliability analysis indicated that the alpha-level would

decline were the items deleted (fi'om .838 to .830 and .829 respectively). Because of these

61

reasons, I opted to retain them. The reliability of the scale was a = .838. The scale’s
constituent items are found in appendix E.

Finally, I factor analyzed the 9-item ‘Interactions with Peers’ scale. Again, I used
quartimax rotation. The results of the analysis indicated that the variable “Most students
in my residential college have values and attitudes different from my own (recoded)” did
not load well on the dominant factor. Furthermore, the reliability analysis indicated that
the reliability of the scale would improve substantially were the item not in the scale
(fiom .821 to .836). I deleted this variable fi'om the analysis and reran the factor analysis.
Two additional variables had low factor loadings, including “About how many hours a
week do you spend participating in co-curricular activities?” (.242) and “Few of the other
students I know would be willing to listen to me and help me ifI had a personal problem
(recoded ” (.203). The reliability analysis indicated improvement in the scale if each item
were deleted. I deleted these variables in turn. The resulting 6—item scale had one
dominant factor accounting for 71.24% of the variance with factor loadings ranging from
.911 to .623. Reliability of the 6-item scale was 0. == .910. The scale’s constituent items
are fomd in appendix F.

Dependent variables. The dependent variables of interest include students’
inclination to inquire and capacity for lifelong learning.

W. The inclination to inquire variable was operationalized
using Cacioppo, Petty, and Kao’s (1984) short form of the Need for Cognition Scale,
which was discussed in more depth in chapter two, and measures “an individual’s
tendency to engage in and enjoy efi‘ortful cognitive endeavors” (p. 306). The short form

contains 18 items, including items such as “I would prefer complex to simple problems,”

62

“I really enjoy a task that involves coming up with new solutions to problems,” and “I
prefer my life to be filled with puzzles I must solve” (p. 307). Some of the tests
administered to establish validity and reliability of the original longer form were
discussed in chapter two. Other tests administered included a correlation analyses
between individual items in the survey and the total score to measure internal
consistency, and a factor analysis, which determined that one factor was dominant in the
scale (Eigenvalue of 10.22, accounting for 30.1% of variance (Cacioppo & Petty, 1982)).
Cacioppo, Petty, and Kao (1984) ranked the 34 items form the original scale in terms of
the absolute value of their factor loadings in the experiment from which the original scale
was developed. Then, they calculated Cronbach’s alpha as each successive item was
addedandconductedaScreetesttodeterminethenumberofitemstoberetained. After
they determined the shortened form would contain 18 items, they correlated students’
scores on the full scale with the shortened scale (r=+.95, p<.001). In addition, they
performed one factor analysis of the students’ responses to the long form, and a second
factor analysis, using the same data, but only including the 18 items on the short form.
They found that the variance attributable to the factor in the short form was 37%, as
opposed to 27% in the long form (Cacioppo, Petty, & Kao, 1984). Furthermore, the
maximized Cronbach’s alpha coefficient was .90 for the 18-item scale, versus .91 for the
longer scale.

In order to ensure the construct validity of the Need for Cognition scale for the
current study, I conducted a factor analysis with the items from the short-form used on
the survey. I opted to use quartimax rotation because I believed that one construct would

explain most of the variance and I wanted to see how highly the variables from the scale

63

loaded on to that dominant factor. I found that there was one dominant factor that
explained 33.9% of the variance. However, upon examining the results of the rotated
matrix, I found that three ofthe factor loadings were low (.153, .169, and .180). The
questions associated with the low loadings included, “I feel relief rather than satisfaction
after completing a task that required a lot of mental effort,” “I only think as hard as I have
to,” and, “I prefer to think about small daily projects to long-term ones.” The reliability
analysis indicated that the reliability of the scale would stay virtually the same with each
of the three items deleted (changing from .879 to .878, .873, and .873 respectively). I
deleted the items in turn and found that one dominant factor explained 36.8% of the
variance. Factor loadings on the rotated matrix ranged from .688 to .277, with 11 of the
15 items loading above .4. The reliability ofthe 15-item scale was o=.s73, compared to
a=..879 for the l8-item scale. I opted to retain the lS-item scale because of its stronger
construct reliability and validity. The scale’s constituent items are found in appendix G.
W The capacity for lifelong learning variable was
operationalized using the Capacity for Life-Long learning index (Hayek & Kuh, 1999),
which is a measure created by smnming students’ responses to 11 of the Estimate of
Gains items on the College Student Experiences Questionnaire developed by Kuh,
Vesper, Connolly, and Pace (1997). The “Estimate of Gain items represent the ability to
“learn to learn” and interact effectively with others in a complex, information-based
society, indicating the extent to which students have acquired continuous learning skills”
(Hayek & Kuh, 1999, p. 4). According to Hayek and Kuh (1999) the index “is reliable
(.84), with item-score correlations ranging fi'om .42 to .69 and item intercorrelations

ranging from .22 to .56” (p. 4). The index includes items asking students to indicate the

extent to which their college experiences have led to their progress in such areas as
“writing clearly and effectively,” “understanding new scientific and technical
developments,” and “ability to learn on your own, pursue ideas, and find information you
need” (Hayek & Kuh, 1999, p. 29). (All items in the Capacity for Lifelong Learning
index were used with permission from the CSEQ Assessment Program, Indiana
University, Copyright 1998, The Trustees of Indiana University.)

To ensure the construct validity and reliability of the Capacity for Lifelong
Learning scale for the current study, I conducted a factor analysis with the items fi'om the
scale. Again, I used quartimax rotation because of my expectation that there would be
one dominant construct. I found that one construct explained 46.2% of the variance, and
factor loadings on the rotated matrix ranged fiom .800 to .428. The reliability of the scale
was (F .88. The scale’s constituent items are found in appendix H.

Prior to administering the survey, I piloted it with two sets of individuals. The
first set was comprised of colleagues with knowledge and expertise in postsecondary
education and residential college environments. I asked them to provide me feedback
about the clarity of the questions, ease of navigating the web-based survey, and any
general observations they had. As a result of their feedback, I shortened several of the
questions. The second set of individuals who took the pilot included students in a new
residential college that was not included in the sampling flame of the study. Based on
feedback from these students, I clarified several demographic questions, including

parents’lfamily income, and how long students have resided in a residential college.

65

Data Analysis Procedures

Many existing studies examining the impact of living-learning community or
residential college environments on student outcomes approach these hierarchical
environments in one of three ways. Often, they pool all students in various living-
learning communities or residential colleges together, disaggregating differing
environmental variables to the student level, and then comparing those students to
students living in traditional residence halls. This pooled approach ignores between-
group variation among the different living-learning community or residential college
environments, resulting in the larger clusters dominating the coefficient estimates.
Furthermore, if the dependence of various units within a cluster is not accounted for in
the model, the variance will be smaller, which will result in smaller standard errors of
parameter estimates and a greater chance that the estimates will be statistically significant
(Krefi & De Leeuw, 1998).

Another approach taken in existing research is to aggregate the observations
within a living-learning community or residential college to the group level (level-2) and
then compare them to other living-learning community or residential college
environments. Although this approach preserves the variation between different living-
learning community and residential college environments, it ignores within-group
variation, ultimately disregarding such information (Krefi & De Leeuw, 1998).
Furthermore, although this approach may illustrate that one environment is more
effective than another for producing a certain outcome or students, it does not aid in
explaining for what type of students the environment is more effective or why the

environment is more effective (Kreft & De Leeuw, 1998).

A third approach to examining the impact of various living-learning communities
and residential colleges is to conduct multiple student level (level- 1) regression models,
one for each environment within the study. Not only is this approach cumbersome when
there are a large number of environments in the sample, it also multiples the likelihood of
type I or type H errors because different models are fitted for each context, resulting in
error structures not being specified properly (Kreft & De Leeuw, 1998). Furthermore, the
results from the regressions are difficult to compare across different environments, as
standard errors of the different models often vary considerably (Krefi & De Leeuw,

l 998).

Using Hierarchical Linear Modeling to analyze the data collected addressed the
analytic challenges formd in many existing studies because it accormts for the existence
of varying levels of data, allowing for level-two variables to explain between-group
variance in the level-one intercept and for the exploration of cross-level interactions
(Kreft & De Leeuw, 1998; Raudenbush & Bryk, 2002). Hierarchical Linear Modeling
was also appropriate given the nature of my conceptual framework, as it enabled me to
examine how students’ sociodemographic characteristics and residential college
environments were associated with their appraisal of the environment with regard to
liberal arts experiences, their motivation, and the interaction of these elements in
promoting or impeding their inclination to inquire or capacity for lifelong learning.

I used SPSS 16 and HLM 6.02 software to assist with data analysis. I began my
analysis by examining the descriptive statistics of the survey data, checking for missing
data, assessing normality, examining outliers, and conducting bivariate correlations of the

variables (presented in Appendices I and J). I also evaluated the reliability and construct

67

validity of the scales using confirmatory factor analysis, the results of which I discussed
above. Then, I turned my attention to my research questions, which I used as a guide in
constructing the multilevel model I used to analyze the data. Below I describe how I
answered each research question.

Does students ’ inclination to inquire or capacity for lifelong learning vary across
residential college environments?

I began my analysis by creating models of each of the outcome variables with no
predictor variables included in them. These one-way Analysis of Variance models with
random effects, also often called the null models, allowed me to estimate the variation in
students’ inclination to inquire and capacity for lifelong learning at the individual level
(level one) and across residential environments (level two), and to obtain baseline values
of deviance, which I would subsequently use to assess model fit of subsequent models

(Hox, 2002). Equation 3.1 displays the null model at level-one (the individual level)

Yij =50} +rii (3.1)
where Yy is the outcome variable (Inclination to Inquire or Capacity for Lifelong
Learning) for individual i in residential college j, ,601- is the intercept for residential
college j, and ry- is the level-one (individual) residual. Equation 3.2 displays the implied
model at level-two (residential college level)

[301° = 7’00 + “0 j (3.2)
where intercept fig}- becomes the outcome variable, 700 is the grand mean efi‘ect (the
fixed intercept at Level-2), and no} is the level-two residual. Equation 3.3 displays the

combined level-1 and level-2 equation,

68

Y5 = 700 + “Oj + r5]- (3.3)
where, again, Yij is the outcome variable (Inclination to Inquire or Capacity for Lifelong
Learning) for individual i in residential college j, 700 is the grand mean effect (the fixed
intercept at Level-2), and no} and "ij are the residuals at the residential college level and

the individual level respectively. Partitioning the residuals into within-residential college
and between-residential college enabled me to ascertain information about the degree of
variation in students’ inclination to inquire and capacity for lifelong learning within and
across different residential college environments. I did so by determining the intra-class
correlation, which is calculated by dividing the variance in level-two (between-
residential college) residuals by the total variance of level-one and level-two. In short, the
null model allowed me to determine that the means of students’ inclination to inquire and
capacity for lifelong learning were statistically different across residential college
environments. Furthermore, it provided me with a baseline from which to gauge the
reduction in variance in subsequent models.

How are students ’ sociodemographic characteristics and motivation related to their
inclination to inquire and capacity for lifelong learning?

In the second step of the modeling process, I created the within-residential college
(also called the individual or level—one) models. I entered four blocks of variables into
each model at level-one. The first block was comprised of dummy—coded race/ethnicity
variables and sex. Because of insufficient numbers, for the analysis I combined the
categories ‘American Indian or Alaskan Native’ students and ‘Native Hawaiian’ students
into a category, ‘other.’ By doing so, I did not mean to imply that these two groups of

students necessarily have similar experiences in college. Rather, because the low

69

numbers in both Categories limit any meaningful assertion, I opted to combine them in
favor of parsimony of the statistical models.

The second block was comprised of sociodemographic and pre-college
characteristics (parents’ income and education, and students’ high school grade point
average). I opted to exclude the ACT score variable fiem the analysis because it would
have severely limited the sample size. I excluded the SAT scores variable because
students’ responses to the survey indicated confusion regarding the question as a result of
scoring changes initiated in 2005 and as a result I was not confident about the validity of
responses.

The third block was comprised of students’ college experience and motivation
variables (including how many years they lived in the residential college, their
motivation, and their degree aspirations). The fourth block of variables was comprised of _
the group mean centered liberal arts experience variables. I opted to enter these at level-
one of the equation to capture students’ individual views regarding their liberal arts
experiences. I centered them to reduce multicollinearity. Also, since each residential
college’s mean of the liberal arts experience variables will be entered at level-2 of the
model, it was important to account for individual students’ views so as to better interpret
the environmental context. Equation 3.4 shows the equation with all four blocks of

variables entered.

70

,st (Intly- ) + [36}- (MulttRacij ) + [37 j (RaceOthg ) + ,st (NoRespRaceyo ) +

,69 1 (F lrstGeni j ) + ,6] 0 . (IncLowij ) + £1 1 j (IncHtghij ) +

,612] (HS'GPAiJ- - HSGPA» ) + ,613j (Motivationij — Motivations. ) + 3 4
,614 J. (YrsLivedij ) + ,6151-(DgASpNOBaChg-)+ s16]. (DgAspMoBadrij ) + ( - )
,617j (T eachClassij - T eachClass. ,- ) + [31 81- (TeachOutij - TeachOutoj ) +

,61 91- (Ac.Challenge,-j — Ac.Challenge, j ) + ,6201-(Diversityij - Diversity. j ) +

,6211- (Peersij — Peers. j ) + rij

 

 

In equation 3.4, Y3]- was the outcome variable (students’ Inclination to Inquire or

Capacity for Lifelong Learning depending on the model), the i subscripts indicate
individuals whereas the j subscripts indicate residential colleges. The symbol - indicates
mean of a context, so when a replaces the i, it indicates that it is the mean of individuals
in group j, whereas when 0 replaces both i and j, it indicates the grand mean across
residential college. In this model, I allowed only the intercept to vary across residential
college environments. As is indicated in equation 3.4, I grand-mean centered the high
school GPA variable and the motivation variable in order to make the intercept
meaningful, as prior to transformation, a zero value for these variables is outside the
range of probable responses. When grand-mean centered, however, the intercept can be
interpreted as the adjusted estimate of the outcome for an “average” student — those who
have mean characteristics (Raudenbush & Bryk, 2002). I also group-mean centered the
liberal arts experience variables (Good Teaching & Quality Faculty Interaction -
Classroom Practices; Good Teaching & Quality Faculty Interaction - Out-of-Class
Interactions; Academic Challenge and High Expectations; Diversity Experiences; and
Interactions with Peers). I chose to center these variables around the group mean in order

to make the intercept meaningful and reduce the multicollinearity of the variables

71

(Raudenbush & Byrk, 2002). The individual-level coefficients were “fixed,” meaning
that their slopes (or efi‘ects) were constrained to be the same across all residential colleges
(Raudenbush & Bryk, 2002).
Does the association between students ’ sociodemographic attributes, pre-college
characteristics, and motivation and their inclination to inquire and capacity for lifelong
learning vary across residential colleges?

In the next step of the modeling process, I examined each of the level-one
predictors in turn at level-two of each model to see if there was any variation in the
variable’s association with students’ inclination to inquire and capacity for lifelong
learning across residential colleges. I was curious to examine whether the association
between the predictor variables and response variables remained constant across
residential colleges (called a fixed effect) or whether the association changed
significantly depending on the residential college context (called a random effect)
(Raudenbush & Byrk, 2001). In order to determine whether there was variation in the
association of each level-one variables and students’ inclination to inquire and capacity
for lifelong learning, I modeled the variance of each level-one variable in turn at level
two and examined the change in deviance and the chi-square value to determine whether
the change was significant. After determining which slopes varied significantly, I
modeled the full within-residential college models. Then, using the deviance test as a
guide, I deleted variables that were non-significant throughout the modeling process in

order to obtain the most parsimonious models that explained the most variance.

72

How is the environmental context of a residential college related to students ’ inclination
to inquire and capacity for lifelong learning? (Specifically, Liberal Arts Experience
Environmental Variables and Disciplinary Focus)

With the level one model complete, I turned my attention to the group-level (i.e.,
the level-2 or residential college) model. I was interested in determining which of the
group-level variables explained the between-residential college variation in students’
inclination to inquire and capacity for lifelong learning. Starting with the outcome
inclination to inquire, I modeled the level-one intercept with two difi'erent sets of
contextual variables, one with each residential college’s mean of the liberal arts
experience variables entered in turn, and one with the disciplinary focus of the residential
colleges. I then did the same for Capacity for Lifelong learning, modeling the intercept of
the level-one model with the two sets of outcomes. Equations 3.5 and 3.6 show the level-

two models, which I added separately to the previous equations.

Lib. Arts Em .- floj = 700 + 701(GrpMnExTCIassj) + 702 (GIpMnExT Out I) + (3 5)
703 (GrpMnExChj) + 704 (GrpMnExDj) + 705 (GrpMnExPj) + qu .

Discp. Focus : [30]- = 700 + 701(DiscSocScij) + 702 (DiscNatScij) +

703 (DisHumj) + 1401- (3'6)

These models were random intercept models, since I allowed the intercept to vary
relative to the contextual characteristics of the residential colleges. Specifically, intercept
700 was a function of the grand mean across all residential colleges on the outcome
variable (inclination to inquire or capacity for lifelong learning) as well as the group-level
liberal arts experience and disciplinary focus variables plus the random error for a
specific residential college. Any significant reduction in the between-residential college

variance was due to the explanatory power of the group-level variables (Hox, 2002).

73

Does the association between student-level motivation and students ’ inclination to
inquire and capacity for lifelong dtfiizr by residential college-level liberal arts experience
variables (good teaching and high quality interaction with faculty, academic challenge
and high expectations, diversity experiences, and peer interactions)?

In the final step of the modeling process, I examined the cross-level interactions
between level-one and level-two variables. I was most interested in the roles of individual
motivation attributes and residential college environments in promoting students’
inclination to inquire and capacity for lifelong learning. Therefore, I focused my attention
on exploring whether the association between environmental liberal arts experience
variables (level-two) and individual motivation (level-one) had any relation to students’
inclination to inquire or capacity for lifelong learning.

Having already modeled the level-one intercept of each outcome variable with the
level-two liberal arts environment variables (which I described in Research Question
Four) and obtained a baseline deviance score, I added each group-level liberal arts
experience in turn to the slopes of two different level-one variables, starting with ‘Degree
Aspirations — More Than Bachelor’s’ and then focusing on ‘Motivation.’ In order to
determine whether the cross-level association between each individual-level variable and
liberal arts environment variable was significant, I examined the change in deviance and
chi-square statistic. Equations 3.7 and 3.8 show cross-level models, which I added
separately to the equation with all environmental liberal arts variables included. Although
the group-level variables are all included in the equations for simplicity of the
presentation, I added them one at a time in order to ascertain whether the association of

each one with the motivation variables was related to student’s inclination to inquire or

capacity for lifelong learning.

74

Deg. Asp - MoBach : [39] = 790 + 791(GrpMnExTClassj) + 792 (thMnErTOutj)
+ 793 (GrpMnExChj) + 794 (GrpMnExDj) + (3.7)
795 (GmMnExPj )[+u9j]

Motivation: [9131- = 7130 + 7131(GrpMnExTClassj) + 7132 (GrpMnExTOutj) +

7133 (GmMnExChj) + 7134(GmMnExDj) + 7135 (GrpMnExPj) (3-3)

The ‘Degree Aspirations-More than Bachelor’s’ slope was random for ‘Inclination
to Inquire’ and fixed for ‘Capacity for Lifelong Learning.’ The decision to set one slope
as random and one as fixed was based on my discovery that the association between high
degree aspirations and students’ inclination to inquire varied across residential college
context (see question three). For the ‘Motivation’ slope, both s10pes were fixed. The
examination of the cross-level interactions answers my fiml research question.
Limitations

Although careful stepswere takento ensurethatthedatalcollected and
subsequently analyzed were reflective of students and their experiences in residential
colleges, there are several limitations that are important to note. The first limitation
relates to my data collection procedures, and specifically the likelihood of coverage error,
sampling error and non-response error that occurred because of the way in which I
collected my data. I discussed these limitations in detail earlier in the chapter, but raise
them again to acknowledge that they may have implications for the external
generalizability of the findings.

The second limitation relates to my use of self-report data. Several researchers
have raised questions about the validity of self-report data (Anaya, 1999; Gonyea, 2005;
Pace, 1985; Pike, 1995, 1996). As a result, I have attempted to follow guidelines that
increase the validity of the data, including: 1) ensuring that the survey questions address

questions that respondents would know, 2) phrasing questions clearly and

75

lmambiguously, 3) referring to recent activities, 4) asking questions that merit a serious
response by the respondents, and 5) ensuring that questions do not embarrass or threaten
the respondents (Bradburn & Sudman, 1988; Converse & Presser, 1989; Gonyea, 2005;
Pace, 1985; Pike, 1995). By following these guidelines, I sought to mitigate the concerns
related to self-report data.

In this chapter I outlined my research design, analytic approach, and limitations.
In chapter4,ldetailtheresults ofthe study, answering eachresearchquestioninturnand
providing a full description of the relationship between the predictor variables with
students’ inclination to inquire and capacity for lifelong learning. In chapter 5, I discuss
the results of the study in light of existing literature and discuss implications for theory,

research, practice, and policy.

76

CHAPTER F OUR
Results
This chapter details the results of examining the association between students’
motivation and sociodemographic attributes, their residential college environments, and
their inclination to inquire and capacity for lifelong learning. I discuss the findings in the
order of the research questions posed in chapter three and in the sequential order in which
I added blocks of variables to the prediction models. I provide a full description of their
association with students’ inclination to inquire and capacity for lifelong learning.

Researc 'on e' Doesstudents’ inc ' 'on ' ' or i for ' el
11mm vary across midgtifl college en_v_r_r2' nments?

For this research questiOn, I was interested in the extent to which students’
afliliation with a particular residential college environment was associated with their
inclination to inquire or capacity for lifelong learning. In other words, how much of the
variation in students’ inclination to inquire and capacity for lifelong learning was
attributable to differences in their residential college environment. In order to determine
the association between students’ inclination to inquire, capacity for lifelong learning,
and their residential college environment, I conducted a one-way AN OVA with random
efi‘ects (also called a null model) to partition the variance in the outcome variables
(inclination to inquire and capacity for lifelong learning) into between-residential college
and within-residential college components. I found that a significant amount of the
variance in students’ inclination to inquire and capacity for lifelong learning was
attributable to their residential college environment.

Having found that the partitioning of variance to within-environment and

between-environment components was significant, I calculated the intra—class correlation

77

(ICC), or the proportion of the variance in each outcome that was explained by the
grouping structure (which in this case was the particular residential college environment)
(Hox, 2002). I determined the ICC by dividing the between-environment variance by the
total variance for each of the outcome variables. I found that the proportion of the total
variance that existed between residential college environments was 8.8% for ‘inclination
to inquire’ and 5.3% for ‘capacity for lifelong learning.’ I detail the partitioning of
variance in the null model in Table 4.1.

Table 4.1 .: Null Models: Partitioning Variance in Inclination to Inquire and Capacity for

Lifelong Learning

Capacity
Inclination for Lifelong
to Inquire Learning

 

Null models
Variance components -
Between-res. college environments (intercept 6.097 2.322
Within-res. college environments 63.420 41.449
Reliabilities 0.829 0.752
ICC 0.088 0.053

 

The between-residential college variance in students’ inclination to inquire and
capacity for lifelong learning may seem inconsequential. However, as Seifert (2006) and
Porter and Swing (1996) argued, higher education survey research often only explains
about 30% of the total variance in a given outcome. Furthermore, since in studies of
college immct the majority of variation in outcome is expected at the individual level
(Pascarella & Terenzini, 2001 , 2005), even a small amount of variation at the
environmental level merits exploration, as it may provide insight as to what aspects of the
environment are influential to a given outcome. Also, since residential college
environments are often resource rich and assumed to be more similar than different

(Ryan, 1993; Smith, 1994), with missions, resources, and organizational structures

78

designed to promote curricular and co-curricular integration, the environmental variation
is particularly interesting. Determining environmental influences on students’ inclination

to inquire and capacity for lifelong learning will most certainly have implications for

policy and practice.

W Two How are WW

 

 

For this research question, I was interested in examining the extent to which
individual sociodemographic and pre—college characteristics, college experience,
motivation, and aspirations, and students’ perceptions of liberal arts experiences were
associated with their inclination to inquire and capacity for lifelong learning within
residential colleges. Having decomposed the total variance into between-residential
college variance and within-residential college variance, for this question I focused on the
within-residential college variance, or how different individual attributes and experiences
(as opposed to environmental attributes) were associated with students’ inclination to
inquire and capacity for lifelong learning.

With regard to my conceptual fi-amework, which was Moos’s social-ecological
fi'amework (see Chapter 1), the within-residential college (level-one) model encapsulates
the personal system (sociodemographic variables), students’ cognitive appraisal
(individual liberal arts experience variables), and their activation or arousal (motivation,
degree aspiration and college experiences) on their efforts at adaptation and coping
(outcome variables). The stability or change is cross-sectional as it is measured by how
the number of years lived in the residential college is associated with students’ inclination

to inquire and capacity for lifelong learning.

79

Using HLM 6.02, I regressed the inclination to inquire and capacity for lifelong
learning outcomes on four blocks of variables to explore what relationship, if any, these
student attributes had on their inclination to inquire and capacity for lifelong learning. In
the first model, I regressed the outcome variables on sex and race/ethnicity variables. In
the second model, I added pre-college characteristics including parent/family education,
income, and high school grade point average. In the third model, I added college
experience, motivation, and aspiration variables including the number of years students
lived within the residential college, their motivation as determined by the survey of basic
needs, and their degree aspirations. In the fourth model, I added the group-mean centered
individual liberal-arts experience variables (good teaching and high quality interactions
with faculty (classroom practices and out-of class interactions with faculty), academic
challenge and high expectations, diversity experiences, and high quality interactions with
peers). These variables reflect the difference between students’ scores on the variables
and the mean of all the students in the sample fiom the same residential college.

In this section, I discuss the effects of each of these blocks of variables on
students’ inclination to inquire and capacity for lifelong learning. Table 4.2 presents all
the coefficients of the complete within-residential college models. Appendices K and L
display the step by step modeling process for both inclination to inquire and capacity for
lifelong learning respectively, starting with the first within-residential college (level-l)
model through the final between-residential college (level-2) model which includes the
environmental characteristics. Also in the appendices are the within- and between-
residential college variance explained calculations to which I refer throughout the results

section. The within- and between- residential college variance explained calculations are

80

determined at level level-one by subtracting the within- residential college variance of a
particular model fi'om the baseline within-residential college variance estimated in the
null model and then dividing by the within-residential college variance in the null model

2 __ 2
(e. g. R2 = (0" ’b 2 a, / "0) where 032.”, is the level-one variance estimate fiom the null

ar/b

model (baseline) and a}, m is the level-one variance estimate fi'om the model in which

the variance is explained (Hox, 2002). At level two, the between-residential college
calculation is the same, except that level-two variance components are used in each of the
calculations (Hox, 2002).

Table 4.2 Association between Students’ Sociodemographic Characteristics, Motivation,

and College Experiences and their Inclination to Inquire and Capacity for Lifelong
I . g .

 

 

Capacity for
Inclination Lifelong
to Inquire Learning
Coefficient sig. Coefficient sig.
Intercept 51.648 *" 26.050 “"
Sociodemographic variables
White (reference group)
African-American -0.786 n. s. 0.048 n. s.
Asian-American -3.280 "* 0.141 n. s.
Hispanic/Latino -l.530 n. 8. 1.355 *
International 4.528 " 1.543 n. s.
Multiracial -0.457 n. 3. -0.770 n. s.
Race-Other 3.632 n. s. 4.291 *
No response 4.200 n. s. -l.074 n. 3.
Male 1.629 ” 0.327 n. s.
First-generation student -0.270 n. 3. 0.268 n. s.
Family Income 850-110): (reference group)
Family Income Below $50K -0.462 n. 5. 0.278 n. s.
Family Income Above $110K 0.320 n. 8. 0.169 n. 3.
HS GPA -0.176 n. s. . 4.059 A

81

Table 4.2 (cont’d)

College Aspiration variables
Motivation 0.138 "‘ 0.126 *"
Number of Yrs Lived in RC 0.749 * 1.426 "*
Degree Aspirations —
- Bachelor's (reference group)
- Less than Bachelor’s 0.193 n. 3. —0.052 n. s.
- More than Bachelor's 2.748 *“ 1.844 *"
Liberal Arts Experiences
Teaching & Quality Interactions
w/ Faculty
Classroom Practices 0.184 “W 0.049 n. s.
-0ut-of-Class Interactions 0.1 12 " 0.171 **
Acad. Challenge & Expectations 0.041 " 0.096 W"
Diversity Experiences 0.137 " 0.077 "
Quality Interactions w/Peers -0.175 “ 0.004 n. s.
Variance Components
Between-Residential Colleges
(intercept) 3.015 ~"* 1.579 "*
Between-Res. Colleges explained 0.506 0.320
Within-Residential Colleges 54.995 30.224
Within-Residential Colleges explained 0.133 0.271
Reliabilities
Intercept 0.663 0.659
Deviance (FML) 7713.752 7201.752

 

"p<.l; *p<.05; "p< .01; m p<.001

Sociodemographic Variables
The first block of sociodemographic variables, race/ethnicity and sex, explain
little within-residential college variation in students’ inclination to inquire and capacity
for lifelong learning (5.1% and 0.3% respectively). As noted in Table 4.2, for some
students, race/ethnicity and sex had significant association with their inclination to
inquire and capacity for lifelong learning, even after family economic and educational

circumstances, educational aspirations and motivation, and liberal arts experiences were

82

accounted for. Specifically, Asian-American students living in residential colleges
reported a significantly lesser inclination to inquire than their White peers. Even after
controlling for first-generation status, income, and high school grade point average,
Asian-American students reported scores in the inclination to inquire scale that were 3.28
points lower tlmn their White counterparts. Despite their lesser inclination to inquire,
there was no difference in Asian-American students’ capacity for lifelong learning. So,
although Asian-American students may have reported a slightly lower proclivity toward
or value for learning, their capacity for lifelong learning was not statistically different
from that of their White peers.

International students, on the other hand, reported a significantly greater
inclination to inquire than their White (domestic) peers, reporting scores that were on
average four points higher. However, International students’ capacity for lifelong
learning was not statistically difi‘erent fiom White students.’

In terms of sex, male students also reported a significantly greater inclimtion to
inquire than their female peers in residential colleges, reporting scores that were 1.6
points higher. However, like international students, male students did not report a
statistically different capacity for lifelong learning than their female peers. So whereas
male students’ value for or proclivity toward learning may have been slightly greater,
their capacity for lifelong learning was not statistically different fiom their female peers.

Hispanic/Latino students and students in the combined Native American and
Alaskan Native ‘other’ category reported a significantly greater capacity for lifelong
learning than their White counterparts, even after controlling for confounding influences

and experiences. However, despite their greater capacity for lifelong learning, these

83

students’ inclination to inquire was not significantly different from their peers after
controlling for confounding influences.

Interestingly, although the sociodemographic variables explained little of the
within-residential college variance, these variables account for 42.2 % of the between-
residential college variation in students’ inclination to inquire and 4.2% of the between-
residential college variation in students’ capacity for lifelong learning. This finding
suggests that the effect of race/ethnicity and sex that I detailed above may be more
pronounced in certain residential colleges than others, with residential colleges that have
a higher concentration of Asian-American students, Hispanic/Latino students,
international students, or female or male students having greater variation fiom those
with fewer students’ of color or an equal gender distribution.

Pre—college Characteristics

In addition to race/ethnicity and sex, the second block of variables on which I
regressed the inclination to inquire and capacity for lifelong learning variables dealt with
students’ other sociodemographic characteristics and pre-college experiences. As noted in
Table 4.2, these variables (which included parents’ education, family income, and high
school grade point average) had very little association with students’ inclination to
inquire and capacity for lifelong learning. The addition of these variables only explained
.2% more of the variation in students’ inclination to inquire and 1.9% more of the
variation in students’ capacity for lifelong learning. The only variable in this block that
had a marginally statistically significant association with either of the two outcome
variables was high school grade point average, which after controlling for confounding

influences, was negatively associated with students’ capacity for lifelong learning. In the

84

model, this variable was grand-mean centered, and thus the interpretation of the finding is
that for every point above the grand mean a students’ grade point average was, there was
a decrease of -1 .059 in their capacity for lifelong learning score.
Motivation, College Experiences, and Aspirations

Whereas the sociodemographic and pre-college blocks of variables explained only
5.3% of the within—residential college variation in students’ inclination to inquire and
2.2% of the within-residential college variation in students’ capacity for lifelong learning,
the addition of the third block of variables bolstered the explained within-residential
college variance to 12.5% and 20.7% respectively. Specifically, the motivation variable
had a statistically significant positive association with students’ inclination to inquire and
capacity for lifelong learning. For every point increase in students’ self-reported score on
the basic needs survey (measuring students’ satisfaction in their sense of autonomy,
competence, and relatedness), there was a small, but significant increase in their
inclination to inquire and capacity for lifelong learning. Furthermore, students’
aspirations to obtain more than a bachelor’s degree were also positively associated with a
greater inclination to inquire and capacity for lifelong learning, even after controlling for
potentially confounding influences. Students who aspired to obtain more than a
bachelor’s degree reported on average a 1.8 point greater inclination to inquire on the
index and a 2.7 point greater capacity for lifelong learning. On the other hand, students’
aspirations to obtain less than a bachelor’s degree had no statistically significant
association with their inclination to inquire or capacity for lifelong learning. However,
because there were only 26 students in the sample who aspired to obtain less than a

bachelor’s degree, caution should be used when interpreting this result, as it is possible

85

that if more students with low degree aspirations were included in the sample, a
statistically significant association between the variable and students’ inclination to
inquire and capacity for lifelong learning may have been found.

The number of years students resided in the residential college environment also
had a significant positive association with their inclination to inquire and capacity for
lifelong learning, as for every year increase in their residency was an associated increase
of .75 points in their inclination to inquire and 1.4 points in their capacity for lifelong
learning. What remains unclear, however, is whether the positive association is due to
students’ interaction with their residential college environment over time, or rather is as a
result of their maturation.

Also noteworthy was how much variation the college experience, degree
aspiration, and motivation variables explained across residential colleges. The addition of
these variables to the model increased the between-residential college variation explained
in students’ inclination to inquire by almost 26%, raising it to 67%. For the capacity for
lifelong learning outcome variable, the addition of the college experience, degree
aspiration, and motivation variables increased the between-residential college variance
explained fiom virtually nothing to 46%. These findings suggest that the effect of college
experiences, degree aspirations, and motivation that I detailed above may again be more
pronounced in certain residential colleges than others.

Liberal Arts Experience Variables

The final block entered into the within-residential college model was comprised

of the group-mean centered liberal arts experience variables. The entry of these variables

at level one served two purposes; it enabled me to ascertain the association between

86

students’ individual experiences with their variables and their inclination to inquire and
capacity for lifelong learning, which I could then compare to the environment (level-two)
results. In addition, the addition of these variables at level—one served to control for
students’ individual beliefs about the environment, allowing me to hone in on the
potential environmental impact. In reference back to my conceptual fiamework, these
variables captured students’ cognitive appraisal of their environment.

With the addition of the group-mean centered liberal arts experience variables
(which included individual students’ perceptions of ‘good teaching and high quality
interactions with faculty’ [both ‘classroom practices’ and ‘out-of-class interactions with
faculty’], ‘academic challenge and high expectations,’ ‘diversity experiences,’ and ‘high
quality interactions with peers’), the final within-residential college models explained
13.3% of the individual-level variation in students’ inclination to inquire and 27.1% of
the individual-level variation in students’ capacity for lifelong learning. The ‘classroom
practices’ component of ‘good teaching and high quality interactions with faculty’ had a
positive association with students’ inclination to inquire, albeit small (for each point
increase in students’ self-reported classroom practices, a corresponding .184 increase in
their inclination to inquire). Interestingly, the ‘classroom practices’ component did not
have a statistically significant association with students’ capacity for lifelong learning. So
although students with a high inclination to inquire also reported that their instructors
employed effective classroom practices, the high quality instruction they received did not
immediately translate to students’ reporting a higher capacity for lifelong learning. That
said, students’ out-of-class interactions with faculty were positively associated with their

capacity for lifelong learning (.171 increase in capacity for lifelong learning for every

87

point gain on their out-of-class interactions scale), and marginally significantly associated
with the inclination to inquire (.112 increase in their inclination to inquire for every point
gain on their out-of-class interaction scale).

Students who reported that their academic experience was challenging and that
expectations were high also reported a greater capacity for lifelong learning.
Furthermore, there was a marginally significant positive association between students’
reports of their academic experience as challenging and their inclination to inquire,
however, in a practical sense the association was quite small (for each point increase in
students’ self-reported academic challenge and high expectations, a corresponding .041
increase in their inclination to inquire). So, whereas students’ experience of their
collegiate environment as challenging may have promoted the requisite skills for lifelong
learning, it did not necessarily deepen their value for or proclivity toward learning.

The positive association among students’ inclination to inquire, capacity for
lifelong learning, and their reports of diversity experiences was small, but significant.
Because the model is not causal, it remains unclear whether students who have a greater
inclination to inquire and capacity for lifelong learning are also students who would be
more apt to seek out diversity experiences, or whether students’ exposure to diversity
experiences was influential in promoting their inclination to inquire and capacity for
lifelong learning. Modeling the group mean of the liberal arts experience variables at
level-2 of the model (research question 4) will provide a better sense of how an overall
environment or ethos favoring diverse interactions may facilitate or impede students’

inclination to inquire and capacity for lifelong learning.

88

Finally, there was a negative association between students’ inclination to inquire
and their interactions with peers. Again, the practical significance of the association was
small (.175 point decrease in students’ inclination to inquire for every 1 point above the
mean of their residential college they scored). The association between students’ capacity
for lifelong learning and their interactions with peers was not significant. Again, because
there is not a causal relationship between the variables, it remains unclear whether the
deep and meaningful relationships with peers actually detract from students’ inclination
to inquire or rather that students’ who value deep and meaningful relationships with peers
might place that value over their proclivity toward learning.

Summary of Within—Residential College Model Findings

The within-residential college model, which included sociodemographic
information such as students’ sex, race/ethnicity, parents’ education and income, high
school grade point average, motivation, degree aspirations, years lived in the residential
college and the group-mean centered liberal arts variables explained 13% of the variation
within residential colleges of students’ inclination to inquire and 27% of the variation
within residential colleges of students’ capacity for lifelong learning. The addition of the
college experience, degree aspiration, and motivation variables explained the most
within-residential college variation for both inclination to inquire and capacity for
lifelong learning. These findings highlight the importance of considering such individual
factors when examining the effectiveness of residential college environments.

Consistent with ecology theory (Bronfenbrenner, 1979; Moos, 1979), although the
addition of the college experience and motivation variables into the model explained

more of the within-residential college variation, none of the significant associations were

89

overwhelmingly large. As a result, it is impossible to point to one individual attribute or
experience to explain the variation in students’ inclination to inquire or capacity for
lifelong learning. Furthermore, much of within-residential college variance remains
unexplained, as the final level-one model only accounted for 13.3% of the within-
residential college variation in students’ inclination to inquire and 27.1% of the within-
residential college variation in students’ capacity for lifelong learning. These findings
indicate that the individual characteristics that may be most closely associated with
students’ inclination to inquire and capacity for lifelong learning remain unknown.

The amount of variation between residential colleges explained by the
individual/within-residential college model is also noteworthy, as the final within-
residential college model explained 50.6% of the variation between residential colleges in
students’ inclination to inquire and 32% of the variation between residential colleges in
students’ capacity for lifelong learning. As noted earlier, these findings indicate that
many of the individual factors that are associated with students’ inclination to inquire or
capacity for lifelong learning are concentrated in certain residential college environments,
and support the findings that these environments vary fiom one another.

Research (Mica Three; Does the asggiation between mg giodemomhic
attributes, pig-college characteristics, and motivation and their inclination to m uire or
wig for lifelong learning vg across residential colleges?

 

 

For this research question, I was interested in examining whether the association
of students’ inclination to inquire or capacity for lifelong learning and the within-
residential college variables I explored in Research Question Two varied across
residential colleges. In reference to Moos’ Social-Ecological Framework (discussed in

chapter 1), in this question, I was examining the relationship of the arrows between the

90

environmental system and students’ cognitive appraisal, activation or arousal and the
outcome variables. Such variation across residential colleges would indicate that the
slopes of the predictor variables (the within-residential college/individual variables) and
the response variables were random, in other words that they changed depending on the
residential college context (Raudenbush & Byrk, 2001). Educational researchers are
increasingly interested in examining these conditional effects (Pascarella, et al., 2005;
Seifert, 2006). Within HLM, it is possible to examine conditional effects by randomizing
the slope of a coefficient, as opposed to fixing it, which is done by modeling the variance
of the level-one predictor at level-two (by randomizing the error term).

Starting with students’ inclination to inquire, I modeled the variance of each
level-one predictor in turn at level two and examined the change in deviance and the chi-
square value to determine whether the change was significant. I then turned my attention
to students’ capacity for lifelong learning, again modeling the variance of each level-one
predictor in turn at level-two and examining the change in deviance and chi-square value
to determine whether the change was significant. Table 4.3 presents the baseline deviance
from the Full Maximum Likelihood model of the complete within-residential college
model followed by the deviance, chi-square change statistic, and significance of the
model with each the level-one predictor modeled in turn at level-two.

Table 4.3. Deviance, Chi-Square Change, and Significance of Random Level-One Slopes

 

 

Inclination to Inquire Capacity for Lifelong Learning
Deviance (FML) Chi-Sq. Sig._ Deviance (FML) Chi-Sq. 8L
(baseline) u0 7713.752 7201.752
Male ul 7713.792 0.040 n. 5. 7201.567 0.700 n. s.
AfAm u2 7713.703 0.049 n. 3. 7200.963 0.789 n. s.
AsAm u3 7709.843 3.909 n. 5. 7200.171 1.581 n. s.

91

Table 4.3 (cont’d)._

 

His/Lat u4 7713.785 0.033 n. 5. 7197.887 3.865 n. s.
Intl u5 7713.123 0.629 n. s. 7201.217 0.535 n. s.
MultiRac u6 7713.758 0.006 n. 5. 7200.864 0.888 n. s.
RaceOth u7 7712.936 0.816 n. 3. 7201.573 0.179 n. s.
' NoRespRace u8 7713.557 0.195 n. s. 7200.792 0.960 n. s.
HSGPA u9 7713.045 0.707 n. 5. 7200.81 1 0.941 n. s.
NoBach u10 7713.434 0.318 n. 5. 7200.742 1.010 n. s.
MoBach ull 7705.629 8.123 "' 7201.565 0.187 n. s.
FirstGen u12 7713.310 0.442 n. 5. 7201.770 0.018 n. s.
1.0me u13 7711.759 1.993 n. 3. 7198.703 3.049 n. s.
HiInc ul4 7713.583 0.169 n. 3. 7201.729 0.023 n. s.
YrsLived u15 7712.453 1.299 n. 3. 7201.769 0.017 n. s.
MO'I'IV u16 7713.690 0.062 n. 5. 7201.787 0.035 n. s.
TeachClass u17 7713.275 0.478 n. 3. 7200.788 0.964 n. s.
TeachOut u18 7713.025 0.728 n. 5. 7197.236 4.516 "
AoChall ul9 7713.625 0.127 n. 5. 7201.041 0.711 n. 5.
Diversity u20 7709.212 4.541 " 7198.930 2.822 n. 5.
Peers u21 7709.644 4.108 n. 5. 7201.718 0.034 n. s.

"p<.l; * p < .05; 9* p < .01; m p < .001

As illustrated in table 4.3, very few of the random level-one slopes significantly
reduced the deviance score of the Full Maximum Likelihood models. Such reduction
would have indicated that more variance was explained by randomizing the slope of the
coefficient, and thus, that there was significant variation in the association of the
variables across environments. Specifically, only the association between students’
aspirations to obtain more than a bachelor’s degree and their inclination to inquire varied
by residential context at the p<.05 significance level. The variance in slope across
residential colleges that was associated with obtaining more than a bachelor’s degree was
9.77. The other associations of the within-residential college/individual variables and
students’ inclination to inquire or capacity for lifelong learning did not vary significantly
based on residential college context. The marginal significance of the random level-one

liberal arts experience variables will be explored in more detail in Research Question

92

Four, which examines the association of the overall group mean of the liberal arts
experience variables and students’ inclination to inquire or capacity for lifelong learning.
Because the variance associated with the random slope of ‘degree aspirations-

more than bachelors’ was small, I opted to keep the slope fixed for subsequent analyses
in favor of model parsimony. In addition, I used the deviance score of the model and
Akaike’s Information Criterion AIC (Akaike, 1987 as cited in Hox, 2002), which is
calculated using the deviance and number of parameters, to guide the deletion of
variables that were non-significant throughout all of the models so as to further increase
the parsimony of the level-one model. I deleted the following variables fiom the
inclination to inquire model: ‘degree aspirations-less than Bachelor’s’ and ‘family
income-Below $50,000.’ Despite the fact that the variables ‘high school GPA,’ ‘family
income-Above $150,000,’ and ‘first-generation student’ were non-significant throughout
the models, I opted to keep them based upon the amount of deviance each explained in
the final parsimonious model. Equation 4.1 shows the final level-one model for
inclination to inquire.

Inclolnqij == floj +fllj(MaIe,J-)+,sz(AfAmy)+fl3j(AsAmiJ-)+fl4j(His/Laty)+

,st (Intlij ) + fl6j (MultiRacij ) + ,67 j (RaceOth?- ) + 381‘ (NoRespRaceij ) +

fig j (F irstGeni j ) + ,610 .(IncHighU ) + 3111' (HSGPAij - HSGPA» ) +
6121(Motivationij —Motivation..)+ ,613j(YrsLived,-j)+ (4-1)
,614 j (DgASpMoBachij ) + B] 51- (TeachClass i]. — T eachCIass of ) +

,616 j (T eachOut ,3- — T eachOut .j ) + [917 ,- (Ac.ChalIenge ,3- — Ac.Challenge ,1 )

+ ,618 j (Diversity ,3- - Diversity ,j ) + ,619 1 (Peers ,3- — Peers .j ) + rij

 

 

 

For the capacity for lifelong learning model, I deleted the following variables:
‘male,’ ‘degree aspirations-less than Bachelor’ s,’ and ‘family income-Below $50,000.’
Even though they were not significant throughout the models, I opted to keep ‘family

income-Above $150,000,’ and ‘first-generation student’ because of their influence on the

93

deviance score of the model. Equation 4.2 shows the final, parsimonious, level-one model
for capacity for lifelong learning.

CapoeLrnij = 1301 +fllj(AfAmgj)+fl2j(ASAmy)+fl3j(Hi’/Latfi)+fl4j(lnfly’)
+flsj(MuItrRac ij)+ ,B6j(Race0th,j)+,B7J-(No Re spRacey)+ (4.2)

fl8j(FirstGen ,1. ) + ,69j(IneHigh ,1. ) + ,Bloj(HSGPA ,1. — 2755543. )
+ 311 j( Motivation ij - Motivation .. ) + ,6121.(YrsLived ij )

+ fll3j(DgAspMoBac hij- ) + ,614j(TchCIass ij - TchCIass ej)+
fl15j(Tch0ut i]- - TchOut oj)+ ,616j(Ac.Chally- - Acffl ej)

+ ,Bl7j(Diversit)z ij — Diversity ,1. ) + ,6181-(Peersij — Peers 0] ) + rij

Research Question F our: How is the environmental context of a residengl' college
;e_lated to students’ inclination to inquireand camcitv for lifelong learning (Specifically,
Liberal Arts Exm’ence Environmental Variables and Disciernag’ Focus 1?

For this research question, I was interested in examining the relationship between

 

 

the characteristics of the residential college environment and students’ inclination to
inquire or capacity for lifelong learning. Specifically, I was curious about how an overall
ethos marked by the liberal arts experiences (good teaching and high quality interactions
with faculty; academic challenge and high expectations; diversity experiences; and quality
interactions with peers) and disciplinary focus were associated with students’ inclination to
inquire or capacity for lifelong learning. With regard to my conceptual framework, which
was Moos’s social-ecological framework (see Chapter 1), the between-residential college
(level-two) model encapsulates the environment system, including the mean score of the
liberal arts experiences in a residential college as well as disciplinary focus of the college
on students’ inclination to inquire and capacity for lifelong learning (efforts at adaptation
and coping).

Having decomposed the total variance into between-residential college variance
and within-residential college variance, for this question I focused on the between-

residential college variation, or whether there was a relationship between the different

94

environmental attributes and students’ inclination to inquire or capacity for lifelong
learning. I focused my analysis on two sets of contextual characteristics: environments
marked by the liberal arts experience variables and the disciplinary focus .of the
residential college.
Environmental Liberal Arts Experience Variables

I began my analysis by focusing on liberal arts experiences. Whereas in the level-
one model I focused on how individual students’ self-reported assessment of the liberal
arts experience variables was associated with their inclination to inquire or capacity for
lifelong learning (thus capturing students’ cognitive appraisal of their environment), in
this question I was interested in how an environment marked by an overall ethos of each
of the liberal arts experiences (good teaching and high quality interactions with faculty;
academic challenge and high expectations; diversity experiences; and quality interactions
with peers) was associated with students’ inclination to inquire or capacity for lifelong
learning. I used the mean score of individuals within each residential college as a proxy
for the ethos of each residential college environment with regard to the liberal arts
experience variables. I added each contextual variable to the parsimonious level-one
models in turn and then created a model with all of the contextual liberal arts experience
variables. To aid in interpreting my findings, I used two tools, the between-residential
variance explained calculation (which I also used in to interpret the level-one model) and
each model’s deviance score and chi-square change statistic. A significant drop in the
deviance score (and a significant chi-square change statistic) would indicate that the
addition of the environmental variable was explaining the variation between residential

college environments. Because I was entering variables one at a time, these scores

95

provided insight into the magnitude of the association between each of the environmental
variables and the outcome variables. I present the coefficients of the association between
the contextual liberal arts experiences and students’ inclination to inquire or capacity for
lifelong learning, the variance and variance explained, the reliability, and the deviance
and chi-square change statistics in Tables 4.4 and 4.5.

As illustrated in table 4.4, the ‘classroom practices’ component of ‘good teaching
and high quality interactions with faculty’ and the ‘academic challenge and high
expectations’ group-mean variables explained the most between-residential college
variation in students’ inclination to inquire. The ‘classroom practices’ component
dropped the deviance score by 13.806 and raised the between-residential college variance
explained fiom 50.2% to 83.7%. The ‘academic challenge and high expectations’
variable dropped the deviance score by 13.273 and raised the between-residential college
variance explained from 50.2% to 74.6%. Although the ‘out-of-class interactions with
faculty’ component of the ‘good teaching and high quality interactions with faculty’
group-level variable was also significant when added on its own, its significance
dwindled when the final model with all the group-level liberal arts experience variables. '

In the final model, which accounted for the overall ethos of all the liberal arts
experiences, the ‘academic challenge and high expectations’ variable remained
significant, the ‘classroom practices’ variable dropped to marginal significance, and the
‘out-of-class interactions with faculty’ variable dropped to nonsignficance. These
changes in significance are likely a result of the high correlations between the level-two

liberal arts experience variables (reported in Appendix J). The other two group-level

96

:5. V a 1.... 2o. V a a... ”no. V a s £.Vn_<

 

m a a _ _ _ a:
t «8.2 < an." .3. an: t «5.2 t 2:: r... 8%— 2.5% 82.520
R a a mu 8 mm «a assuage
”$.32. REE 34.2: 48.8: 03.32. cages 80:? as: 833:
is $3 485 83 82 33 33 58.35
again:
RS 23 2:. £3 83 55 3.2. 8:330
_ .30 sees;
85.: new: 23m 83m Swen 23a ”8.: .30 seas“?
and :3 83 33 £3 53 83 8:33
.60 .em-§m
. £3 t. 82 r... :3 t 2.2 t; :3 2.. e8... .2. Rod osv._oo.s¢-§um
3.3.—.5
.9: :3. a... 82 soon? .2: £130
.2 22. .2 £2 «88:89.» £939
9 «ad a... mmmd doame.EU.u3<
.m d 8:. a... need mas—0.3.50.
< ”and t. 23 so? seems?
. e385; 83335
£30 £388.
3333;330
.. SEN t. :93. t. 82.. t. at: t. 49.3 .3. 2a.: 2: an: “9285
we Sm 52 mm as; + ma bemoan new $9.30 1% as... a new eoélmmm .882 dam...
EBB a :23 + _ 93 + _ 35 so + £6 + :33
+ _ .23 _ E3 _ .23 can

 

233 2 83:2: .383"... as 80883“ 32 335 .3 332 fineness“ 8058 83834 ”4.... 0:3

97

:5. Va .3... ”8. V a a... "we. V a e m—.Va<

 

m _ _ _ _ _ a:
a: Smfim 8a.: a... mN—S a:- egk 9...... mcmd— .md ~34 UOQSMQHIHUW EU
8 an an «N a an a cages;
83.8: 298:. 233:. 303: 3.3»: 8285 38.83 SE: 85.6:
on; .33 ~33 38.3 2.2. 38.3 38.3 30235
835353:
:3 :3 ES :3 E... 2.3 Ed 85293
.30 .3353
323 83.8 38.8 38.8 «8.8 $23. 32.3. .30 .8m.§§
E3 Rm... 38 84.3 83 82 and 8:335
.30 amazon
_. 32 E: t... 22 at £2 2. £20 3... 2.2 t. #2 .Aauesé
.30 .8353
00an>
.3. a 83 83 3.8%; .35 £35
.3. a $3.3 . ”23 88380 £3.35
.33 :33. t 33 .38me .3033
a. no: I... mead mun—oeoéo.
< :20. .3. 5.3 3.835 338230.
e385; £88.23
£35 a «338.;
89:23am“.—
E< .82.: - .530
._ 33.3 8:: : 83.. t. 8:: : Se: < ”.33 t. 208 33235
#3. ea 22 1% use + .ww £2.95 0mm ofifi0lmm 3% 3 am 888... new 383 42mm
3.8: .3 :23 + _ 35 + _ .23 so 820+ :23
+ _ 35 + _ .23 :93 .3.—

 

mefiafi muses 3e N33.3.00 .9835 as eofifiaxm as. ES: 3 80.32 acoaaesam 3858 83832 ”2. as;

98

variables, ‘diversity experiences,’ and ‘interactions with peers’ were not significant when
entered on their own, nor were they significant in the final model. In sum, for every point
increase in an environment marked by academic challenge and high expectations (as
measured by the group mean) there was an associated gain of .214 points in students’
inclination to inquire score. Likewise, although it was only marginally significant, for
every point increase in an environment marked by effective teaching practices, there was
an associated gain of .598 points in students’ inclination to inquire. Although these gains
are modest, when one considers that individual’s responses to the liberal arts experience
variables are already controlled for, they are noteworthy.

Table 4.5 illustrates the association between the group-level liberal arts variables
and students’ capacity for lifelong learning. All of the group-level liberal arts experience
variables were significant when added on their own into the parsimonious level-one
model, with the exception of the ‘classroom practices’ component of ‘good teaching and
high quality interactions with faculty.’ The ‘out-of-class interactions with faculty’
component of ‘good teaching and high quality interactions with faculty’ clearly had the
strongest association with students’ capacity for lifelong learning. When added to the
parsimonious level—one model, ‘out-of-class interactions with faculty’ lowered the
deviance score by 16.565. In addition, it raised the explained variance of the between-
residential college model from 32% to 70.9%. The variable ‘interactions with peers’ was
also significant when entered on its own, lowering the deviance score by 11 and the
raising the explained variance of the between-residential college model to from 32% to
55.7%. However, when entered into the final model, ‘interactions with peers’ dropped to

non-significance.

99

Interestingly, the ‘classroom practices’ component of ‘good teaching and high
quality interactions with faculty’ rose to marginal significance in the model with all
group-level liberal arts experiences included. The negative association between the
‘classroom practices’ variable and students’ capacity for lifelong learning in the final
model is likely a result of the high correlations among the other level-two variables.
Given the fact that the significance is marginal, and the negative association between the
variables is small, the practical significance of the finding is limited. However, the
finding that an environment marked by an ethos of out-of-class experiences with faculty
is associated with students’ capacity for lifelong learning was significant, and can be
interpreted as for every 1 point increase of the mean score of students’ out-of-class
interactions with faculty there was a corresponding increase of 1.1 points in their self-
reported capacity for lifelong learning. Although the gain is modest, when one considers
that individual’s responses to the liberal arts experience variables are already controlled
for in addition to the overall ethos of liberal arts experience, it is interesting that out-of-
class interactions with faculty was still significant.

Disciplinary Focus

After examining the relationship of the environmental liberal arts experience
variables and students’ inclination to inquire or capacity for lifelong learning, I’tumed my
attention to exploring the relationship between the disciplinary focus of the residential
college environment and students’ inclination to inquire or capacity for lifelong learning.
I was particularly interested in examining how the disciplinary focus of the residential
college related to students’ capacity for lifelong learning, as past research indicated their

might be a relationship but failed to control for students’ motivation (Hayek & Kuh,

100

1998). Unfortunately, the sample of residential college environments I collected did not
include enough variation in disciplinary focus to make meaningful inferences about its
association to students’ inclination to inquire or capacity for lifelong learning, as 20 of
the residential colleges were classified as interdisciplinary/other, 1 was situated in the
social sciences, 2 in the natural sciences, and 1 in the humanities. As a result, I was
unable to run the analysis.
Summary of the Between-Residential College Model Findings

The between-residential college model, which included variables pertaining to an
ethos marked by the liberal arts variables (good teaching and high quality interactions
with faculty, academic challenge and high expectations, diversity experiences, and
interactions with peers) explained 82.2% of the between-residential college variation in
students’ inclination to inquire (3 2% of which was explained by the environmental liberal
arts variables). The model explained 77.5% of the between-residential college variation
in students’ capacity for lifelong learning (45.4% of which was explained by the
environmental liberal arts variables). The majority of the between-residential college
variation explained for inclination to inquire resulted from the addition of an
environmental ethos promoting ‘academic challenge and high expectations.’ For the
capacity for lifelong learning outcome, out-of-class interactions with faculty explained
the greatest amount of between-environment variation. These findings highlight the
importance of an ethos marked by these experiences.

Even though the environmental liberal arts experience variables explained much
of the variation between residential colleges, their association with the outcome variables

was not large. Furthermore, as illustrated by the first research question, the between-

101

residential college variation in students’ inclination to inquire was 8.8% and for capacity
for lifelong learning it was 5.3%, and so while these models explain a large amount of the
variation in the between-residential college outcomes, their overall contribution to
understanding the variation in students’ inclination to inquire and capacity for lifelong
learning is considerably less. I raise this point not to diminish the importance of the
findings, but rather to make sure that the findings are set in a larger context.

Does the association between student-level motivation and students’ inclination to inquire
and capacity for lifelong differ by residential college-level liberal arts experience

variables (good teaching and high quality interaction with faculg, academic challenge
fli high expectations, diversity experiences, and peer interactiofl?

In this final research question, I was interested in examining the interaction
between individual motivation attributes and residential college environments in
promoting students’ inclination to inquire and capacity for lifelong learning. Because
these interactions are between level-1 (individual) and level-2 (contextual/environment)
variables, they are often called cross-level interactions. With regard to Moos’s social-
ecological framework (see Chapter 1), in exploring the cross-level interactions I was
examining the relationship between the personal system and environmental system,
which was illustrated in Moos’ model by the arrows connecting the two systems.

Having already modeled the level-one intercept of each outcome variable with the
level-two liberal arts environment variables (which I described in Research Question
Four) and obtained a baseline deviance score, I added each group-level liberal arts
experience in turn to the slopes of two different level-one variables. I started with
‘Degree Aspirations — More Than Bachelor’s’ and then focused on ‘Motivation.’ I

determined whether the cross-level interaction between the individual-level motivation

variables and environmental liberal arts experiences were significant by examining the

102

chi-square change statistic and drop in deviance score. Table 4.6 presents the baseline
deviance from the Full Maximum Likelihood model of the between-residential college
model with all the environmental liberal arts experiences included followed by the
deviance, chi-square change statistic, and significance of the model with the cross-level
interaction of the motivation variables and each environmental liberal arts experience
variable.

Table 4.6: Deviance, Chi-Square Change, and Significance of Cross-Level

Interactions between Motivation (Student) variables and Liberal Arts Experience
(Environment) variables

 

Inclination to Inquire
Deviance (FML) Chi-Sq.

Capacity for Lifelong Learning
Deviance (F ML) Chi-Sq.

 

baseline 7684.737 7180.650
MoreBach/TeachPrac 7684.154 0.583 n. 3. 7180.650 0.000 n. s.
MoreBach/OutofClass 7683.864 0.873 n. 5. 7180.225 0.425 n. s.
MoreBach/Challenge 7684.574 0.163 n. 5. 7180.641 0.009 n. s.
MoreBach/Diversity 7684.730 0.007 n. 5. 7180.648 0.002 n. s.
MoreBach/Peers 7684.734 0.003 n. 5. 7178.848 1.802 n. s.
Motivation/TeachPrac 7684.662 0.075 n. 3. 7173.960 6.690 *
Motivation/OutofClass 7684.709 0.028 n. 3. 7179.963 0.687 n. s.
Motivation/Challenge 7684.588 0.149 n. 5. 7179.488 1.162 n. s.
Motivation/Diversity 7684.729 0.008 n. 8. 7180.400 0.250 n. s.
Motivation/Peers 7684.691 0.046 n. 3. 7180.251 0.399 n. s.

Ap<.l; It p<05; #11: p < .01; its”: P < .001

As illustrated in table 4.6, very few of the cross-level interactions between the
individual motivation variables (level-one) and environmental liberal arts experience
variables (level-two) significantly reduced the deviance score of the Full Maximum
Likelihood models of students’ inclination to inquire and capacity for lifelong learning.
Specifically, only the cross-level interaction between students’ motivation and the

‘classroom practices’ component of ‘good teaching and high quality interactions with

103

faculty’ was significantly related to students’ capacity for lifelong learning at the p<.05
level. The interaction effect was -.0203 7, meaning that a rise in teaching practices was
associated with a very slight decline in students’ motivation. From a practical
perspective, the interaction coefficient was so small as to be all but insignificant.

The other cross-level interactions of motivation variables (level-one) and
environmental liberal arts experience variables (level-two) and students’ inclination to
inquire or capacity for lifelong learning were not significant across residential college
context.

Summary of Results

The results of the current study indicated that a statistically significant amount of
variation in students’ inclination to inquire and capacity for lifelong learning existed
across residential environments. Although proportion of variance in students’ inclination
to inquire and capacity for lifelong learning was relatively small (8.8% and 5.3%
respectively), existing research has illustrated that even a small amount of variation at the
environmental level merits exploration, as it may provide insight into what aspects of the
environment are influential to a given outcome (Porter & Swing, 1996; Seifert, 2006).

The within-residential college model explained 13% of the variation within
residential colleges of students’ inclination to inquire and 27% of the variation within
residential colleges of students’ capacity for lifelong learning. The addition of the college
experience, degree aspiration, and motivation variables explained the most within-
residential college variation for both inclination to inquire and capacity for lifelong
learning. The amount of variation between residential colleges explained by the

individual/within-residential college model is noteworthy, as the final within-residential

104

college model explained 50.6% of the variation between residential colleges in students’
inclination to inquire and 32% of the variation between residential colleges in students’
capacity for lifelong learning. As noted earlier, these findings indicate that many of the
individual factors that are associated with students’ inclination to inquire or capacity for
lifelong learning are concentrated in certain residential college environments, and support
the findings that these environments vary from one another.

Having built within-residential college model for students’ inclination to inquire
and capacity for lifelong learning, I turned my attention to examining whether the degree
of association of these outcome variables and the within-residential college variables
varied across residential college context. By modeling the variance of each level-one
predictor variable in turn at level-two, I determined whether the slopes of the predictor
variables were random, thus indicating that the association changed depending on
residential college context. I found that only the association between students’ aspirations
to obtain more than a bachelor’s degree and their inclination to inquire varied by
residential college context at the p<.05 significance level. Furthermore, the variance in
slope was minimal, and as a consequence I Opted to keep all of the slopes ‘fixed’ for my
subsequent analyses.

The between-residential college model, which included variables pertaining to an
ethos marked by the liberal arts variables (good teaching and high quality interactions
with faculty, academic challenge and high expectations, diversity experiences, and
interactions with peers) explained 82.2% of the between-residential college variation in
students’ inclination to inquire and 77.5% of the between-residential college variation in

students’ capacity for lifelong learning. The majority of the between-residential college

105

Lilli

v -.
5L1.

r
5w
hail

variation explained for inclination to inquire resulted from the addition of an
environmental ethos promoting ‘academic challenge and high expectations.’ For the
capacity for lifelong learning outcome, out-of-class interactions with faculty explained
the greatest amount of between-enviromnent variation. These findings highlight the
importance of an ethos marked by these experiences.

Having estimated the level-one, within residential college and level-two, between-
residential college models for students’ inclination to inquire and capacity for lifelong
learning, I examined whether the interaction between the group-level variables in each of
the models and the motivation variables had a statistically significant association with the
outcome variables. I found that there was a statistically significant negative cross-level
interaction between students’ motivation and the ‘classroom practices’ component of the
good teaching and high quality interactions with faculty scale. The effect was so small
that there was little practical significance associated with this finding. Having examined
the associations of the variables at all different aspects of Moos’ model, my findings are

complete. I turn now to the discussion of my findings.

106

CHAPTER FIVE
Discussion and Implications

In this final chapter, I discuss the findings of the current study, situating them in
existing research and revisiting their relevance to Moos’s model. Then, I suggest
implications for theory, research, policy, and practice. I begin by revisiting the purpose of
the study and research questions. Next, I provide a brief summary of the findings reported
in Chapter Fourwhile discussing their relevance to existing higher educations research. I
also discuss the findings in relation to Moos’s model, and highlight where the model is
particularly useful in illuminating environmental and personal considerations. Then, I
consider the implications of the findings for theory, research, practice and policy. I
conclude by discussing questions for future research.

As postsecondary education is promoted as a necessity for participation in the 21St
century economy, academics, policymakers, and the public have all voiced concerns
about the quality and coherence of undergraduate education (AAC & U, 2007; Barr &
Tagg, 1995; Boyer Commission, 1998; U. S. Department of Education, 2006). Critics
point to the size, scope, and multiple missions of large, public research universities as
contributing to students’ feelings of anonymity, lack of engagement, and disconnection
from faculty (Astin, 1993; Boyer, 1987; Gaff, 1970; Gamson, 2000; Guskin, 1994;
Hawkins, 1999; Jerome, 2000). Although undergraduates may face additional challenges
at large, public research universities than they might in a more intimate and
undergraduate-focused setting, these institutions remain a likely destination for many
students to begin or complete their baccalaureate education because of their size, relative

affordability, and diversity in educational offerings.

107

University administrators increasingly turn to residential colleges and other types
of living-learning programs to address the size and scale conundrum facing large research
universities. By creating smaller enclaves of students living together initially, taking part
in a shared educational endeavor, and using resources within their environment that stress
academics (Inkelas, Zeller, Murphy, & Hummel, 2006), administrators and faculty
purport to create the atmosphere of a small liberal arts college while still offering students
the resources of a large university, including comprehensive research and library facilities
(Magolda, 1994; Schuman, 2005).

Despite the increasing popularity of residential colleges and other living-learning
programs, research examining their effectiveness is limited. Many studies have focused
on determining whether these environments are more effective than no intervention in
promoting students’ persistence (Pike, Schroeder, & Berry, 1997), academic achievement
(Pasque & Murphy, 2005; Pike, Schroeder, & Berry, 1997), and involvement and social
integration (Pike, 1999; Pike, Schroeder, & Berry, 1997). Virtually no attention has been
paid to whether and how these environments promote values associated with a liberal arts ,
education, including whether they deepen students’ inclination to inquire and capacity for
lifelong learning. Identified by the Center of Inquiry in the Liberal Arts at Wabash
College as one distinctive outcome of a liberal arts education, having a deep inclination
to inquire would suggest that a student has a strong desire to learn and continues to
pursue intellectual growth. Closely connected to students’ inclination to inquire is their
capacity for lifelong learning, which is defined by Hayek and Kuh (1999) as students’
ability to “‘learn to learn’ and interact effectively with others in a complex, information-

based society” (p. 4). Whereas a deepened inclination to inquire promotes a value for

108

continuing to pursue knowledge, a robust capacity for lifelong learning provides students
the tools to act upon their value for inquiry.

Another concern with existing research on residential colleges and living-learning
communities is that much of it is plagued with problems of analysis, as researchers have
often aggregated data they collect to the environmental level, disregarding the individual
differences of students in these environments, which may include their motivation,
sociodemographic characteristics, and experiences. By using an ecological approach that
accounts for environmental context and individual characteristics (Moos, 1976, 1979,
1986), in the current study I sought examine how students’ attributes (including their
motivation and other sociodemographic characteristics) and residential college
environments facilitate or impede students’ inclination to inquire and capacity for
lifelong learning. Specifically, I investigated the following questions:

1. Does students’ inclination to inquire or capacity for lifelong learning vary

across residential college environments?

2. How are students’ sociodemographic characteristics and motivation related to
their inclination to inquire and capacity for lifelong learning?

3. Do the associations between students’ sociodemographic attributes and
motivation and their inclination to inquire and capacity for lifelong learning
vary across residential colleges?

4. How is the environmental context of a residential college related to students’

inclination to inquire and capacity for lifelong learning? Specifically,

109

a. Environmental Liberal Arts Experience variables - (good teaching and
high quality interaction with faculty, academic challenge and high
expectations, diversity experiences, and peer interactions)

b. Disciplinary Focus

5. Does the association between student-level motivation and students’

inclination to inquire and capacity for lifelong differ by residential college-
level liberal arts experience variables (good teaching and high quality
interaction with faculty, academic challenge and high expectations, diversity
experiences, and peer interactions)?

From October 2008 through January 2009, I collected data from over 1800
undergraduate students in 24 residential colleges across the United States to examine the
association between their sociodemographic characteristics, college experiences and
motivation, residential college environments, and their inclination to inquire and capacity
for lifelong learning. I used multilevel analysis (Hox, 2002; Krefi & De Leeuw, 1998;
Raudenbush & Bryk, 2002) to ascertain the significant individual, contextual, and cross-
level associations of variables with students’ inclination to inquire and capacity for
lifelong learning.

Discussion of Findings

I frame my discussion of the findings using the different levels of analysis as a
guide. I begin by addressing the null model, which partitioned the variance into within-
and between-residential college contexts, and then focus more acutely on the
individual/within-residential college (level-one) and environment/between-residential

college (level-two) models of the associations between variables and students’ inclination

110

to inquire and capacity for lifelong learning. Throughout the discussion I connect
findings from the current study to existing research.
Null Model — Estimating Individual and Contextual Variation

In fitting the null model, I was interested in determining the extent to which
students’ affiliation with a residential college environment was associated with their
inclination to inquire or capacity for lifelong learning. Conducting a one-way AN OVA
with random effects enabled me to partition the variance in the outcome variables into
within-residential college and between-residential college components. I found that a
statistically significant amount of the variance in students’ inclination to inquire and
capacity for lifelong learning was attributable to students’ residential college
environment. Specifically, the residential college environment accounted for 8.8% of the
variation in students’ inclination to inquire and 5.3% of the variation in students’ capacity
for lifelong learning.

At first blush the amount of variation across environments may not seem
noteworthy. However, when placed in the context that residential colleges are often
considered more similar than different (Ryan, 1993; Smith, 1994), and that in collegiate
settings more variation is found between individuals than between environments
(Pascarella & Terenzini, 2001, 2005), it became clear that the variation across
environments merited exploration. The differences in outcomes across environments held
potential insight into what aspects of the environment are most closely associated with
students’ inclination to inquire or capacity for lifelong learning. The variation across
environments also justified the use of a multilevel model to explore the associations

between the variables.

111

In addition, the findings from the null model support existing literature calling for
a need to take an ecological approach to exploring liberal arts outcomes, as the multitude
of factors affecting student outcomes include both environmental and personal factors
(Pascarella & Terenzini, 2005). These factors often are interacting and may influence one
another (Bronfenbrenner, 1979; Moos, 1976, 1979, 1986).

In emphasizing the need for an ecological approach to examining residential
college environments, the findings of the null model also help to contextualize some of
the mixed findings from existing studies. Specifically, those studies that aggregate the
variance in student outcomes to the group-level and compare across environments (e. g.
Inkelas & Weisman, 2003; Pasque & Murphy, 2005) may ignore the greatest source of
variation, which are often individual characteristics and experiences (Pascarella &
Terenzini, 1991, 2005). The results of the null model in the current study were consistent
with college impact research (Pascarella & Terenzini, 1991 , 2005) in that there was much
more variation in students’ inclination to inquire and capacity for lifelong learning within
residential colleges than between residential colleges. By failing to examine the variation
within residential environments, existing research may overlook important factors
affecting student outcomes, or may overstate the environmental impact on student
outcomes. On the other hand, the variation between environments, although small, was
significant, and lends support to the argument that residential college environments and
living-learning programs are not all the same (Inkelas, Longerbeam, Leonard, & Soldner,
2005; Wawrzynski & J essup-Anger, in press), and thus caution should be taken when
grouping students from these environments together without a strong rationale for doing

SO.

112

Within-Residential College Model —— The Association of Individual Attributes with
Students’ Inclination to Inquire and Capacity for Lifelong Learning

In fitting the within-residential college (level one) model, I was interested in
examining the extent to which students’ sociodemo graphic and pre-college
characteristics, college experiences and motivation, and individual perceptions of their
liberal arts experiences were associated with their inclination to inquire and capacity for
lifelong learning. By regressing the inclination to inquire and capacity for lifelong
learning outcomes on the four blocks of variables, I was able to determine how closely
associated these variables were with the outcome variables. I found that the overall
within-residential college (level one) model explained 13.3% of the within—residential
college and 50.6% of the between-residential college variation in students’ inclination to
inquire. The model explained 27. 1% of the within-residential college and 32% of the
between-residential college variation in students’ capacity for lifelong learning. I frame
my discussion of how the findings of the within-residential college (level one) model
relate to existing research using the blocks of variables I entered as a guide.

Sociodemographic Attributes

Although the addition of the sociodemographic attributes to the model explained
little of the within-residential college variation in students’ inclination to inquire and
capacity for lifelong learning (5.1% and .3% respectively), several of the significant
associations between variables are worth discussing. First, it is noteworthy that Asian-
American students scored on average 3.28 points lower on the inclination to inquire scale
than their White peers despite there being no difference in their scores on the capacity for

lifelong learning scale. One possible explanation for their lower average scores is that the

113

scale does not provide a true measure of Asian-American students’ inclination to inquire.

In recent years, researchers have raised questions about the applicability of
psychological models that were developed and tested using primarily White and often
male students to women and students of color and have suggested alternative models
(Cross & Fhagen-Smith, 2005; Josselson, 1987, 1996; Kim, 2001; Kodarna, McEwen,
Liang, & Lee, 2001; Torres, 2003). In their critique and proposed alternative to
Chickering’s model of psychosocial identity development, Kodarna, et al. (2001)
contended that two external domains are necessary in a model that is inclusive of Asian-
American students. These domains include the tensions between Western values and
racism from the US and Asian values for family and community. The researchers argued
that underlying many existing Western models are individualistic values, which promote
independence and self-exploration, whereas underlying traditional Asian familial and
cultural norms are collectivist values, which promote placing the needs of others above
one’s self and striving toward interdependence (Kodama, et al., 2001).

In examining the scale used to measure students’ inclination to inquire with an
eye toward why Asian-American students might have scored lower than White students, I
found several items that when set in the context of an individualistic versus collectivist
mindset, may be biased toward an individualistic outlook on valuing thinking. Items such
as ‘The idea of relying on thought to make my way to the top appeals to me,’ and ‘I like
to have the responsibility of handling a situation that requires a lot of thinking’ may have
been construed by some students as necessitating acting in a way that would set
themselves apart from their peers, a notion that would likely be met with more

apprehension in a collectivist culture.

114

In addition to Asian-American students scoring lower than their White peers,
female students scored on average 1.629 points lower than males. The finding is
especially interesting when one considers that the ‘need for cognition’ scale, which was
used to measure students’ inclination to inquire was tested for gender-bias when it was
developed (Cacioppo & Petty, 1982). However, the tests were done almost three decades
ago, and student demographics have changed considerably since that time with more
diverse students attending college for myriad reasons. Like the argument made for why
Asian-American students as a group scored lower on the scale, a similar argument can be
made for why female students scored lower than males.

Several researchers have argued that gender and gender role socialization may
play a part in cognitive (Belenky, Clinchy, Goldberger, & Tarule, 1986), epistemological
(Baxter Magolda, 1992), moral (Gilligan, 1993), and identity development (Josselson,
1987) of women and that existing theories and models of development have historically
overlooked these differences. Researchers posit that women often take a more relational,
interpersonal approach to knowledge acquisition than their male peers, who often prefer a
more impersonal, debate centered approach (Baxter Magolda, 1992; Belenky, et al.,
1986). If these arguments are extended to women students’ value for learning and
expression of their enjoyment of thinking, it is possible that some of the scale items are
emphasizing a pattern of development that is slightly more appealing to most men than
most women.

Another possible explanation for the lower average scores of Asian-American and
women students may be because both of these groups of students have historically had

less access to higher education than their White and male peers (Thelin, 2004). As a

115

result, these students may be socialized by family and society to focus more on the
outcome of their college years in terms of degree attainment for their future success as
opposed to developing a value for thinking. I will discuss the implications of this
possibility later in this chapter.

A third interesting finding was that international students scored on average 4.5
points higher on the inclination to inquire scale than their White peers, even after
accounting for such factors as family income and education, motivation, and degree
aspirations. Although it is difficult to interpret the exact meaning of this finding because
all international students are grouped together regardless of country of origin, it may be
most indicative of the reality that international students often must overcome more
obstacles to get into college in the United States than their White (domestic) peers, and as
a result, may have a more pronounced or thought-out value for learning than their peers.
This finding aligns with Pizzolato’s (2003) argument that high-risk students are often
more self-authored than their peers because of the additional obstacles they have
overcome in order to arrive at college.

In examining the association of sociodemographic variables and students’
capacity for lifelong learning, two noteworthy results included the fact that
Hispanic/Latino students scored 1.355 points higher, and Race-Other students (which was
a category that combined Native Hawaiian and Native American students) scored 4.291
points higher on average than their White peers on the capacity for lifelong learning
scale, even after accounting for potentially confounding variables. These results may be
indicative again that these students have a clear sense of why they are in college

(Pizzolato, 1993), and thus are making greater gains than their white peers, or

116

alternatively it could be that these students as a group entered with a lower benchmark
from which to compare their progress than their peers, and thus are making greater gains.

Finally, although the sociodemographic variables did not explain a large amount
of variation within residential college environments, these variables did explain a
considerable amount of the variation between residential college environments (50.6% of
between-residential college variation in students’ inclination to inquire and 32% of
between-residential college variation in students’ capacity for lifelong learning). These
findings indicate that many of the significant sociodemographic characteristics that are
associated with students’ inclination to inquire and capacity for lifelong learning are
clustered in certain residential college environments. The findings support the argument
that if residential college research is aggregated to the environmental level, ignoring
individual/within-environment variation, there is a risk of overstating the environmental
influence, when in fact the findings may stem from the characteristics of individual
students who are attracted to and clustered in specific environments.

Pre-college Characteristics

The addition of the pre-college characteristics (which included parents’ education
and income levels, and students’ high school grade point average) explained little of the
within-college and between-college variation in students’ inclination to inquire and
capacity for lifelong learning. The only marginally significant association was the
negative association between students’ high school grade point average and their capacity
for lifelong learning score. This finding seems antithetical to conventional wisdom,
however, Seifert (2006) found similar results in her study of the effect of major on 21St

century competency development. Seifert argued that in self-report studies of college

117

impact, students who come into college with higher grades may have a higher benchmark
from which to compare their progress, and thus may report lower gains. In the case of the
current study, that rationale makes sense, as the capacity for lifelong learning index
measures students’ self-reported gains in areas associated with the skills necessary for
lifelong learning. It could be that students who had a higher high school grade point
average did not believe they had made as much progress in developing these skills as
their counterparts who entered the residential college environment with lower high school
grade point averages.

The non-significant association between the parental income and education
variables and students’ inclination to inquire and capacity for lifelong learning was
somewhat surprising considering that Gorard and Selwyn (2005) found both of these
attributes to be predictive of a commitment to lifelong learning, which they measured by
one’s enrollment in formal education during the adult years. However, because Gorard
and Selwyn’s study was conducted in the United Kingdom and used a different outcome
variable, the dissimilar findings may be explained by contextual differences in
postsecondary education between the US and United Kingdom. Access is much more
limited in the United Kingdom and thus parental income and education are likely more
important factors to continuing one’s education.

Motivation, College Experiences, and Degree Aspirations

The addition of the motivation, college experience, and degree aspiration
variables to the model explained the most variation in students’ inclination to inquire and
capacity for lifelong learning of any of the level-one blocks of variables. Although the

positive association between students’ motivation and their inclination to inquire and

118

capacity for lifelong learning was small, its significance supports McCombs (1991)
assertion that to promote lifelong learning, educational settings should develop
supportive climates conducive to cultivating personal relationships, a sense of control,
and personal choice in students’ learning process. The positive association also extends
the reach of self-determination theory to the development of lifelong learners. However,
the small significance of the association merits further exploration to determine whether
the instrument is effective in capturing students’ sense of autonomy, competence, and
relatedness and also if the association deepens over time.

When taken together, the significant positive association between students’
motivation, aspirations to obtain more than a bachelor’s degree, and their inclination to
inquire and capacity for lifelong learning support Hayek and Kuh’s (1998) supposition
that students’ motivation to learn may play a role in the development of their capacity for
lifelong learning, and thus should be accounted for when examining factors that influence
students’ capacity for lifelong learning. Furthermore, although small, the results of the
varying slope for the association between students’ desire to obtain more than a
bachelor’s degree and their inclination to inquire suggest that this association may be
stronger in certain circumstances, a finding that ought to be examined further in future
studies.

The positive association between the number of years students lived in the
residential college and their inclination to inquire and capacity for lifelong learning is
indicative that these outcomes may deepen over time as students become more integrated
into the collegiate setting. However, as I alluded to in chapter four, it remains unclear

whether the positive association between years spent in the residential college and the

119

outcome variables is a result of students’ interaction with their residential college
environment, or rather a product of their maturation. A longitudinal study of students
who leave the residential college setting and those who stay would help to tease out these
findings.

Individual Liberal Arts Experiences

Admittedly, I was most interested in the addition of the liberal arts experiences
variables (good teaching and high quality interactions with faculty; academic challenge
and high expectations; diversity experiences and quality interactions with peers) at level-
two of the model because they were more indicative of the potential influence of the
environment on the outcome variables. That said, the addition of the group-mean
centered liberal arts experience variables at level one enabled me to ascertain the
association between students’ individual experiences with these variables and their
inclination to inquire and capacity for lifelong learning, which I could then compare to
the environment (level-two) results. In addition, the addition of these variables at level-
one served to control for students’ individual beliefs about the environment, allowing me
to hone in on the potential environmental impact. I discuss the associations between the
individual liberal arts experience variables briefly here, reserving some of the discussion
for their entry into the model at level-two.

When entered into the model at level one, all of the individual liberal arts
experience variables were at least marginally statistically significant in their association
with students’ inclination to inquire. However, none of these associations was
particularly large. The most significant positive associations between students’

inclination to inquire and the individual liberal arts experiences variables were the

120

‘classroom practices’ component of the ‘good teaching and high quality interactions with
faculty’ variable and the ‘diversity experiences’ variable. These results, in part, support
research conducted by Blaich. and Wise (2008) who in their summary of the findings
from the first year of the Wabash National Study of Liberal Arts Education reported that
good teaching and diversity experiences were both positively related to growth in
students’ need for cognition (or inclination to inquire) over the course of one year.
However, Blaich and Wise also found Academic Challenge to be related to growth in
students’ need for cognition, which was only marginally significant in the current study.
Perhaps the most interesting finding was the statistically significant negative
association between students’ ‘high quality interaction with peers’ and their inclination to
inquire. The negative association was somewhat surprising given the weight of evidence
that peers and co-curricular involvement play a pivotal role in positively influencing
student learning (Astin, 1984; Kuh, Schuh, Whitt, & Associates, 1991; Pascarella &
Terenzini, 1991, 2005). However, the finding extends that of Pike, Schroeder, and Berry
(1997) who found that although living-learning community participation may heighten
social integration and institutional commitment, it does little to enhance academic
achievement, indicating that the role of peers might be more indirect. More research is
warranted to tease out whether the negative association between high quality interactions
with peers and students’ inclination to inquire is an indication that deep and meaningful
relationships with peers actually detract from students’ inclination to inquire, or that
students who value deep and meaningful relationships with peers might place that value

over their value for learning. It is also possible that the scale used in the current study is

121

not reflecting accurately the association between peer relationships and students’
inclination to inquire.

Similar to the small associations with students’ inclination to inquire, several of
the individual liberal arts experiences had a small but significant association with
students’ capacity for lifelong learning. Interestingly, aside from diversity experiences,
which had a significant positive association with both students’ inclination to inquire and
capacity for lifelong learning, the most significant associations between the liberal arts
variables and students’ capacity for lifelong learning were different from those that were
significant with students’ inclination to inquire. Whereas the ‘out-of-class interactions
with faculty’ component of the ‘good teaching and high quality interactions with faculty’
variable had a marginally significant association with students’ inclination to inquire, it
was significant at the p<.01 level for students’ capacity for lifelong learning. This finding
contradicts that of Hayek and Kuh (1999), who found no relationship between faculty-
student interaction and students’ capacity for lifelong learning. The difference in findings
is potentially explained by differences in the two samples. Existing research suggests that
students in residential colleges or living-learning communities have more opportunities to
interact with faculty than other students (Cox & Orehovec, 2007; Pike, 1999; Smith,
1994), which may have contributed to the significant positive association between out-of-
class interactions with faculty and students’ capacity for lifelong learning.

The differing significant associations between the individual liberal arts
experience variables and the two outcome variables allude to the possibility that different
environmental features are important in eliciting and supporting an inclination to inquire

(value) and capacity for lifelong learning (skill set). The entry of the variables at level

122

two of the model provides additional substantiation of this claim. I turn to my discussion
of the between-residential college model now.
Between-Residential College Model — The Association of Residential College
Environments with Students ’ Inclination to Inquire and Capacity for Lifelong Learning
In fitting the between-residential college (level-two) models, I was interested in
determining the extent to which various aspects of the residential college environment
were associated with the outcome variables. Specifically, I was curious about how an
overall ethos marked by the liberal arts experiences (good teaching and high quality
interactions with faculty; academic challenge and high expectations; diversity
experiences; and quality interactions with peers) and disciplinary focus were related to
students’ inclination to inquire or capacity for lifelong learning. Because there was
insufficient variation in the disciplinary foci of the residential colleges included in the
sample to run the disciplinary focus component of the level-two model, my discussion in
this section focuses solely on the addition of the environmental liberal arts to the models.
By adding each of the environmental liberal arts variables to level-two of the
models in turn before adding them all together, 1 determined the unique influence of each
in explaining the between-residential college variance in the outcome variables. With the
addition of the final model (which included all of the environmental liberal arts
variables), I determined which associations remained significant despite the high
correlations between the environmental (group mean) liberal arts experience variables
(see Appendix J for level-two correlations). I begin the discussion of the associations
between the outcome variables and the environmental liberal arts variables by focusing

on the inclination to inquire models. I then turn to the capacity for lifelong learning

123

models. Throughout the discussion, I offer potential explanations for the findings and
connect them to existing literature.
Inclination to Inquire

The between-residential college model explained 82.2% of the between-
residential college variation in students’ inclination to inquire (32% of which was
explained by the environmental liberal arts variables). Although three of the
environmental liberal arts variables were significantly associated with students’
inclination to inquire when entered individually into the level-two model (including both
the ‘classroom practices’ and ‘out-of-class interactions with faculty’ components of the
high quality interactions with faculty scale and the ‘academic challenge and high
expectations’ variable), only the ‘academic challenge and high expectations’ variable
remained significant at the p<.05 level in the final model. The association between the
environmental ‘academic challenge and high expectations’ variable and students’
inclination to inquire was modest (.214 points), but noteworthy given that individuals’
reactions to the liberal arts experiences were controlled for at level-one of the model.

There is extensive research supporting the notion that an environment marked by
academic challenge and high expectations would enhance students’ inclination to inquire.
Researchers examining undergraduate student success have found that students who are
challenged in their academic environment report higher levels of development in a
variety of areas (Astin, 1993; Chickering & Gamson, 1991; Cruce, Wolniak, Seifert,
Pascarella, & Blaich, 2005; Kuh, Schuh, Whitt, & Associates, 1991; Pascarella &
Terenzini, 1991, 2005; Seifert, 2006). It is worth mentioning that even in a residential

college environment, which arguably is deliberate in its attempt to provide students’ with

124

a comprehensive undergraduate experience, there is still variation in students’
perceptions of the level of challenge provided by the environment. Faculty and
administrators who desire to engage students more deeply in the life of the mind might
focus their attention on increasing the level of challenge and the expectations they have
of their students.

Also interesting to note were the differences in the results of the group-mean
centered individual liberal arts variables entered at level one of the model and an
environment marked by these variables (as indicated by the group-mean) entered at level
two. The fact that the ‘out-of-class interactions with faculty’ variable, ‘diversity
experiences’ variable, and ‘quality interactions with peers’ variable were significant at
level one of the model, but not in the final level-two model (which included all of the
level-two liberal arts variables) may indicate that these factors are most effective in
influencing students’ inclination to inquire only when they connected directly with
students. On the other hand, the ‘academic challenge and high expectations’ variable,
which was significant in both the level-one and level-two final models, and the
‘classroom practices’ variable, which was significant in the level-one model and
marginally significant in the level-two model, may have a more distal impact on students’
inclination to inquire, potentially creating an overall ethos that deepens students’
inclination to inquire even when it does not directly engage a student. That rationale
holds when one considers that the ‘classroom practices’ and ‘academic challenge and
high expectations’ variables are less about relationships and more about students’ general

sense of their environment. These findings are supported by Mayhew, Wolniak, and

Pascarella’s (2008) findings that instruction-based practices, which were those that

p 125

promoted active learning, discussion, multiple viewpoints, and self-reflection had an
indirect effect on encouraging students’ inclination to inquire (a term they labeled need
for cognition, which was true to the original name of the scale).

Also important to examine was the non-significant associations between students’
inclination to inquire and environments marked by diversity experiences and quality
interactions with peers. Neither of these two variables was significant when entered into
the level-two model on their own, nor were they significant in the final model, which
included all of the level two variables. Mayhew, Wolniak, and Pascarella (2008) found
that negative interactions with diverse peers hampered growth in students’ inclination to
inquire. The differences in findings could be attributable to differences in the diversity
scales, as the scale used in the current study focuses more on the frequency of
interactions with diverse peers and ideas as opposed to the quality of those interactions. It
is possible that by focusing on how an environment marked by positive or negative
interactions with diverse peers, as opposed to frequency of these interactions, I would
find more of an association between an environment marked by students’ diversity
interactions and their inclination to inquire.

As with the negative association between high quality interactions with peers at
level-one of the model, the non-significant association between students’ inclination to
inquire and an environment marked by high quality interactions with their peers at level-
two of the model was curious given the evidence that peers play a role in engaging
students’ in learning, especially in living-learning communities (Wawrzynski, Jessup-
Anger, Helman, Stolz, & Beaulieu, in press). More research is warranted to determine if

there are particular interactions with peers that are not captured in the ‘high quality

126

interactions with peers’ scale, which might be more associated with a deep inclination to
inquire.
Capacity for Lifelong Learning

Turning now to the association of the environmental liberal arts variables and
students’ capacity for lifelong learning, the level-two model explained 77.5% of the
between-residential college variation in students’ capacity for lifelong learning (45.4% of
which was explained by the environmental liberal arts variables). With the exception of
the classroom practices component of good teaching and high quality interactions with
faculty, all of the environmental liberal arts experience variables were significant when
entered on their own into the level-two model. However, in the final model, only the out-
of-class interactions with faculty remained significant at the p<.05 level. This finding
contradicts that of Hayek and Kuh (1999) who found that faculty-student interaction had
little to no effect on deepening students’ capacity for lifelong learning. As I argued earlier
when discussing the individual level variables, the difference in findings is potentially
explained by differences in the two samples, as students in the current study might have
had more opportunity to interact with faculty, especially outside of class because of their
participation in the residential college.

The marginally significant negative association between ‘classroom practices’ and
students’ capacity for lifelong learning was puzzling. However, examining the scale
items of the ‘classroom practices’ component in the context of their association to
students’ capacity for lifelong learning raised questions about the fact these items were
mostly focused on instructor preparation, organization, and clarity. A classroom

environment marked by these standards is unlikely to translate directly into deepening

127

students’ ability to learn to learn and apply concepts independently, unless coupled with
developing a mastery orientation for learning (Dweck & Leggett, 1988), deepening
students’ appreciation for why the material is important (Brophy, 2004), and creating
opportunities for active engagement with material (Chickering & Gamson, 1987, 1991).
In fact, some of the practices might be associated with an instructor maintaining control
over the classroom as opposed to developing autonomy in students (McCaslin & Good,
1992)

The disconnect between these instructional practices and students’ capacity for
lifelong learning may also explain the significant negative cross-level interaction
(Research Question Five) between students’ motivation and an environment marked by
these classroom practices. Although the cross-level interaction between the two variables
was so small as to not have practical significance, it adds credence to questioning
whether a different measure of classroom practices which might take into account
practices more geared toward developing self-regulated learners might more effectively
capture the relationship between classroom practices and students’ capacity for lifelong
learning. I

Similar to the changes in the significance of the association of the liberal arts
experience and students’ inclination to inquire at level-one and level-two of the model,
there were also changes in the significance of these variables and students’ capacity for
lifelong learning at the different levels of the model. Specifically, only the ‘out-of class
interactions with faculty’ component of the ‘good teaching and hi gh-quality interactions
with faculty’ variable remained significant in the final models at both level one and level

two. The ‘academic challenge and high expectations’ and ‘diversity experiences’

128

variables, while significant at level-one of the model, were not significant at level two.
These findings may indicate that for this particular construct, which deals with building
skills as opposed to deepening a value, the relationships students develop with faculty are
important in creating an environmental ethos that supports building a capacity for lifelong
learning. Alternatively, it may suggest that students who are already committed to
deepening their capacity for lifelong learning are especially likely to interact with faculty
outside of the classroom.

Revisiting Moos ’5 Social Ecological Framework

The findings of the current study illustrate the importance of using an ecological
approach to examining the impact of sub-environments on students, especially in
residential colleges and living-learning programs, as students bring many socio-
demographic and motivational characteristics into these environments that are
instrumental to their learning and development outcomes. In this section, I frame the
findings using Moos’s model as a guide, critique the model in light of the findings, and
discuss where the model was particularly useful. Figure 5.1 provides a visual of Moos’s
model, including an illustration of how my research questions pertained to each aspect of
the model.

I began my investigation by examining whether there was variation in students’
inclination to inquire and capacity for lifelong learning across residential college
environments (Research question 1). By conducting the one-way AN OVA with random
effects, I partitioned the variance in students’ inclination to inquire and capacity for
lifelong learning into within-residential college and between-residential college

components.

129

 

 

 

Res. Question 4

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

\
\ \ \ Res. Question 3
\ l‘
‘ / I \\
/ I \ \
. A I 1 \ t
\\ system V fl
\\ Res. Question 1 V
\\ v Student
\ Stability
‘ Cognitiv. Activation mm B .t ‘
| Ads tlon
Appraisal :5 °' Arousa d 3:14:22.)an : Change
‘\ A
Personal \\ : l‘ 1
system \ l / / / A
4 \ 5 ’ /
Res. Question 1 4 \\ 1 7 1 / /
A f \ l / /
““““ \ l ///
~ ‘ g &

 

Res. Question 2

 

 

 

Figure 5.1: Relationship between my Research Questions and Moos (1979) Model of
the Relationship between Environmental and Personal Variables and Student
Stability and Change

The partitioning of variance provided a foundation for examining the personal and
environmental systems and their association with students’ appraisal of their
environment, motivation, and inclination to inquire and capacity for lifelong learning,
each of which I detail below.

The Environmental System. To assess the environmental system (research
question 4), I used the group mean of the liberal arts experience variables for each
residential college. Having already controlled for students’ individual perceptions
(cognitive appraisal) of the residential college environment at level-one of the model, in

assessing the environmental system, I was interested in the overall ethos of the

130

environment specifically related to the liberal arts experiences. The liberal arts
experiences focused most heavily on what Moos would call the organizational factors,
human aggregate, and social climate of the residential college setting (these were
described in more detail in chapter one). The physical setting was more implicit, and may
have affected students’ assessment of their environment in terms of studying with peers,
involvement in activities, and interacting with faculty.

In addition to assessing how an overall ethos of the environment related to the
liberal arts was associated with students’ inclination to inquire and capacity for lifelong
learning, Moos’s model illustrated the potential that the degree of the association between
students’ pre-college characteristics, motivation, and individual liberal arts experiences
and the outcome variables may have varied across residential colleges (research question
3). Furthermore, it demonstrated the potential importance of exploring the cross-level
interactions between the personal system and the environmental system (research
question 5), as the environmental system may have had an indirect association with
students’ inclination to inquire and capacity for lifelong learning by way of their
motivation and degree aspirations.

For the most part there was little practical significance in the degree of variation
in the association between students’ pre-college characteristics, motivation, and
individual liberal arts experiences “and the outcome variables by residential college
context (research question 3). Perhaps if there were more variation in the types of
communities examined (e. g., living-learning communities in addition to residential
colleges), the degree of these associations would have varied more widely across

residential colleges. Likewise, there was little practical significance in the cross-level

131

interactions between the environmental variables and the motivation variables, as the
only significant cross-level interaction was the negative association between students’
motivation and the ‘classroom practices’ component of the good teaching and high
quality interactions with faculty scale (discussed in detail in chapter 4). Again, the
minimal cross level associations were perhaps a result of the fact that there was not a
tremendous amount of variation across the residential college contexts. These
relationships seem more important to consider when the environmental contexts are more
distinct.

The Personal System. To assess the personal system, I built within-residential
college/individual (level-one) models examining the association between students’
sociodemographic and pre-college characteristics and the outcome variables. Moos
explained that such characteristics may influence the way that students may perceive their
environments and respond to them. As I discussed in the findings, the first two blocks of
variables I entered into the level-one model explained little of the variation within
residential college environments. Moos’ suggestion that additional personal
characteristics, including “attitudes[and] expectations” may provide additional insight
into the personal system, and should be considered when investigating the roles of the
personal system in future studies.

Mediating Factors. I added two blocks of variables representing Moos’s
mediating factors. The first was students’ cognitive appraisal of their environment, which
Moos (1979) described as students’ evaluation of their environment as being “potentially
harmful, beneficial, or irrelevant” (p.11). I deemed these the individual liberal arts

experience variables and explored their association with students’ inclination to inquire

132

and capacity for lifelong learning at level one of the model. As I mentioned in my
discussion of the findings, the addition of students’ cognitive appraisal of their
environment enabled me to ascertain the association between students’ individual
experiences with the liberalarts variables and their inclination to inquire and capacity for
lifelong learning, which I could then compare to the environment (level-two) results.
Furthermore, the addition of these variables at level-one served to control for students’
individual beliefs about the environment, allowing me to hone in on the environmental
system.

The second mediating factor represented by the model was students’ activation or
arousal, or what I deemed students’ motivation, degree aspirations, and college
experiences. The addition of these variables to the model explained the most within-
residential college variation in students’ inclination to inquire and capacity for lifelong
learning and was clearly an important consideration of the personal system. These
variables were mediated to a lesser extent by the environmental system, as only the
degree of association between students’ desire to obtain more than a bachelor’s degree
varied by residential context (illustrated by research question 3, and also mentioned
above).

" Eflorts at Adaptation. Moos explained that common transitions and every day
situations demand coping and adaptation responses, and that coping and adaptation are
not only mediators of outcomes, but are also outcomes themselves. My interest was in
examining students’ inclination to inquire and capacity for lifelong leaming in light of
their personal characteristics and residential college environments. As a result, in the

current study, the outcome variables (students’ inclination to inquire and capacity for

133

lifelong learning) are considered students’ efforts at adapting to the collegiate
environment and preparing for lifelong learning. As such, they are associated with the
environmental and personal system and mediated by students’ cognitive appraisal and
motivation.

Student Stability and Change. Assessing the stability or change of specific
students’ inclination to inquire and capacity for lifelong learning was outside the scope of
the current study. However, by including a cross-sectional sample, I was able to ascertain
that there was a statistically significant association between students’ inclination to
inquire and capacity for lifelong learning and the number of years they lived in a
residential college. However, as I mentioned in my discussion, more research is needed to
determine whether the association is as a result of the residential college environment or
merely and indication of students’ maturation. Regardless, the findings illustrate that the
students’ inclination to inquire and capacity for lifelong learning may deepen as they
mature.

In summary, Moos’s model provided a useful framework for examining the
relationship between the environmental system, personal system, and students’
inclination to inquire and capacity for lifelong learning. The model allowed for the
consideration of a confluence of personal and environmental factors to be associated with
students’ inclination to inquire and capacity for lifelong learning. In addition, the
inclusion of mediating factors (including activation or arousal and cognitive appraisal)
was important as well, as there were statistically significant associations between these
factors and students’ inclination to inquire and capacity for lifelong learning. Perhaps less

important to the current study was the consideration that the degree of variation in the

134

outcome variables may have changed across environmental context and the exploration
of cross-level interaction. However, if the residential college environments were more
distinct from one another (as they often are in living-learning community research) these
considerations may have been more important. Additional studies could also make better
use of Moos’s focus on the stability or change of these outcomes over time by taking a
longitudinal approach. With my discussion of the findings and their relationship to
existing literature and Moos’s model complete, I turn now to the implications of the
study.
Implications for Theory, Research, Practice, and Policy

In light of the findings of the current study and the relationship of these findings
with existing research, I offer implications for theory, research, practice, and policy. I
also detail suggestions for future research.
Implications for Theory

The findings of the current study offer an important contribution to higher
education theory by reinforcing the calls of researchers (Pascarella & Terenzini, 2005;
Renn, 2003, 2004; Renn & Arnold, 2003; Strange & Banning, 2001) to employ
ecological models to understand student outcomes, especially when seeking to
understand the influence of collegiate sub-environments on student learning and
development. As was illustrated by partitioning the variance in students’ inclination to
inquire and capacity for lifelong learning into within-residential college and between-
residential college components, there was a greater amount of variation within residential
college environments than between them. F urtherrnore, almost half of the variation in the

outcome variables between residential college environments was attributable to

135

individual sociodemographic characteristics that were clustered within certain residential
college environments. If the different levels of analysis were not accounted for, the
environmental influence may have been overstated, because of these clustered
individuals. Likewise, by examining solely the characteristics that students bring to their
sub-environment, researchers miss an opportunity to explore and make meaning of how
different interventions are associated with deepening student outcomes. The ecological
approach enables researchers to have the best of both worlds, in that they can examine
and explore personal and environmental considerations together, acknowledging that
these components are constantly interacting and informing one another. By employing an
ecological approach, researchers will gain a better understanding of the influence of
students and their environments in promoting student learning and development.
Implications for Research

The contributions of the current study to higher education research are also worth
noting. First, the current study continues the departure away from much of the early
research on living-learning communities and residential colleges, which sought to I
ascertain whether these environments were more effective in promoting student outcomes
than no intervention (i.e., Pasque & Murphy, 2005; Pike et al., 1997; Pike, 1999). The
next generation of living-learning community and residential college research focuses
instead on ascertaining which aspects of the residential college environment are most
effective in promoting student outcomes and why. In addition, the study is the first to
examine the effectiveness of these environments in promoting the liberal arts outcomes
they purport to emulate. In part, the current study supports the notion that the more

closely aligned a residential college is with promoting liberal arts experiences (such as

136

academic challenge and high expectations and out-of-class interactions with faculty) the
more likely students are to report higher liberal arts outcomes (in this case, a deeper
inclination to inquire and capacity for lifelong learning). However, the current study also
raises additional questions about whether the emphasis of residential college
environments in promoting deep peer relationships actually translates to deeper student
learning. More research is needed to understand the role of peers in these settings beyond
that they provide students’ a sense of social integration.

A second research implication advanced by the current study is the conceptual
distinction between students’ inclination to inquire, or their value for learning, and their
capacity for lifelong learning. The results of the current study indicate that these two
outcomes are distinct and associated with different sociodemographic characteristics and
liberal arts experiences. Thus, care should be taken to define them accurately when
conducting research. As I discussed in Chapter 2, existing research using the ‘need for
cognition’ scale has used different terminology to describe the outcome associated with
the scale, including ‘need for cognition’ (Blaich & Wise, 2008; Cacioppo & Petty, 1982),
‘inclination to inquire and lifelong learning’ (Seifert, et al., 2008), and ‘lifelong learning
orientation’ (Mayhew, Wolniak, & Pascarella, 2008). With the results of the current study
illustrating that there are different sociodemo graphic characteristics, experiences, and
environments associated with students’ inclination to inquire (value) and capacity for
lifelong learning (skill set), it is vital that researchers reflect these differences when
defining their constructs how they relate to a deepened value for learning and capacity for

lifelong learning.

137

In addition to taking care when defining constructs related to students’ value for
learning and capacity for lifelong learning, another implication for research is that the
results of the current study raise questions about the ‘need for cognition’ scale itself,
which was used to measure students’ inclination to inquire. The factor analysis (discussed
in chapter 3) indicated that for the current sample, several factor loadings of the scale
were surprisingly low (.153, .169, and .180), especially when one considers how much
empirical testing of the scale has been done on the scale (Cacioppo & Petty, 1982;
Cacioppo, Petty, Feinstein, & Jarvis, 1996). However, in examining those items that were
low (and which I ultimately opted to exclude), namely, “I feel relief rather than
satisfaction after completing a task that required a lot of mental thought,” I only think as
hard as I have to,” and, “I prefer to think about small daily projects to long-term ones
(recoded),” I wondered whether students’ answers to these questions were reflective more
of their attempts to cope with a demanding and intellectually-pressured collegiate
environment and less about their enjoyment and value for thinking. In fact, one of the
items, “I prefer to think about small daily projects to long-term ones (recoded),” is often
included in time management advice that college and university administrators give to
students (see for example, Campus Calm website, n. d.; George Washington University
Academic Success Center, n. d.; University of Kansas Counseling and Psychological
Services Website, n. (1). Another issue raised by the results of the current study about the
need for cognition scale is whether the scale appropriately captures Asian-American and
women students’ enjoyment of thinking.

The questions raised about potential bias in the need for cognition scale, coupled

with the low factor loadings of several items are especially alarming when one considers

138

that the scale is currently widely used in the Wabash National Study of Liberal Arts
Education to conduct research and assessment on the extent to which a variety of
postsecondary institutions are successful in developing students’ inclination to inquire.
More research is warranted to determine whether the questions raised about the scale in
the current study extend to the findings of the Wabash study, and if so, whether caution
should be taken in interpreting the results of that study.

Implications for Practice — Prospective Students, Administrators, and Faculty

The results of the current study provide parents and prospective students with
additional information to consider when making the decision to enter a residential
college. First, the results illustrate that there is substantially more variation in students’
inclination to inquire and capacity for lifelong learning within residential college
environments than between them. The implication of this finding is that in addition to
choosing a residential college based upon its educational offerings, prospective students
should get to know students living in the residential college and determine whether these
students have a similar level of motivation and eagerness to learn as the prospective
student.

In addition, as parents and prospective students weigh the educational offerings of
aresidential college, the results of the current study demonstrate the importance of
looking for an environment that has is a scholarly ambiance with high expectations of
students. Again, in addition to examining course offerings, prospective students may be
able to gauge the atmosphere by talking to students living in the residential college in

order to get a sense of their perceptions of the environment.

139

Finally, prospective students should meet the faculty in the residential college and
ask them about opportunities for interaction outside of class. These opportunities might
include working on research projects with faculty, participating in service-learning
projects, or engaging in some other type of co-curricular learning opportunity. By having
semi-structured activities to participate in with faculty, students may engage with faculty
in a meaningful way.

In addition to providing important information to students and parents, the results
offer an important contribution to higher education administrative and faculty practice,
both within residential colleges and within postsecondary institutions more broadly.
There are several findings that offer clear evidence of areas in which concentrated
attention could potentially enhance students’ inclination to inquire and capacity for
lifelong learning, especially in the realms of student advising and support and faculty and
student interactions.

Student Advising and Support

The current study offers evidence that residential college environments, to the
extent that they provide a challenging academic atmosphere and offer out-of-class
interactions with faculty, are associated with deepening students’ inclination to inquire
and capacity for lifelong learning. Advisers and others charged with helping students
navigate their academic decision-making can use the results of the current study to assist
students in developing a more nuanced understanding of residence college environments.
For example, instead of offering a general recommendation that students join a residential
college, advisors may give specific advice about how to make the most of a residential

college experience, getting involved in activities that foster deeper levels of student and

140

faculty interaction or engaging in activities that provide a deeper level of intellectual
challenge. In addition, if an institution offers multiple residential colleges, advisors can
assist students in finding one that is the right fit, especially if the student has the desire to
deepen their value for learning and capacity for lifelong learning. Ideally, students should
be able to weigh the specific programmatic features of a residential college and evaluate
how they align with their goals for undergraduate education.

The results of the current study also highlight that Asian-American students and
women students have a lesser inclination to inquire than their white and male peers.
Although more research is needed to determine whether these findings are a reflection of
a biased scale or truly a reflection that these two groups as a whole tend to engage in and
enjoy thinking less, it raises questions about how the messages we convey to students
about the purpose of a postsecondary education facilitate or impede their value for
learning. When faculty members, advisors, and other university personnel appeal to
historically underrepresented students by encouraging them to partake in activities for
strictly utilitarian reasons (i.e., serve as an intern, volunteer, or study abroad solely to
better their prospects for future employment), are there ramifications for what students
take away from that activity? Are these messages more often conveyed toward students
who have historically been excluded from postsecondary education? Brophy (2004)
suggests that there is value in helping “learners begin to see the value in potential
learning opportunities that they have not yet come to appreciate (and might never come to
appreciate) on their own” (p. 258). In order to assist all students (and specifically Asian-
American and women students) in deepening their value and engagement in thinking and

learning, faculty and administrators must reflect on how educational practices and

141

activities they encourage students to get involved in promote a value for learning, and

then work to connect those deeper values to students’ beliefs.

Faculty and Student Interaction

In addition to offering implications for student advising and support, the results of
the current study also have significance for faculty and student interaction, especially
related to motivation, creation an environment marked by academic challenge and high
expectations, and out-of-class interactions with faculty.

Although the association between students’ motivation and their inclination to
inquire and capacity for lifelong learning was small, it was significant and provides
evidence that the more students’ feel a sense of autonomy, competence, and relatedness
in their educational pursuits, the deeper their inclination to inquire and capacity for
lifelong learning. Existing motivation and teaching and learning literature sheds light on
how faculty might deepen students’ beliefs in these areas, including by providing
students’ with opportunities to take charge of their educational pursuits, coupled with
enough support and direction to do so (Baxter Magolda, 2004; Brophy, 2004), providing
appropriately challenging material so that students feel accomplished yet continually
strive for excellence (Brophy, 2004; Chickering & Gamson, 1987, 1991; Svinicki, 2004),
and by creating a classroom environment that fosters collaboration and encourages risk
taking (Brophy, 2004; Svinicki, 2004).

The results of the current study clearly indicate that an environment that is
marked by academic challenge and high expectations is associated with a deepened

inclination to inquire. As I indicated previously in this chapter, there is substantial

142

research supporting the positive influence a challenging academic environment has on a
variety of student outcomes. Although research universities as a whole might be too large
and have too many competing missions to provide a coherent and appropriately narrow
message promoting academic excellence and high expectations, a residential college
environment is more suited to maintaining a consistent, scholarly message across all
aspects of the college. The first step in this process may be to ensure that there is a
consistent and coherent message to convey, which means that residential colleges must
examine their missions and educational purpose to determine if they promote academic
challenge and high expectations (Kuh, 1999). After the message is determined, it should
be conveyed through all aspects of the residential college, from admissions through
graduation, by faculty, administrators, and students alike.

Whereas an environment marked by academic challenge and high expectations
was indisputably related to students’ inclination to inquire, out-of-class interactions with
faculty were unequivocally associated with students’ capacity for lifelong learning. Even
with individual experiences accounted for at level one of the model, the ‘out—of-class
interactions with faculty’ variable rose to significance at level-two. As I discussed
previously, because the model is not causal, it remains unclear if out-of-class interactions
with faculty deepen students’ capacity for lifelong learning or rather if students’ who
have a deeper capacity for lifelong learning are also inclined to interact with faculty
outside of class. Regardless, it is clear that if the opportunity to interact with faculty
outside of class presents itself, these students are likely to engage, and will likely benefit

from these interactions.

143

Cox and Orehovec (2007) examined faculty-student interactions within a
residential college environment and developed a typology detailing the nature of the
interaction and extent to which it happened. The five types of faculty-student interaction
ranged from disengagement to mentoring, with incidental contact, functional interaction,
and personal interaction defining the middle. The authors argued that even in a residential
college environment, which is marked by an expectation that faculty and students will
interact outside of class, the greatest type of interaction between faculty and students is
disengagement, as often faculty and students have little common ground on which to
build a relationship. The authors suggested examining the cultural norms of the
institution to determine how faculty-student interaction is valued. For example, are
faculty rewarded for connecting with students out of the classroom? Is their sufficient
time built into their positions to enable them to connect with students outside of class?
Are there alternative activities competing for student time? Perhaps most important, are
there common experiences through which faculty and students can connect? Faculty and
administrators seeking to improve liberal arts outcomes in their residential college
environments can examine their organizational structures to determine how deliberately
they promote faculty and student interaction outside the classroom. Adding such
initiatives as an undergraduate research or professorial assistant program may increase
the opportunities for meaningful student-faculty interaction outside of class.
Implications for Policy

Finally, the contributions of the current study to higher education policy are also
worth noting. The results illustrate that residential colleges, to varying degrees of success,

were able to create environments marked by ‘academic challenge and high expectations’

144

and ‘out-of-class interactions with faculty’ and that these qualities were associated with
deepened inclination to inquire and capacity for lifelong learning. Although the
relationship between these environmental variables and the outcome variables was not
causal, it offers specific areas that administrators and faculty might target should they
desire to foster students’ proclivity toward and capacity for lifelong learning. Thus, one
policy implication stemming from the results is that offering residential colleges as a way
to scale down the size and scope of large research universities may be the first step in
creating a more intimate environment. However, to create an environment that is marked
by liberal arts experiences and promotes liberal arts outcomes, administrators and faculty
must also ensure that these environments are challenging and offer ample opportunities
for faculty and student interaction.

Furthermore, given the finding illustrating importance of creating an environment
marked by out-of-class interactions with faculty, a second policy implication stemming
from the current study is that some faculty may be more successful in these environments
than others. Specifically, faculty who see value in engaging students in co-curricular
activities and desire to expand their teaching beyond the boundaries of the classroom may
be especially suited to teaching students in a residential college setting. Those charged
With recruiting faculty to teach in residential colleges should clearly indicate their interest
in attracting faculty with these values to the residential college. They should also evaluate
how they reward these values for promotion and tenure.

Perhaps the most enduring policy question related to residential colleges is: Do
they work? The results of the current study indicate that residential colleges, to varying

degrees of success, do promote environments that are marked by liberal arts experiences.

145

Even after accounting for students’ sociodemographic and background characteristics,
pre-college experiences, motivation, and individual perceptions of liberal arts
experiences, these environments were associated with students who had a deepened
inclination to inquire and capacity for lifelong learning.
Suggestions for Future Research

There is much more research to be done to understand whether and how
residential college environments promote the liberal arts outcomes they purport to
emulate, not the least of which is examining their association with other liberal arts
outcomes. The Center of Inquiry at Wabash College has identified several outcomes
associated with a liberal arts education (Blaich, et al., 2004) and recently launched the
Wabash National Study of Liberal Arts Education to examine how these outcomes are
fostered in liberal arts colleges and other types of institutions. To date, the study has not
examined sub-environments, including residential colleges and living-learning
communities. As these types of interventions continue to gain popularity at large public
research universities, it is important that researchers incorporate them into their study
designs. It is no longer possible to generalize that all students at large research
universities have the same educational experience, as many of them are involved in
initiatives that to varying degrees of success, scale down the campus and provide a more
intimate educational experience.

In addition to focusing on other aspects of the liberal arts, enduring questions
remain related to how residential college environments promote students’ proclivity
toward and capacity for lifelong learning. One of the most pressing questions remaining

is whether the need for cognition scale, which was used to measure students’ inclination

146

to inquire, is effective in capturing students’ value for learning. The low factor loadings I
discovered when analyzing the validity and reliability of the scale, coupled with its
significantly lower association to Asian-American students and women students,
highlight potential bias in the scale that should be investigated in other studies using the
scale.

Another question remaining at the end of the current study is whether there are
more objective measures that might capture students’ capacity for lifelong learning. As I
discussed in the limitation section in chapter three, there are legitimate concems raised
about the use of self-reported data in measuring college impact (Anaya, 1999; Gonyea,
2005; Pace, 1985; Pike, 1995, 1996). Future research could ameliorate this self-report
limitation by including reliable and valid objective measures detailing the gains students’
make in their capacity for lifelong learning.

A final question raised by the results of the current study relates to the negative
association between the level-one ‘high quality interactions with peers’ and students’
inclination to inquire. As I discussed previously in this chapter, given the weight of
evidence suggesting that peers play a pivotal role in learning and development, and the
assumption made by residential colleges and living-learning communities that peers are a
vital part of students’ collegiate experience, more research should be done to tease out the
role of peers in promoting a value for learning in residential college environments.
Because the current study and previous studies attending to the role of peers in
influencing student outcomes are both quantitative (Pike, Schroeder, & Berry, 1997), is
possible that the constructs being used are not accurately reflecting the role of peer

relationships in promoting these outcomes. A qualitative portrayal examining how

147

students’ make meaning of the role of their peers in deepening their value for learning

might tease out the relationship further.

Conclusion

The purpose of the current study was to examine the ways in which students’
sociodemographic characteristics, motivation and college experiences, and residential
college environments were associated with their inclination to inquire and capacity for
lifelong learning. The results support the contention that students’ motivation is
associated with a deepened inclination to inquire and capacity for lifelong learning.
However, moving beyond motivation, the results illustrate that certain aspects of the
residential college environment itself, namely the degree to which the environment
promotes academic challenge and high expectations as well as faculty-student
interactions outside the classroom, have potential to deepen these outcomes in the
students’ that reside in them.

The current economic climate is making college decisions even harder for
students, while also highlighting the need for them to continue learning throughout their
lives as they prepare to contribute to the 21St century knowledge economy. It is also
forcing postsecondary institutions to take a hard look at the effectiveness of their
programs in light of declining budgets. With these pressures in mind, the results of the
current study indicate that postsecondary institutions can assist students in developing a
value of learning and capacity for lifelong learning by focusing on creating challenging
environments that are marked by deliberate opportunities for faculty and student

interactions outside of class.

148

APPENDIX A

SURVEY

149

Dear Participant:

This survey is part of a research study designed to frnd out about students’
participation in a residential college, their motivation, and their feelings about
leaming. Your participation will contribute to research on the residential college
experience and assist college and university practitioners in understanding how to
best support students.

Should you complete the survey, you will be entered into a drawing to receive a $100
gift card to amazon.com. There will be 5 gift cards given out. The information you
provide for the giftcard drawing will not be connected in any way to your survey
answers. The survey should take about 15 minutes to complete. Your privacy will be
protected to the maximum extent allowable by law.

Your participation is confidential, completely voluntary, and you may choose not to
participate. The IP address of your computer will not be linked to your survey. You may
refuse to answer any particular question. All responses will be numerically coded for
statistical analysis. Any identifiable information will be removed. You may discontinue
your participation at any time. There are no known risks associated with participation in
this study. You must be 18 years old to participate in this study.

If you have any questions about this study, please contact Dr. Kristen Renn, Associate
Professor in Educational Administration by phone: (517) 353-5979, email address:
renn@msu.edu, or regular mail: 428 Erickson Hall, East Lansing, MI 48824-1034. If you
have any questions or concerns about your role and rights as a research participant, or
would like to register a complaint about this research study, you may contact,
anonymously if you wish, Michigan State University Human Research Protection
Program at 517-355-2180, FAX 517-432-4503, or e-mail irb@msu.edu, or regular mail
at: 202 Olds Hall, MSU, East Lansing, MI 48824.

Thank you for participating!

150

1. Sex:

0 Male

0 Female

O Transgender

2. Citizenship:
0 US
0 Other: (Please list)

3. Race/Ethnicity (choose all that apply):

I] African-American E] Native American or Alaskan Native
El Asian American [I] Native Hawaiian or other Pacific Islander
13 Hispanic/Latino El White/Caucasian

C] International

4. What university do you attend?

5. Class year:

0 Freshman

O Sophomore

0 Junior

0 Senior

0 Graduate Student

6. Is this your first year at your current university
0 Yes

0 No

7. Did you transfer from another university, college, or community college?
0 Yes

0 No

8. What is your major? (If you haven’t decided, write undecided)

9. What residential college are you affiliated with?

10. Including this year, how long have you been affiliated with this residential college?

0 1 year 0 4 years
0 2 years 0 5 years
0 3 years O 6 or more years

151

ll. Including this year, how many ears have you lived in the residential college?

0 1 year 0 4 years
0 2 years 0 5 years
0 3 years 0 6 or more years

12. If your residential college did not exist, would you have attended the same university?
O Yes

0 No

If no, what college/university would you have attended?

13. What was your unweighted cumulative high school grade point average (gpa) on a 4.0 scale?

14. What was your SAT score? (write n/a if you did not take the SAT)
Verbal
Math

15. What was your ACT score? (Write n/a if you did not take the ACT)

16. What is the highest degree you plan to obtain in your lifetime?
0 Less than a bachelor’s degree

0 Bachelor’s degree (B.A., B.S., etc.)

0 Master’s degree (M.A., M.S., M.B.A., M.F.A., etc.)

0 Law Degree (J.D.)

0 Doctorate (Ph. D., Ed. D., M. D.)

17. What is the highest degree obtained by your mother/guardian?
0 Less than High School

0 High School Degree

0 Some college, no degree

0 College degree

0 Graduate or professional degree

18. What is the highest degree obtained by your father/guardian?
0 Less than High School

0 High School Degree

0 Some college, no degree

0 College degree

0 Graduate or professional degree

19. What is your parents’lguardians’ annual family income (approximately)?
0 Below $20,000

0 $20,001-$30,000

0 $30,001-$50,000

0 850,001-870,000

O' $70,001-$90,000

0 $90,001-$110,000

O $110,001-8130,000

O $130,001-8150,000

0 More than 150,000

152

20. Unique Identifier — (Will be used to identify your first set of answers, were you to fill out a
follow up survey. You are under no obligation to fill out a follow up survey, regardless of
whether or not you provide the information.

- First two letters of the city you were born in

- First two letters of your mother’s maiden name (if unknown, put UK)

- First two letters of the name of your elementary (or primary) school

- First two letters of your father’s first name (if unknown, put UK

 

21. Ifyou are willing to discuss your experiences in your residential college in more detail,
please supply your email address (note, you are under no obligation to supply your email
address:

22. The following questions ask about your preferences related to inquiry. There are no
right or wrong answers, just answer as genuinely as possible. Use the scale below to answer
the questions.

Extremely Uncharacter- Neither Character— Extremely
Uncharacter- istic of me Uncharacter- istic of me Characteristic
istic of Me istic nor - of Me

Characteristic
l. I would prefer complex to
simple problems. 0 O O O O
2. I like to have the
responsibility of handling a
situation that requires a lot

of thinking. 0 O O O O
3. Thinking is not my idea of
fun 0 O O _ O O

4. I would rather do

something that requires little

thought than something that

is sure to challenge my

thinking abilities. O O O O O
5. I try to anticipate and

avoid situations where there

is likely chance I will have

to think in depth about

something. 0 O O O O
6. I find satisfaction in

deliberating hard and for

long hours. 0 O O O O
7. I only think as hard as I

have to. O O O O O
8. I prefer to think about

small daily projects to long-

terrn ones. 0 O O O O
9. I like tasks that require

little thought once I’ve

learned them. 0 O O O O
10. The idea of relying on

thought to make my way to

the top appeals to me. O O O O O

153

Extremely Uncharacter- Neither Character- Extremely
Uncharacter- istic of me Uncharacter- istic of me Characteristic
istic of Me istic nor of Me
Characteristic

11. I really enjoy a task that

involves coming up with

new solutions to problems. 0 O O O O

12. Learning new ways to

think doesn’t excite me very

much. 0 O O O O

13. I prefer my life to be

filled with puzzles I must

solve. O O O O O
14. The notion of thinking
abstractly is appealing to me. O O O O O

15. I would prefer a task that

is intellectual, difficult, and

important to one that is

somewhat important by

doesn’t require much

thought. 0 O O O O
16. I feel relief rather than

satisfaction after completing

a task that required a lot of

mental effort. O O O O O
17. It’s enough for me that

something gets the job

done; I don’t care how or

why it works. 0 O O O O

18. I usually end up

deliberating about issues

even when they do not

affect me personally. O O O O O

23. In thinking over your experiences in college up to now, to what extent do you feel you have
gained or made progress in each of the following areas? Indicate your response by checking the
box that best describes your experience.*
Very Little Some Quite a Bit of Very Much
Progress Progress Progress Progress
1. Acquiring background and
specialization for further education
in some professional, scientific, or
scholarly field. ’ O O O O
2. Gaining a broad general
education about different fields of

knowledge. 0 O O O
3. Writing clearly and effectively. 0 O O O
4. Understanding other people and

the ability to get along with

different kinds of people. 0 O O O
5. Understanding new scientific

and technical developments. 0 O O O

154

Very Little Some Quite a Bit of Very Much
Progress Progress Progress Progress
6. Ability to think analytically or -
logically. O O O O
7. Ability to put ideas together, to
see relationships, similarities, and
differences between ideas. 0 O O O
8. Quantitative thinking —
understanding probabilities,
proportions, etc. 0 O O O
9. Ability to learn on your own, ‘
pursue ideas, and find information

you need. 0 O O O
10. Acquiring familiarity with the

use of computers. 0 O O O
11. Ability to function as a team

member. 0 O O 0

*All questions in this section were taken from the Capacity for Lifelong Learning index and used with
permission from the CSEQ Assessment Program, Indiana University, Copyright 1998, The Trustees of
Indiana University.

24. Please read each of the following items carefully, think about how it relates to your life, and
then indicate how true it is for you. Indicate your response by checking the box that best
describes your experiences.
Not Hardly Some Neither Some Often Very
at all ever true what true or what true true
true untrue untrue true
1. I feel like I am free to decide for
myself how to live my life.
2. I feel pressured by life.
3. I generally feel free to express my
ideas and opinions.
4. In my daily life, I frequently have
to do what I am told.
5. People I interact with on a daily
basis tend to take my feelings into
consideration.
6. I feel like I can pretty much be
myself in my daily situations. 0 O - O O O O O
7. There is not much opportunity for
me to decide for myself how to do
things in my daily life. 0 O O O O O O

O O

O O OO
O O OO
O O OO
O O OO
O O OO

O O
O O
O O

O
O
O
O
O
O
O

155

25. Please read each of the following items carefully, thinking about how it relates to your life,

and then indicate how true it is for you. Indicate your response by checking the box that best

describes your experience.
Not Hardly Some Neither Some Often Very
at all ever true what true or what true true
true untrue untrue true

1. Often, I do not feel very

competent. ‘ O O

2. People I know tell me I am good

at what I do. 0

3. I have been able to learn

interesting new skills recently. 0

4. Most days I feel a sense of

accomplishment from what I do. 0

5. In my life I do not get much of a

chance to show how capable I am. 0

6. I often do not feel very capable. O

OO O O O O
OO O O O O
OO O O O

OO O O O O
OO O O O O
OO O O O O

26. Please read each of the following items carefully, thinking about how it relates to your life,
and then indicate how true it is for you. Indicate your response by checking the box that best
describes your experience.
Not Hardly Some Neither Some Often Very
at all ever true . what true or what true true

true untrue untrue true
1. I really like the people I interact
with. O O O O O O O
2. I get along with people I come
into contact with. O O O O O O O
3. I pretty much keep to myself and
don’t have a lot of social contacts. 0 O O O O O O
4. I consider the people I regularly
interact with to be my friends. 0 O O O O O O
5. People in my life care about me. O O O O O O O
6. There are not many people that I
am close to. O O O O O O O
7. The people I interact with
regulme do not seem to much like
me. O O O O O O O
8. People are generally pretty
friendly toward me. O O O O O O O

156

27. The following pertain to your experiences in your residential college. Click on the response
that best describes your experience within your residential college.
Strongly Disagree Neither Agree Strongly
Disagree Agree or Agree
Disagree
1. Most faculty with whom I have
had contact are genuinely interested
in students. 0 O O O O
2. Most faculty with whom I have
had contact are interested in helping
students grow in more than just

academic areas. 0 O O O O
3. Most faculty with whom I have
had contact are outstanding teachers. 0 O O O O

4. Most faculty with whom I have

had contact are genuinely interested

in teaching. 0 O O O O
5. Most faculty with whom I have

had contact are willing to spend time

outside of class to discuss issues of

interest and importance to students. 0 O O O O
6. My non-classroom interactions

with faculty have had a positive

influence on your personal growth,

values, and attitudes. O O O O O
7. My non-classroom interactions

with faculty have had a positive

influence on your intellectual growth

and interest in ideas. 0 O O O O
8. My non-classroom interactions

with faculty have had a positive

influence on your career goals and

aspirations. O O O O O
. 9. Since entering the residential

college, I have developed a close,

personal relationship with at least

one faculty member. 0 O O O O
10. I am satisfied with the

opportunities to meet and interact

informally with faculty members. 0 O O O O

28. The following pertain to your experiences since entering your current college/university.
Click on the response that best describes your experiences, particularly in the current school
year.

Very
Never Rarely Sometimes Often Often
1. Frequency that faculty informed
you of your level of performance in
a timely manner. 0 O O O O

157

Very

Never Rarely Sometimes Often Often
2. Frequency that you received
prompt written or oral feedback from
faculty on your academic
performance.M O O O O O
3. Frequency that faculty checked to
see if you learned the material well

before going on to new material. . O O O O O
4. Frequency that faculty gave clear
explanations. O O O O O

5. Frequency that faculty made good

use of examples and illustrations to

explain difficult points. 0 O O O O
6. Frequency that faculty effectively

reviewed and summarized the

material. 0 O O O O
7. Frequency that faculty interpreted
abstract ideas and theories clearly.

8. Frequency that faculty gave
assignments that helped in learning
the course material.

9. Frequency that the presentation of
material was well organized.

10. Frequency that faculty were well
prepared for class.

1 1. Frequency that class time was
used effectively.

12. Frequency that course goals and
requirements were clearly explained.
13. Frequency that faculty had a
good command of what they were
teaching. 0 O O O O

O
O
O
O
0

00000
CO 000
O 0 000
OO O CO
CO 000

The following questions pertain to your experiences in your classes.

29. In your experience during the current school year, about how often have you worked harder
than you thought you could to meet an instructor’s standards or expectations?“

0 Never

0 Rarely

0 Sometimes

O Often

0 Very Often

 

** Items used with permission from Indiana University

158

30. Over the course of the current school year, how many assigned textbooks, books, or book-
length packs of course readings have you/or will you read?“

0 None

0 Between 1 and 4

0 Between 5 and 10

0 Between 11 and 20

0 More than 20

31. Over the course of the current school year, how many papers or reports between 5 and 19
pages did you write?**

0 None

0 Between 1 and 4

0 Between 5 and 10

0 Between 11 and 20

0 More than 20

32. In a typical week, how many problem sets of homework take you more than an hour to
complete?“

0 None

0 1-2

0 3-4

0 5-6

0 More than 6

33. Please click on the button that best describes your experience in your current
college/university, particularly in your residential college courses.

. Very
Never - Rarely Sometimes Often Often
I. How often have your

examinations challenged you to do

your best work?" 0 O O O O
2. How often does your residential

college emphasize spending

significant amounts of time studying

and on academic work?“ 0 O O O O
3. During the current school year,

about how often have you asked

questions in class or contributed to

class discussion?M O O O O O
4. In a typical academic year, about

how often will you make a class

presentation?" 0 O O O O
5. In a typical academic year, about

how often will you prepare two or

more drafts of a paper of assignment

before turning it in?" O O O O O
6. During the current school year,

about how often have you come to

class without completing readings or

assignments?" O O O O O

 

** Items used with permission from Indiana University

159

7. In a typical academic year, how
often do exams or assignments
require you to write essays?

8. Since entering college, how often
have exams of assignments required
you to use course content not
presented in the course to address a '
problem?

9. In your experience since entering
college, how often have exams or
assignments required you to compare
or contrast topics or ideas from a
course?

10. Since entering college, how often
have exams or assignments required
you to point out the strengths and
weaknesses of a particular argument
or point of view?

11. Now often have exams or
assigmnents required you to argue
for or against a particular point of
view and defend your argument?

12. How often have faculty asked
challenging questions in class?

13. How often have faculty asked
you to show how a particular course
concept could be applied to an actual
problem or situation?

14. How often have faculty asked
you to point out any fallacies
(erroneous beliefs) in basic ideas,
principles, or points of view
presented in the course?

15. How often have faculty asked
you to argue for or against a
different point of view?

16. How often have faculty
challenged your ideas in class?

17. How often have students
challenged each other’s ideas in
class?

Never

160

Rarely

Sometimes

Often

Very
Often

34. The following questions correspond to your experience in your current college/university,
and particularly in your courses affiliated with your residential college. Please click on the

button that best describes you revel of agreement with each statement.
Strongly Disagree Neither Agree
Disagree Agree or
Disagree
1. The courses have helped me
understand the historical, political
and social connections of events. 0 O O O
2. The courses have helped me see
the connections between my
intended career and how it affects
society. 0 O O O
3. My out-of-class experiences have
helped me connect what I have
learned in the classroom with life
events. 0 O O O
4. My out-of-class experiences have
helped me translate knowledge and
understanding from the classroom
into action. 0 O O O

35. Please click on the button that best describes your experience in your current

college/university, particularly in your courses affiliated with your residential college.

Never Rarely Sometimes Often

1. During the current school year,

about how often have you worked on

a paper or project that required

integrating ideas or information from

various courses?“ 0 O O O
2. During the current school year,

about how often have you put

together ideas of concepts from

different courses when completing

assignments or during class?“ 0 O O O
3. During the current school year,

about how often have you discussed

ideas from your readings or classes

with others outside of class

(students, family, co-workers,

etc)?" 0 O O O
4. During the current school year,

how much has your coursework

emphasized synthesizing and

organizing new ideas, information,

or experiences into new, more

complex interpretations and

relationships?“ O O O O

Strongly
Agree

Very
Often

 

** Items used with permission from Indiana University

161

. Very
Never Rarely Sometimes Often Often
5. During the current school year,

how much has your coursework

emphasized making judgments about

the value of information, arguments,

or methods, such as examining how

others gathered and interpreted data

and assessing the soundness of their

conclusions?" 0 O O O O

36. Please click on the button that best describes your experience since arriving in your

residential college.

Never Rarely Sometimes Often Very
Often

I. How often have you attended a

debate or lecture on a current

political/social issue? 0 O O O O

2. How frequently have you had

serious discussions with student

affairs staff (e.g., residence hall staff,

career counselor, student union or

campus activities staff, etc.) whose

political, social, or religious views

were different from your own? 0 O O O O

3. To what extent does you

residential college emphasize

encouraging contact among students

from different economic, social, and

racial or ethnic backgrounds?" O O O O O

4. In your experience in your

residential college during the current

school year, about how often have

you had serious conversations with

students of a different race or

ethnicity than your own?“ 0 O O O O

5. In your experience in your

residential college during the current

school year, about how often have

you had conversations with students

who are very different from you in

terms of their religious beliefs,

political opinions, or personal '

values?" 0 O O O O

6. How often have you participated

in a racial or cultural awareness

workshop during this academic year? 0 O O O O

 

** Items used with permission from Indiana University

162

. Very
Never Rarely Sometimes Often Often
7. How often have you had

discussions regarding intergroup

relations with diverse students (e. g.,

students differing from you in race,

national origin, values, religion,

political views) since you became a

part of your residential college? 0 O O O O
8. How often have you had

meaningful and honest discussions

about issues related to social justice

with diverse students (e.g., students

differing from you in race, national

origin, values, religion, political

views) since you became a part of

your residential college? 0 O O O O
9. How often have you shared

personal feelings and problems with

diverse students (e. g., students

differing from you in race, national

origin, values, religion, political

views) which you became a part of .

your residential college? 0 O O O O

37. About how many hours in a typical week do you spend participating in co-curricular
activities (organizations, campus publications, student government, fraternity or sorority,
intercollegiate or intramural sports, etc.)?**

0 0 hours

0 1-5 hours

0 6-10 hours

O 11-15 hours

0 16-20 hours

O 21-25 hours

0 26-30 hours

0 More than 30 hours

38. The following questions pertain to your experiences with peers in your residential college.
Please click on the button that best describes the extent to which you agree with the following
statements.
Strongly Disagree Neither Agree Strongly
Disagree Agree or Agree
Disagree
1. I have developed close personal
relationships with other students
since coming to this residential
college. 0 O O O O
2. The student friendships I
developed in my residential college
have been personally satisfying. 0 O O O O

 

** Items used with permission from Indiana University

163

3. My interpersonal relationships

with other students have had a

positive influence on my personal

growth. 0
4. My interpersonal relationships

with other students have had a

positive influence on my intellectual

growth and interest in ideas. 0
5. The quality of my relationships

with other students in my residential

college have been friendly,

supportive, and given me a sense of

belonging." O
6. It has been difficult for me to meet

and make friends with other

students. 0

7. Few of the other students I know

would be willing to listen to me and

help me if I had a personal problem. 0
8. Most students in my residential

college have values and attitudes

different from my own. 0

 

** Items used with permission from Indiana University

164

APPENDIX B»

VARIABLE INFORMATION -

165

VARIABLES INFORMATION
Dependent Variables

Inclination to Inquire (InctoInq) — a scaled measure with a mean of 54.45 and a standard
deviation of 8.27. The scale was comprised of 15 items detailing students’ self-reports of
their enjoyment of thinking (see appendix G for all 15 scale items). The reliability of the
scale was .873

Capacity for Lifelong Learning (CapLifeLrn) — a scaled measure with a mean of 29.71
and a standard deviation of 6.58. The scale was comprised of 11 items from the Estimate
of Gains items on the College Student Experiences Questionnaire, which measured I
students’ self-reported “ability to “learn to learn” and interact effectively with others in a
complex, information-based society, indicating the extent to which students have
acquired continuous learning skills” (Hayek & Kuh, 1999, p. 4). The scale was comprised
of students’ self-reports of the impact of their collegiate experience on the following
items: Ability to synthesize ideas; think analytically; learn on one’s own; function as a
team member; understand others; write clearly; gain a broad education; become familiar
with computers; acquire background information for further education; think
quantitatively; understand new scientific developments. The reliability of the scale was
.880.

Independent Variables —Level One
Socio Demographic Characteristics

Sex (Male) was a dichotomous variable, coded 0 for women; 1 for men. Men comprised

32.6% of the analytic sample (n=591).

Race/Ethnicity comprised a series of dichotomous variables with regard to the

 

race/ethnicity to which students identified at the time they completed the survey. See

below for the dummy variables entered into the analytic models.

166

Afiican-American (AfAm) was a dichotomous variable, coded 0 for non-Afiican
American; 1 for African-American. 1.8% or 33 students in the sample identified

as Afiican-American.

Asian-American (AsAm) was a dichotomous variable, coded 0 for non-Asian
American; 1 for Asian-American. 21.3% or 386 students in the sample identified

as Asian-American.

Hispanic/Latino (His/Lat) was a dichotomous variable, coded 0 for non-
Hispanic/Latino; 1 for Hispanic/Latino. 7.1% or 129 students in the sample

identified as Hispanic/Latino.

White (white) was a dichotomous variable, coded 0 for non-White; 1 for White.
56.2% or 1017 students in the sample identified as White and this served as the

reference group for all of the analyses.

International (Intl) was a dichotomous variable, coded 0 for non-Intemational; 1
for International. 2% or 36 students in the analytic sample identified as

International.

Race-Other (RaceOth) was a dichotomous variable, coded 0 for non-Race-Other;
1 for Race-Other, which was a combination of Native American/Hawaiian and
Hawaiian or Pacific Islanders. .8% or 15 students in the analytic sample identified

as Race-Other.

Multiracial (MultiRac) was a dichotomous variable, coded 0 for non-Multiracial;
l for Multiracial. 9.8% or 177 students in the analytic sample identified as more
than one race/ethnicity.

167

No Response for Race (NoRespRace) was a dichotomous variable, coded 0 for
provided a response for race; 1 for no response for race. 1% or 18 students did not
identify a race.

Pre-College Characteristics

Family Income was originally a nine-point ordinal variable ranging from 1 (=below
$20,000) to 9 (above $150,000). For analytic purposes, I calculated the percentiles of the
original scale and used the lower 25%, middle 50% and upper 25%.

Famin Income below $50,000 (IncLow) was a dichotomous variable, coded 0 for
family income above $50,000; 1 for family income below $50,000. 25.4% or 460
students in the analytic sample described their family income as below $50,000.

Family Income between $50,000 and $110,000 (IncMed) was a dichotomous
variable, coded 0 for family income below $50,000 or above $110,000. 41.1 % or
745 students in the analytic sample described their family income as between
$50,000 and $110,000, and this served as the reference group for all of the

analyses.

Family Income above $110,000 (IncHi) was a dichotomous variable, coded 0 for
family income below $110,000; 1 for family income above $110,000. 29.5% or
535 students described their family income as above $110,000.

High School GPA (HSGPA) was a scale variable with a mean of 3.63 and a standard

deviation of .321

F irst-Generation (F irstGen) was a dichotomous variable, coded 0 for non-First generation
and l for first-generation. For the purposes of the current study, first generation was
defined ‘neither of a students’ parents obtained a bachelor’s degree. 22.3% or 404
students disclosed that neither of their parents/guardians had a college degree.

168

College Experience and Motivation

Years Lived in the Residential College (YrsLived) was a scaled variable with a mean of

1.45 and a standard deviation of .730.

Degree Aspirations was originally a five-point ordinal variable, coded 1 for less than
bachelor’s; 2 for bachelor’s; 3 for master’s; 4 for law degree; and 5 for Ph. D., Ed. D., or
M. D. For analytic purposes, I recoded it to less than bachelor’s, bachelor’s and more

than bachelor’s.

Degree Aspirations — Less than Bachelor’s Degree (DgAspNoBach) was a
dichotomous variable, coded 0 for aspiring to a bachelor’s degree or more; 1 for
aspiring to less than bachelor’s. 1.6% or 26 students in the analytic sample aspired

to less than a bachelor’s degree.

mgree Aspirations — Bachelor’s Degree (DgAspBach) was a dichotomous
variable, coded O for aspiring to obtain less or more than a bachelor’s degree; 1
for aspiring to obtain a bachelor’s degree. 15.3% or 277 students in the analytic
sample aspired to obtain a bachelor’s degree, and this served as the reference

group for all of the analyses.

Degree Aspirations — More t_h,an Bachelor’s Degree (DgAspMoBach) was a
dichotomous variable, coded 0 for aspiring to obtain a bachelor’s degree or less; 1
for aspiring to obtain more than a bachelor’s degree. 82.9% or 1499 students in

the analytic sample aspired to obtain more than a bachelor’s degree.

Motivation (Motivation) was a scaled measure with a mean of 89.87 and a standard
deviation of 13.45. The scale was comprised of 17 items which detailed students’ beliefs
about their level of autonomy, competence and relatedness, which serve as a foundation
for motivation (see appendix C for all 17 scale items). The reliability of the scale was

.884.

169

Individual Liberal Arts Experience Variables

Good Teaching and High Quality Interactions with F acultv
Classroom Practices (T eachClass) was a scaled measure with a mean of 42.42
and a standard deviation of 6.42. The scale was comprised of 11 items which
detailed students’ beliefs about the preparedness of their instructors. The
reliability of the scale was .911.

Out-Of-Class Interactions with Faculty (TeachOut) was a scaled measure with a
mean of 17.03 and a standard deviation of 4.03. The scale was comprised of 5
items which detailed students beliefs about their out of class interactions with

their instructors. The reliability of the scale was .844.

Academic Challenge and High Expectations (AcChallenge) was a scaled measure with a
mean of 72.29 and a standard deviation of 12.57. The scale was comprised of 21 items
which detailed students’ beliefs about the level of challenge in their collegiate

environment (see appendix D for all 21 scale items). The reliability of the scale was .912.

Diversity Experiences (Diversity) was a scaled measure with a mean of 25.24 and a
standard deviation of 6.16. The scale was comprised of 8 items detailing students’ beliefs
about their exposure to diversity in their collegiate environment (see appendix E for all 8

scale items). The reliability of the scale was .83 8.

High Quality Interactions with Peers (Peers) was a scaled measure with a mean of 22.78
and a standard deviation of 5.38. The scale was comprised of 6 items detailing students’
beliefs about their interactions with their peers (see appendix F for all 6 scale items. The

reliability of the scale was .910.

Independent Variables —Level Two

Environmental Liberal Arts Experience Variables

170

Good Teaching and High Quality Interactions with Faculty

Clasgroom Practices (GrpMnExTClass) was comprised of the mean of the level-
one ‘classroom practices’ measure for all students in a given residential college.

The mean was 42.45 and standard deviation was 1.70.

Out-Of-Class Interactions with Faculfl (GrpMnExT Out) was comprised of the
mean of the level-one ‘out-of-class interactions’ measure for all students in a

given residential college. The mean was 17.25 and standard deviation was 1.48.

Academic Challegm and High Expectations (GrpMnExCh) was comprised of the mean
of the level-one ‘academic challenges’ measure for all students in a given residential

college. The mean was 72.74 and standard deviation was 6.63.

Diversity Experiences (GrpMnExD) was comprised of the mean of the level-one
‘diversity experiences’ measure for all students in a given residential college. The mean

was 25.29 and standard deviation was 2.00
High Quality Interactions with Peers (GrpMnExP) was comprised of the mean of the

level-one ‘interactions with peers’ measure for all students in a given residential college.

The mean was 22.90 and standard deviation was 1.53.

171

APPENDIX C
FACTOR ANALYSIS OF MOTIVATION (GENERAL NEEDS SATISFACTION)

SCALE

172

Scales for ‘Motivation’

 

 

Residential college students ’ self-reports of their Reliability Factor
‘General Needs Satisfaction ’ Loading
Autonomy a = .746
1. I feel like I can pretty much be myself in daily
situations. .81 8
2. I generally feel free to express my opinions and
ideas. .775
3. I feel like I am free to decide for myself how to
life my life. .730
4. People I interact with on a daily basis take my
feelings into consideration. .664
5. There is not much opportunity for me to decide
for myself how to do things in my daily life
(recoded). .541
Competence a = .745
1. I often do not feel very capable. .832
2 Often I do not feel very confident. .792
3. Most days I feel a sense of accomplishment from
what I do. .692
4. In my life I do not get much of a chance to show .631
how capable I am.
5. People I know tell me I am good at what I do. .536
Relatedness o. = .828
1. I really like the people I interact with. .800
2. .I get along with people I come into contact with. .778
3. People in my life care about me. .753
4. I consider the people I regularly interact with to .753
be my friends. .
5. People are generally pretty friendly toward me. .746
6. The people I interact with regularly do not seem .669
to much like me (recoded).
7. There are not many people that I am close to. .569

Overall scale for Motivation (1 =.884

173

APPENDIX D

FACTOR ANALYSIS OF LIBERAL ARTS EXPERIENCE SCALE-ACADEMIC

CHALLENGE AND HIGH EXPECTATIONS

174

Scale for ‘Academic Challenge and High Expectations’

 

Residential college students ’ self-reports of ‘Academic
Challenge and High Enpectations ’

Reliability

Factor
Loading

 

1. How often have faculty asked you to argue for or
against a particular point of view?

2. How often have exams or assignments required
you to argue for or against a particular point of
view and defend your argument? ’

3. Since entering college, how often have exams or
assignments required you to point out the
strengths and weaknesses of a particular
argument or point of view?

4. How often have faculty asked you to point out
any fallacies (erroneous beliefs) in basic ideas,
principles, or points of view presented in the
course?

5. During the current school year, how much has
your coursework emphasized making judgments
about the value of information, arguments, or
methods such as examining how others gathered
and interpreted data and assessing the soundness
or their conclusions?* *

6. How often have faculty challenged your ideas in
class?

7. How often have students challenged each other’s
ideas in class?

8. During the current school year, how much has
your coursework emphasized synthesizing and
organizing new ideas, information or
experiences into new, more complex
interpretations and relationships?”

9. During the current school year, about how often
have you worked on a paper or project that
required integrating ideas or information from
various sources?* *

10. In your experience in your residential college,
how often have exams or assignments required
you to compare or contrast topics or ideas from
a course?

11. How often have faculty asked challenging
questions in class?

** Used with permission from Indiana University

175

a=.9ll

.718

.713

.708

.696

.675

.666

.662

.661

.660

.632

.631

 

12. In a typical academic year, how often do exams
or assignments require you to write essays?

13. During the current school year, about how often
have you put together ideas or concepts from
different courses when completing assignments
or during class?”

14. The courses have helped me to understand the
historical, political and social connections of
past events?

15. How often have faculty asked you to show how
a particular course concept could be applied to
an actual problem or situation?

16. During the current school year, about how often
have you discussed ideas from your readings or
classes with others outside of class (students,
family, co-workers, etc.)?**

17.During the current school year, about how often
have you asked questions in class or contributed
to class discussion?"

18. My out-of-class experiences have helped me to
translate knowledge and understanding from
classroom to action.

19. My out-of-class experiences have helped me
connect what I have learned in the classroom
with life events?”

20. Over the course of the current school year, how
many papers or reports between 5 and 19 pages
will you write?* *

21. In a typical academic year, about how often
will you prepare two or more drafts of a paper
or assignment before turning it in?M

** Used with permission from Indiana University

176

.610

.595

.591

.577

.573

.472

.470

.470

.425

.387

APPENDIX E

FACTOR ANALYSIS OF LIBERAL ARTS EXPERIENCE SCALE- DIVERSITY

EXPERIENCES

177

Scale for ‘Diversity Experiences’

 

Residential college students ’ self-reports of ‘Diversity Reliability Factor
Experiences’ Loading

 

a = .838
1. During the current school year, about how often
have you had serious conversations with students
of a different race or ethnicity than your own?” .824

2. During the current school year, about how often

have you had conversations with students who

are very different from you in terms of their

religious beliefs, political opinions, or personal .782

values?* "'
3. How often have you had discussion about

cultural differences with diverse students (e. g.,

students differing from you in race, national

origin, values, religion, political views) since

you became part of your residential college? .765
4. How often have you shared personal feelings and

problems with diverse students (e. g., students

different from you in race, national origin,

values, religion, political views) since you

became a part of your residential college? .760
5. How often have you shared personal feelings and

problems with diverse students (e.g., students

differing from you in race, national origin,

values, religion, political views) since you

became part of your residential college? .722
6. To what extent does your residential college

encourage contact among students from

different economic, social and racial/ethnic

backgrounds?" .527
7. How frequently have you had serious discussion

with student affairs staff (e. g., residence hall

staff, career counselor, student union or campus

activities staff, etc.) whose political, social, or

religious views were different from your own? .331
8. How often have you/will you participate in a

racial or cultural awareness workshop during

this academic year? .302

** Used with permission from Indiana University

178

APPENDIX F

FACTOR ANALYSIS OF LIBERAL ARTS EXPERIENCE SCALE —

INTERACTIONS WITH PEERS

179

Scale for ‘Interactions with Peers’

 

Residential college students ’ self-reports of ‘Interactions Reliability
with Peers ’
a = .910
1. The student friendships I have developed in my
residential college have been personally
satisfying.

2. My interpersonal relationships with other
students have had a positive influence on my
personal growth, attitudes, and values.

3. I have developed close personal relationships
with other students since coming to this
residential college.

4. The quality of my relationships with other
students in my residential college have been
friendly, supportive, and have given me a sense
of belonging.* *

5. My interpersonal relationships with other
students have had a positive influence on my
intellectual growth and interest in ideas.

6. It has been difficult for me to meet and make
friends with other students.

** Used with permission from Indiana University

180

Factor
Loading

.911

.893

.886

.865

.853

.623

APPENDIX G

FACTOR ANALYSIS OF INCLINATION TO INQUIRE SCALE

181

Scale for ‘Inclination tolnquire’

 

 

Residential college students ’ self-reports of their Reliability Factor
‘Inclination to Inquire ’ Loading
a = .873
1. I would prefer complex to simple problems .688

2. I would prefer a task that is intellectual, difficult,
and important compared to one that is somewhat

important but doesn’t require much thought. .68]
3. I like to have the responsibility of handing a

situation that requires a lot of thinking. .677
4. I prefer my life to be filled with puzzles I must

solve. .65 1
5. I find satisfaction in deliberating hard and for

long hours. .648
6. The idea of relying on thought to make my way

to the top appeals to me. .631
7. I really enjoy a task that involves coming up

with new solutions to problems. .623
8. The notion of thinking abstractly is appealing to

me. .553
9. I usually end up deliberating about issues even

when they do not affect me personally. .475
10. Learning new ways of thinking doesn’t excite

me much (recoded). .444
1 1. Thinking is not my idea of fun. (recoded) .405

12. I try to anticipate and avoid situations where

there is likely a chance I will have to think in

depth about something (recoded) .360
13. I would rather do something that requires little

thought than do something that is sure to

challenge my thinking (recoded). .350
14. I like tasks that require little thought once I’ve
learned them (recoded) .300

15. It’s enough for me that something gets the job
done; I don’t care how or why it works
(recoded) .277

182

APPENDIX H

FACTOR ANALYSIS OF CAPACITY FOR LIFELONG LEARNING SCALE

183

Scale for ‘Capacity for Lifelong Learning’

 

 

Residential college students ’ self-reports of their Reliability Factor
’Capacityfor Lifelong Learning’ ** Loading
or = .880
1. Ability to put ideas together, to see relationships,
similarities, and differences between ideas. .800
2. Ability to think analytically or logically. .762
3. Ability to learn on your own, pursue ideas, and
find information you need. .738
4. Ability to function as a team member. .718
5. Understanding other people and the ability to get
along with different kinds of people. .718
6. Write clearly and effectively. .711
7. Gaining a broad general education about
different fields of knowledge. .642
8. Acquiring familiarity with the use of computers. .609
9. Acquiring background and specialization for
further education in some professional,
scientific, or scholarly field. .553
10. Quantitative thinking — understanding
probabilities, proportions, etc. ‘ .475
11.Understanding new scientific and technical
developments.
.428

** All items used with permission from the CSEQ Assessment Program, Indiana

University, Copyright 1998, The Trustees of Indiana University.

184

APPENDIX I

BIVARIATE CORRELATIONS OF LEVEL-l VARIABLES

185

 

 

 

8o. 8...
5.8. .238.
Sumo. £6.

.8. EEO
.8. 90.8.
E... :80.
£8... $3...
8.8. Ca...
58.. $8..
58.. E8...
Arr!- .88....
68...- 38..
......t. .88.
8. to.
ERN- :8.
Can. .338-
.:.48. ES.
. $.23
.25.». .
From. 58..
8o.- 3...-
8...- 8o.-
.8- CE..-
”.8 8o.-
?.3~. r38.-
38.- .28:-
Ar..:- 8....
..8. £8.-
5< (new...

 

 

 

8....
8o.
3...

«8..

«N...-
33.

8....

E..-

AstSf
8....
80..

F3:-
.

moo.-
m8.-
:0...
0N0.-
Cup-yo.-
v3.-
30..
Swansoz

 

 

awo-
3°.
wvor
m 90. u
mmo. .
GOO.-
V3.1
#3..
08..
08.
NNor
og-
moo.
onor
ONO.
N8.
30..
NS.-
0N0.-
AlevMOFf
Goo.-

 

8......o

 

v8.-
v8.1
Aevfigf
Asivuhmr
one.-
Ono.-

‘—

Anvkgf
Aiiv POOP
Ply wk 9 .1

m3...
mNo .-
0013—32

 

 

92.8..

08.: NewMWIO.
ALVA-yo. alvfiaf
.8- 2...
2o ALE-o.
2o. .3...
9... 8o. -
2o. 3...
3.... 8o.-
..8. N8.-
Nuo. 38.
8o.- €5.-
35. .8.
2.... Fa:-
ouo. 8...-
8o.- Ema.
so. ESQ
8.... F33.
38. Fia-
ouou r38.-
io. - 8...-
2o.- 8....
..r3 - CE?
8...-

8o.- .
FE... ES...
2? 8o.-
2... 8.
.3. .33.:

 

 

Atvono.

omo. .
Automo-
Aivhm r-

moo.
Emma. -
Aavomo.. r

30.. -

 

.33. 3.3 lore... 32.8.8.8... 8. 3 9m.
Enos-.830

233:.
flood
3.92,...
.35 .2.
39.8.:
83.0283.-
5.32.0.2
853
3822

 

Cum... 8J1
F38. 2o. -
F3:- 2...
CE...- 2...
Cs..- 8....
onw 2....
.ri..- 8o.-
r..8~- 2...
c8.- 8...
8...- 8...
8.... v8.-
53... to.-
.r:8. $8....
.8- so.
:5. ...o8.
8.8. 9....
8o.- 8....
8.. - fr:-
....2.. cm...-
...~8.- 3...-
Cox. - N3. -
EE- 8... -
CE..- 3 8. -
CS..- 80. -
. Exo-
.:.E... F
«5. m8.-
.1|l
E<w< flags.

 

 

 

 

 

186

.8..a.«. 3.3. F33 .902. ...8..a.«8«. 3.3. 8. 3 .85 .

 

 

 

 

 

 

 

 

 

 

 

 

 

F38. F.«F«. .88. «8.- ...F8- F8. «8.- «8- «,3.- «8. «.8...
.288. F.«:. F88. F88. «8.. «3.- «8.- «8. 8...- F38. 3.2on
F.«F«. ««o. F: F F. F.8- 8? 8...- «8. «8.- «8.- ....-8- .35 .2
F38. F.«FF. FE... F3:. «8. «8. «8.- «8.. «8.- «8- 3068»
F38. .8. F88. F.8- «8- «8. F8. F3- ...«8- ...«8. «8.038»
F F38. F.«8. .88.. F88. F8. 8o «3.- ««8.- «8. «88.3.2
F88. F ««o. 8...- «8- «8. «8. «8.- 8...- F.«FF. . 85p;
F38. ««o. F F38. F38. «8. F8. - 8... . ««8.- F38. .88:
.38. - «8.- F.«««- F 88.... «8.- «8. FF..- «8. £88.- .38
F33- «8.- F38. 88...- F 8.. F8- «8. F38. «8.- «8.32
8... «8. «8. «8.- 8.. F F58- F.8«- F.8«.- F.F«F. =3...
8... «8. 8.... «8. F8.- F.8- F F88- ...F8- «8. 3%...
««o. - «8.- 8... FF... - «8. F38. F88- F F38. F.«o«. 33...
«3.- «8.- ««8.- «8. F.«8. F.8«- CF8- F.8«. F .338. 80.2....
«8. F.«:. F38. .88.- «8.- F.F«F. «8. F.«o«. F.8«.- F Em
C8. F.8F. .388. CE. .88.- F.«..F. «8. F.8«- F.«««.- F.«8. F0...
F.«oF. F38. F38. 88...- 8o.- €«8. to. F...8. F.«8.- F.8F. (new...
F.8F. 8o. «8. - F88. F.8F- F.««F. «F... F.«o«- F.oF«.- .282. 2...;
8...- «8.- So. «8. «8.- «8.- ««o. «8.- «8.- ...«8.- 898232
««8.- «8.- 8... «8.- ..«o. 8... «8.- «8. 8°. «8.- 8850
««o. «8.- «F... «8.. 8...- ...«3- «8. «8. ««o. 8.. 88...:
«8.- «8.- «8. «8.- «F... «8.- «8. «8.- 8o. «8.- .3.
«8 - «8.. «8. 88.... 8. CS F- «8.- F.««F. F.o««. F33. .33.:
.288. «8.- 8...- «8.- F.«FF. F.8- «8- F...«F. £88. «8.- :22
«F... «8. «8. «8.. E..- 838- 88. £88. «3. «8.. .52
.38.- «8. «8.- «Fe. 8... «F... 8°. «8.. §.. .288. «3:
«888.2 82.3; :88: .38 .3832 .3... 3%... 3.3:. 80.2w. Em

 

 

187

.888. .95. 8... .882. ..8..s.«. 3.6. 8. 3 .88 -

 

 

 

. «8. F38. F38. F38. 4.38. 838. Ema-:80

«8. F F.8F. F38. F38. F38. F38. 838...
F38. F32. F F38. F38. F38. F38. 28..
F38. F38. F38. F F38. .288. .282. 33.2....
F38. F38. F38. F38. . F38. F38. .35 .2
F38. F38. F38. F38. F38. F .28F«. 32.88
F38. F38. F38. F22. F38. F38. F 8208.8
F38. F38. F38. F32. F38. .288. F...««. 88885.
F32. ..38. F38. F3... .3... F3... FF... 8...»...
F32. .282. ..38. F38. F23. F32. F38. .3822
F32.- F38. ..8.- .288.- F3:- F3F..- F38.- .38
38.- .38.- .88.- 8...- 38.- 2... 88.- .3802
«8.- ...«8. 8... ..8.- 8...- 38. ..8. =3...
8... 8... «8.- 2...- ..8. «8.- 8... 38...
2... F32- 8...- ..8. «8.- 8...- F8.- .83... .
«8. F32- 8...- 8... ..8.- 2...- .88.- 803....
«8.- F38. 8... F38.- .88.- ...8.- ..38. 58
8... F28. .-.«8. 8... «8. «8. ..38. F9.
8... F38. F22. F38. F28. F38. F.8.. <88:
«8. F38. F32. 88. F.2.. F38. F32. 2...;
«8.. F8. ..8.- 8o.- .8. 8... 2... 8288...:
..8. «8. .8. 2...- 38- 8...- ««o.- 8850
8...- 8... «8. 88.- 2...- 88.- 8... 88......
8.... ..38. 8.- ..8. 2... 88. ..F... .3.
.88. .88.- 2... F38. «8. 8...- «8. .33....
F38.- F38. F3F F- 88...- F.««.- F.8F.- F.8F- .53..
..8. ..8.- 2... 2... 38.- 2...- 38.- .52
«.3. F388. ..8.- F38.- .--.8F. 88. .88.- 2......

Ems-.88 8.39.. 28.. 3.228 .35 .2 50.3.3 38.0.3.3

 

 

 

 

 

 

 

 

 

188

APPENDIX J

BIVARIATE CORRELATIONS OF LEVEL-2 VARIABLES

189

Gags-a .96. San gm... Megs-u .92 8. 5 gm .

 

 

 

 

 

 

 

 

 

 

 

 

F 88».- Covo- C89- 53. fat. €39 Haunt ENR. com-dams
I. 0
£03.- €21- F E 5o.- Cmmo. 52b. m8. :8... 03.. cw“:
I. o
Emmo- ano- fra- P from. €80. EEN- Econ. 58,. .8
52.320
ENE. 58”.. Camo. Atxou. . EB: fawn. Fzfi. imam. 28a om
Cot. CNS. Etc. p.33.- r-Et r ESQ Fit 5%. £2on om
$8». $38.- muo. ALE-N.- Cumo. 9.28. . €80. £3». =20 2 oz
Cunt CR".- 58? Cam. £02. 5.8. :80. P 5.2.. so ..8- om
ENE. £53. «3.. r-KE. Can». {an 53m. CE. F can. no» om
sec. :25... am £55 :20 so 8i
cow-.820 -920 ..go $2.83 22:. um oz 2 om :8. oz .3 om

 

 

190

APPENDIX K

FULL MODEL OF INCLINATION TO IN QUIRE

191

 

.m a 2.3. a. .= 8;. .3. at... <6 mm,
.m a 88 .m a 3:. .m a $85. x28 2,3,.- 8.85 beam
.m a «3.? .m a five. -. a 83. van» 320m 2:85 £55
Rae» 822%.: «233 use-é 335
.m a 23- .m a 82.. a. a Se...- aoua... 09:8 Sufism-Em
2.. as.— .. Eh.— .. we; .3 Nam; 032
.m a 8a.... .m a 23 .m a 2:.” .m a 35." 38%.. oz
.3. «Sn .3. :2 .3 83. .3 and. sac-08m
.m a 55- .m a $3- .... a $3. < 52. 30538
. «~3- .. m2... . 83 .m a a: 35355
a. a 83. . «Sa- < «3...- : $2- egg“:
2... own-m- 2:. an-m- 2:. giv- Ii mam-Q. gtosqgma
.m a 83. .m a :2. .m a 2c."- a. a mad. Bog-=35
«gnu 35.8%.: 833
83...: 2...—£25.35
£8 55 288.35 £35 .3:
825:0qu £95
saga E: v5 93030 0388(-
aofipfis 30%-.8 .
3038.5 8859—0 ..
.938". a?
8°33:— baao .3: 2. 35.5 86
33.5
3.83.3. 3 83.8.3.0 .3525
:9 «3.3 :5 coo-Nu o: mmofin o: hue-mm “moo-35
gm €< .3 ”am is: a. ma muo=8 ..mw 5m a. .32
,3. + an. :8 .E 5803. ".082
n :82 + N 382 + .082 s .082
a is: H... .25: a E52 .33

.332 as. 8.5.: 8 3.3-=8.

192

So. Va .3... 3c. vat. "we. va 9 £.VQ<

 

 

So.“ 83. 83. o8.” m:
.2. £59: 2: 82a: .2. 323 3%: 333m 82.5 so
83a . 8:2 8a.: 8a.: 83 585%..
Nng- 828... “8:8. :38”. SSEQ £28 8.3::
83 89c 33 28 «as .8283
82.353.
mm; a; «8... E3 83 Bias 33:8 3825.553
233“ «8.2 :58 «2.8 818 20:8 3825.595,
82 :3 82. «as :8... 3.5 3:335 88:8 Sufism-5033
a: 823 :. $2 5 89m 2: an.» t. 83. €8§£ 89:8 augfix-asavm
8.532.950 con-t!»
: 2:.- 28m 55 Sega £30 :3:
z 5.9 8085.5 €55
< :3 823.03%". .3: Ba 09.236 £883
< 2 _ .c 888.25 30%-.8 -
.2. 3:5 8885 sagas .
33.5 a?
88335 £35 :9: 2a Essa 88
58:35 .5. 1.2.... .525...
.2. $2 :- 68.” ”Egg Se 932 . ”8% 8&8
.m a me .c .m a 3: £2938 8.: 83 . songs? 8&8
«mo—«2&3» {Suficm - 382%“.- 3&3
. 33 < 83 um a 35 285° 25:2
.5. ”2.0 o: n26 859502
. 33...: sataua .950
.fi 32 ..E mm is: 4 3w Lamas % 5m a. 3.2
as + 9m. .=8 .95 .53 ".082
n 332 + N 382 i .032 u. .332 .39—
a. 3% a .082 a .352 .23: as... 2.3.: 3 noun-.7...

 

193

 

.2. 82.. 2. 2.3. .2. 9...... a... 3.... $6 a.
a... 2.6 2. :4... .2. 3.3 .2. 3.... v.2: 2.3... 8.8... has...
........................ .32 32.... 088... 2......
Es... .82....2. .232 8.3.. $3....
.2. 52.. a... 8nd. .2. 8nd. .2. 32.. .83.. 09...... 8.338;...
2. £3 .. £3 2. up... : «a... 0......
6.: 3.? 6.... 10ft 6.... 5a.? a... .86 8596.02
.2. «no... .2. no»... .2. N8.» 2... :3 8.3.02...
.2. 2w... .2. «8.... .2. Bad. .3. 8...... 1.8.2:...
. 89v . o8... . 23 . 83 38.3.5...
.2. 8.2. .2. 81- .2. 3»... 2. 5.. oguaE:
c2. 09...”- «I. 08.0. I. 89.0. 2:. End. gtofiéuma
.2. 80.... .2. «8.... a... 8a.... .2. 8...... 5.55-554.
«5.2% uucfimxmt 8.3.:
83...: «2.385398.
e8. 5.58.9.3... bacon»...
808.....qu b.2030
.. m8... 8858...... a... ..S 9.5.30 c.8083.
.. «3... 238.3... 5.9.35 -
t. and 828...... 58.330-
28.... a...
3.3.8.... 2.30 ..m... ..s. .558. .58
9......
2...-.33.... 83:83:30 153.5
2.. 313 i 32.8 .2. EN! .5. an... .83....
4a... 2.4.3.6 Jam .36 1w... .8... 3.0 #w .232
+ . .26.. .8 so + . .23 .2...
a .03... + . .25.. 6 .332 . .26..
K 6.32 ”n .332 425—

132 ....... 2.2.... s 393......

194

SQVQ 9...... SQVQ: unqva. £.VQ<

 

 

83 83 83 a:
t 2.3. a 22: .z 83. 55% 23$ .6
89m“ 83m 898 83a escape;
332k Stat. 8982 SSE. 325 8.5.5:
:36 8nd auto «8... «9225
83%;...—
R. .c on; E. .o «2 .o 8%.?» 83:8 3628353
856 ~83 ~33 ”8.: 88:8 aaaugéfia
$3 33 5... 82 25 3:3? 83:8 3.825-633
.. at».- .t to; : 83 a: 5.». Aggy £230 133—.8033
ago-SE90 Banta>
: «ta. : 83. t 2.3. : 2.3. can as, gags E30 :3:
a BE 1 53 z. 83 : on; aosgmbmpaé
< :66 < Sod < «36 < :3. 3038.5 :2: Ba casino 28083
a... «F to a... :3 .3. :3 .3 2 2. Sega 223?? .
2... 8rd 1:. hard 1.. 8rd :5 mum—6 «602 809.20.
238 a?
2:835 £30 .3: 8.. uses; 88
53.2% 5.2 .32.: .322...—
s. 83 i :3 z. 23 :. Ea €22.25 85 2o: - EOEQE 89o
..................... ..-..- muo_£8mfia§-§fia&<8aon
«warm?» 3.332% - 38.33 ”EMS
. Sod . 2o... . 83 _. 32 um a 35 as» a. 2:52
2:. word c 12.6 23 wad 2.3 «2.: 859562
8.3.: 8.5.1.”. $2.6
14% $8.16 d5 36 ma .8... ago #5 .032
+ _ .23 8:5 + _ E3 5
a .332 + _ .23 c 3.: _ .23
h .032 "n 3932 guy—

3...: =5— ...3... 8 53:3.

195

 

.2.. and. 6.... 8...? .o... 3.20. $5 m:
.2. 8nd .2. 8...... .2. 9...... 22.352. 8.8... 3.5....
.................. .52 322... 8.8... has...
E3... 2.2%... 3.2.3 3.8... $3..
.2. 32.. .2. 8...... .2. 82.. .83. ”$2.8 8.3.0892...
.2. 8.... z .8. .. 83 2.2
6... 306 .3. «mm... a... 03+ 8%....
.2. o5... .2. .8... .2. one... 350.08..
.2. find. .2. 2...... .2. 8...... 38......8
. 8v... . to... . 53 182.5...
.2. 5... a... new... .2. 3m..- o§.§%.m
coo N86- :0 hmrdn 3i 3N”: gag-flaunt.
.2. 8“,... .2. 83. .2. 8...... §Ea<.§...<
«9.39% 3293.3.» 8.3.:
83...... 3......8333
.2. :3. .2. En... e8. 5.58.9.3... $.30 ..m...
.2. 92.. .2. «no... fifiuaxmb.§>.a
. in... 2338...... ..m... 9... .9330 0.588...
d... 8.6 Saunas... 3.0.»?50 -
< 8m... 888... 8.33.0.
5.8.... 5....
2....85... $.30 a... ..s. 2.232 8....
.9......
3.8.8.. 2 8.3.3325 323.5
. .2.- .2. to...” z. 5.9 .923...
[.23. a2 a... law 28.. + id 5.25 .2....
.2... .... . .96.. 5. + . .25.. .23: =3. 8.3.2.. s 3252..
+ . .26.. .252 a .25...

n: .032

 

196

:5. v a :3 ”S. V a go ”we. v a a u_.Va <

 

 

o8.m 83 83 ,3
: 893 < 83 «B. F 3.3% 23$ .5
Sega 98.2 893 $55
3. 52. S. F E. .2 Stat 32.: 8.5::
:13 wood ..86 «8235
82.3.53
5.: 2:. 2:. 8398 E230 auufimoaééa
85.3 2.3m n83 amazon auscuoméfia
~35 :2. 83 3.5 8:335 83:8 3823.;
. 33 .2. 08d 2.. 3o.» Agé 88:8 gagsméozom
85:33.5 ousat¢>
.. 83. : 9:6. .. 3.3. 28.. a? 808235 £50 .55
t 83 .. 83 a 83 82.32% £35
< ~36 < Sad ¢ Sad 88383. .3: us ”95:20 o§§<
a... :3 .2. up 3 .2. a. to 8888.5 393.30 .
t. 33 i 83 z. 33 802 .8235.

has... 5?
flags £30 a“: as 353 88
«sarong .5 .23: 1:22....—
................ ....... fi.»§m§5§.§§%<8&5
«85.6%.: aS~33m - Eonuhmuw 3&3
. 83 . 8nd . $3 on 5 85 50> a. 35:2
2.. 83 t. 83 .1 83 83.9.2
83...: 818...”.— 32.5
d5 32 [am 98 + % £65 429—
a: .Em =. _ .23 § + _ .23 :3: =3. 8.3.5 8 San-=2:

+ _ E3 .082 5 382

“2 E52

197

APPENDIX L

FULL MODEL OF CAPACITY FOR LIFELONG LEARNING

198

< 96.7 .m .: ~25. .3. and <m0 ml

199

 

.2. 3.... a a 2...... .2. 88. v.2: 23... 8.8... is...
.2. «a... .m ... Sn... .2. ”a... V33 32.... 8.8... 2.3....
«as.» 828%.. «27%... 2.32.. .95....
.. ... g... .. ... .3... .. ... 32 2.33.. 0%..8 Sufiswé...
. ... Rm... .. .. 8N... .. ... «8... .. ... as... 0...:
.3. 2.....- a... *8..- .3. .8... a... 32.. 9.892....
. a? < 33 a... 2m. .3. 83 25.08..
.. ... 2.3 ._ o8..- .. ... 2.3. . ... at... 385......
.3. 2.2 .3. 8.... .3. :2 .3. an... Ease...
. an. . 3...... < «a. .. 5.. 8.32%.:
.. ... .2... .. ... 8...... .. ... 88. . .. a :3. .3553“...
.. ... .3... .m a 8...... .. .. 2...... .. ... 82 §§.§E<
«0.39% ogoguxwh 83.:
8......2. finance;
28.. 5.3 8°38... 3.30 5.:
808.3... £2.25
aofisaxm a... a... 0.8.30 0.8083.

8.3.5.... ”8.32.5 .

888... Sofia... -
5.3.... 5.3
8:883... $.30 a... .2... .553 .55
81.5
38:85.. 83:62.5 135.30
2:. 03.3 2: 03.3 o: ogdu :3 93.8 2.... «3.3 «9285
«5 .52 ..3 .2.. gm 8.3.3: mal 1.3.8.2.. Jflm .am a ....z dam...
+ m .282 + N .082 + .932 .5885. .082 .892 .3.

a. .082 a .352 .u .082 a .352 9.8.3 2.39.3 .3. 2.92.6

 

:5. Va ...... 2c. Va ..5 ”no. Va 8 £.Vn<

 

200

 

88.. 28... 83. 8.... ....
... .82.... .... 8...... ... 8...... .. 34.8 2.228 9......» .8
8...... 8...... 8...: 8...: 8.... 2.88....
.22.... 85.8..
«2...... 83.8 «2.38. .822. 28.82.

.822...
88... «8... 8.3 3...... «a... 8.2.32...
E... 8.... .8... .8... 88... 822...... 88:8 88.82.25...
88.... 82...... 82... an... 8...... 88:8 3288.82-25...»
...... .2.... «8.... N3... 88... 32.822...» 88:8 388.82.803.02
...... .2.... ... 8.8.. 2.. 8.... .... a... .... a... .822... 88:8 22.08.82.858.
8:32....ch 33......»
.. .. .8... 88.. 5.38882... ....25 ......
._ 58.: 8055.5. £835
...: .8... 2.2.2.8.... ...... ...... 88...... 2:888...

: E... .8282... 828.880 .

.. ... 83... 82.8... 88.828 -
.28.... ......
82.88.... 8.8.6 ...... ...... 8.88. .88
8982...... ...... 1.2.... .325...
... 3.... ... 8.... 822.2... ...... p82 . 2.28.82 8.8..
.. .. «3.... .. ... ...... 8.22.82.22.88. 82.882888
882%... 28.2.8... - 28.828... .282
.2. ...... ... :2 o... ... .8»... 28...... 82.52
...: 026 ...: «36 839502
83......» 8.2.5.... 9.2.8
l2. 2.... ...... .8. J88 8.8.82 ......m 82.8.»... 2.. 8. ... ....z .22...
+ n .282 + a .282 + .282 £283... ..282 .282 ......
2. .282 .m .282 a .282 2 .282 ...-...... 28.8.. .... .2828

 

 

. 3...- . fig..- . 8.5...- .. an..- 50 m:

m d 53. m a 388... m d 288... m ... 8.... v.25 28¢. 2:8... baa".

- . .......................... v83 325 8.8:. has...

«9.2» Scan?» 3262. 289: bin...

m ... 33.3 ... a “.822. .m ... $.33 m ... man... .53» om»..8 8.38.3.3“.

uuuuuuuuuuuuuuuuuuuuuuuuuuuuuu 032

v. a 5.8..- .m a 9.323. .m ... 83.8... m ... 3...- 88%». oz

.. «$.83. . 38$... ._ $.83 . «an... 860.83.

m ... «822.. .m ... «283.. .... a «83...... a .a «one. 385......

...... .83.... .m ... R83. .m a 93.8... .m a .3. .2335...—

.. awn?!— . 39mm. .. manna: .. an... Sufi—BEE

m a «$33 .... ... «28...: .m a 88%... m a an... 55.2-8.2

m ... 82.2. .m ... $38... .m ... 238... m ... «8.... 5.85.808...

«3.9% 35(th 8.33

335...: 3.529535

28.. 5.3 «8%... £30 in.

«denim mgr—95

.. Ewan—d Eogxm fiivfiumaozfloomgg

2:. 39.3.6 8°33 8293.30 .
«. ... 33.2. ”8:8... 850.

has“. a...»

EOE... $.80 a... ..8 #283 .58

.....6

3232.33: 8.3.8.9220 .258an

1... 82.2.2 2. Rost— < 382.6— 2: aflen $0885
ma $8.30 .% 2.6 Jam .8... 9.5 am .082

+ . .23 .o :5 + . .23 a...
a .352 + . .23 a .082 . .23 5mm
... .38.). ”n ...qu 28:. :5. 9.2.3 2333 .3. Eva-U

 

201

8:. V a 2.5 ”3. v a 2. ”ma. v a . ”—.Va <

 

 

. . . ....
.. 88.... .... 83...... .. ... Ex... 35...... 23...... ...u
«a S ......Q 8... .N 232.2...
32... 85......

$82.... $3.8: 8:89 ......ch
.93...
«.8... 2...... .8... 9...... 82.333.
E... 2.2. ...”... «a... 8:398 83...... 3823.555
88.... 03...... 8... .... .2.... 88:8 .....E...§.........B
8.... 8:. an... an... ...... 3%.? 83...... 3.3.8:...825
.... 2.32 : $2.... ...... 3.2.. ...... 2.... .3023... 8.2.8 .....fi...8:...83.o:
anaemic”. 3.3....»
.. ... $88... ... ... 92...... .. ... 888... .... ... 8.... e8. ...... 2.2.8.3... ......5 .3:
. .38... . R38... . :88... . 2...... 808.5%: 3.5.5
.... 2 $8... ...: .88....— ..: 8.3.... ... n8... gage... ...: ...... 9.8.3.6 0.5.3..

... 88:... .. .83.... .. 85.... .. n: ... 82.95.... 2.5.2.6 .
< 388... < $33... a a 883... .. ... a3... 30.6.... g... .
......8: ......
egg... ......5 6.: ...... 2:8... .58
38:3... ...... .32.: 1.6.5...
.... ENE... ... can... ... 22...... .... «...... 98.2.8: .3... 2oz . 8°33 8&5
........................ 9.22.8: ...... 33 - 23.3.3. 83x.
3.2%... .838: - 23:23.: 3&3
...: 633*”.— .3... finenm; ...: mug—*4 ...: 23..— DM— 3 v03..— nuaoruo .3892
...: manna—d 2... vac—Ed :5 autumn—A. a: em—d 8.3.502
83......» 88:28 «3.6
1.3m 03...... % .2.... gm .8... 8.6 Am... .08.).
+ . .23 ......o + . .93 .9...

H.. .232 + . .23 H.. .08.... . .26.. .25.
... .332 a .982 .252 ...... .....Eaa 23...... .... baa-u

 

202

... v2.7 .. 32.7 a «2... (..ch

 

...... a... .... n8... .... 2.... 2.533558225.

...................... ...... 32.... 2:8... ......a...

ES.» 828%. .232 38...... ....S...

...... Sm... .. ... 32.2... .m ... 82mm... .52.... 0.2.8 8338......

...................... 2.2

...... 8...... ...... 2:8... .3. $28... 88.3. oz

.. 8.... .. 23...... .. :28... 550.03..

...... 5.... ... a 3...... .. ... «82...... 28......5

a... ...... .. a 2.33.. .. ... 88%.. 38.3.3...

...... 8.... . 5...... < .332 ogfinm...

...... 8...... .. ... 22...... .. a 2.3.”... 80.35.32

.3. 8a.... a. ... :83... . ... «:31... 555.52....

3.8% «SEQ»... 23.:

8...-...: 0.229533

.w ... on... : .88.... 28.. ....s 25.683... ....So ......

.. ... .2.... _. 3.2.... 82.2.99... .23....

.. .. 2...... 8.50%... ...... ...... 0............o 0.80.5...
...... mc— .— 328.83..— mma_0..«c.§O .
< 2...? 82.3.... 892320 -

.....8... ......»

3923.... .....30 ...... ...... ......on .58

.825

.2.-.28.. ... 8.5.38.3... .3589

. 8....” . $22.... : .323... 28.2...

...m 3...... .... .mm 28.. Ana ..ng
.2.... .... + . .26. + . .25..
+ . .23 .... .982 ... .252 .2....

... .082 .232 ...... 2.28.. 23...... .... £850

203

80. Va ...: “3. Va .... ”no. Va . m_.VA_<

 

 

83 _ _ ,3
: 82a .. .38.... : RE ... 3.55m 2.58 .5
89.3 a «N 32.5....
SEE 35::
.898: 288$: 23.3:
5855
8.3 $3 «8... 82.3....»—
E... :3 :3 3:398 89:8 3.3.3.555
x: .8 ”88.8 «88.2 30:8 awfigamééa
E... :3 83 3.5 .3393 83:5 3823...-..9033
. «n3 ...: 3.8.. ... 322 £83.: 88:8 au§_3¢-=8§m
8:32.890 count!»
.2. «a... .m ... a: s... ... ... «.88... 5... 5.3 3°83... £36 .3:
._ o8... . 828... . has... 808.3%”. £35
... n8... ... 8:8... ... 888... so; .3: ...... ”~83“. 25°33.
._ e2... .. 33:... : 3.5.3 Eogé 9.23930 -
...... .8... < 283... .m ... 283... 888: 88.2.6 .
has... a?
gag... 9.26 .9: ...... 3.1.3 38
.8533”.— £< .22: 1.23....—
:. 85 ... $22. _ ... ~58 . £22.08 a... so: . 88.5.5. 89o
..................... .3055 a... $3 - 2°33 8&5
$82.?» ”.338: - 238%.... 9&3
... an. ... £33.. .... 9.82.. 8. ... 35 as»... 252
...: vn—d o: NSMN—d .2. ”33:. E59582
83...: 83.33.. 32.5
wa 2.... 1.5m 28.. ”mm 5.335
.5 gm .... + _ .23 + _ .26..
+ _ .33 a. .80: a .08: dam”.
u: .082 It: ...... 3...!— 2§ ...... £236

 

204

REFERENCES

205

References

Abukari, A. (2005). Conceptualising lifelong learning: A reflection on lifelong learning at
Lund University (Sweden) and Middlesex University (UK). European Journal of
Education, 40, 143-154.

Anaya, G. (1999). College impact on student learning: Comparing the use of self reported
gains, standardized test scores, and college grades. Research in Higher Education,
40(5), 499-526.

Association of American Colleges and Universities (AAC & U). (2007). College learning
and the new global century. Washington, D. C.: Author.

Arnold, K. D. (1995). Lives of promise: What becomes of high school valedictorians. San
Francisco: Jossey-Bass.

Astin, A. W. (1977). Four critical years: Eflects of college on beliefs, attitudes, and
knowledge. San Francisco: Jossey-Bass.

Astin, A. W. (1984). Student involvement: A developmental theory for higher education.
Journal of College Student DeveIOpment, 25. 297-308.

Astin, A. W. (1993). What matters in college? San Francisco: Jossey—Bass.

Astin, A. W., Oseguera, L., Sax, L. J., & Korn, W. S. (2002). The American freshman:
Thirty-five year trends, 1966-2001. Los Angeles: Higher Education Research
Institute, UCLA

Aulepp, U., & Delworth, U. (1976). Training manual for an ecosystem model. Boulder,
CO: WICHE.

Banning, J. H. (1989). Creating a climate for successful deve10pment: The campus
ecology manager’s role. In U. Delworth & G. Hanson (eds.), Student services: A
handbook for the profession (2ml ed., pp. 304.322). San Francisco: Jossey-Bass.

Barr, R. B., & Tagg, J. (1995). From teaching to learning: A new paradigm for
tmdergraduate education. Change, 27(6), 12.

Baxter Magolda, M. B. (1992). Knowing and reasoning in college: Gender-related
patterns in students ’ intellectual development. San Francisco: Jossey-Bass.

Baxter Magolda, M. B. (2004). Learning partnerships model: A fi'amework for promoting
self-authorship. In M. B. Baxter Magolda & P. M. King (Eds), Learning
partnerships: Theory and models of practice for self-authorship (pp. 37-62).

- Sterling, VA: Stylus.

206

Belenky, M. F., Clinchy, B. M., Goldberger, N. R., & Tarule, J. M. (1986). Women ’s
ways of knowing. New York: Basic Books.

. Biggs, J. B. (1993). From theory to practice: A cognitive systems approach. Higher
Education Research and Development, 12, 73 -85.

Blaich, C., Bost, A., Chan, E., & Lynch, R. (2004). Defining liberal arts education.
Unpublished manuscript. Retrieved September 5, 2008 from
http://www.wabash.edu/cila/docs/Defl.ibArtEdFinal.pd£

Blaich, C. & Wise, K. (2008). Overview of findings from the first year of the Wabash
National Study of Liberal Arts Education. Unpublished manuscript. Retrieved
April 26, 2009 fi'om
http://www.wabash.edu/cila/docs/ 1 l .13 .080verview°/o200f°/<:20First"/o20Year°/oZ0
Findings%20All%20website.pdf

Boyer, E. L. (1987). College: The undergraduate experience in America. San Francisco:
The Carnegie Foundation for the Advancement of Teaching.

Boyer Commission on Educating Undergraduates in the Research University. (1998).
Reinventing undergraduate education: A blueprint for America ’s research
universities. Stony Brook: State University of New York at Stony Brook.

Bradburn, N. M., & Sudman, S. (1988). Polls and surveys: Understanding what they tell
us. San Francisco: J ossey-Bass.

Bronfenbrenner, U. (1979). The ecology of human development: Experiments by nature
and design. Cambridge, MA: Harvard University Press.

Brophy, J. (2004). Motivation to learn, (2" Ed.). Mahwah, NJ: Erlbaum.

Brower, A. M., Golde, C. M., & Allen, C. (2003). Residential learning commtmities
positively afl‘ect college binge drinking. NASPA Journal, 40, 132-153.

Brown, M. K., & Rogers, J. C. (2005). Need for Cognition scale. Retrieved on September
7, 2008 from Center of Inquiry in the Liberal Arts at Wabash College website:
http:l/liberalarts.wabash.edu/cila/home.cfin?news_id=26l 1

Bmeau of Labor Statistics. (2006). Number of jobs held, labor market activity, and
earnings growth among the youngest baby boomers: Results from a longitudinal

survey. Washington, DC. Author. Retrieved from on December 1, 2007 from:
http://www.bls.gov/news.release/pdf/nlsoy.pdf

Cacioppo, J. T., & Petty, R. E. (1982). The need for cognition. Journal of Personality and
Social Psychology, 47,116-131.

207

Cacioppo, J. T., Petty, R., E., Feinstein, J. A., & Jarvis, B. G. (1996). Dispositional
differences in cognitive motivation: The life and times of individuals varying in
need for cognition, Psychological Bulletin, 119, 197-253.

Cacioppo, J. T., Petty, R. E., & Kao, C. F. (1984). The efficient assessment of need for
cognition. Journal of Personality Assessment, 48(3), 306—307.

Campus Calm website, (11. d. ). http://www.campuscalm.com/time_management.html

Center of Inquiry in the Liberal Arts at Wabash College. (n.d). Wabash National Study
of Liberal Arts Education: Liberal arts outcomes. Retrieved on September 7, 2008
from http://staged.wabash.edu/cila/home.cfin?news_id=2338

Chapman, J. & Aspin, D. (1997). Schools as centres of lifelong learning for all. In M.
Hatton (Ed). Lifelong learning: Policies, practices and programs, 97. (pp. 154-
167). APEC Publication.

Chickering, A., & Gamson, Z. (1987). Seven principles for good practice in
undergraduate education. AAHE Bulletin, 39. 3-7.

Chickering, A., & Gamson, Z. (1991). Applying the seven principles for good practice in
undergraduate education. San Francisco: Jossey-Bass.

Cohen, A. R., Stotland, E., & Wolfe, D. M. (1955). An experimental investigation of
need for cognition. Journal of Abnormal and Social Psychology, 51, 291-294.

Converse, J. M., & Presser, S. (1989). Survey questions: Handcrafiing the standardized
questionnaire. Newbury, CA: Sage.

Cox, B. E., & Orehovec, E. (2007). Faculty-student interaction outside the classroom: A
typology from a residential college. The Review of Higher Education, 30, 343-
362.

Creswell, J. W. (2003). Research design: Qualitative, quantitative, and mixed methods
approaches. Thousand Oaks, CA: Sage.

Cropley, A. J. (1977). Educational brokering: Access to lifelong education. International
Review of Education, 23, 339-343.

Cropley, A. J. (1978). Some guidelines for the reform of school curricula in the
perspective of lifelong learning. International Review of Education, I, 21-33.

Cross, W.E., & Fhagen-Smith, P. (2005). Nigrescence and ego identity development —
Accounting for differential black identity patterns. In M. E. Wilson & L. E. Wolf-
Wendel (Eds), ASHE reader on college student development theory (pp. 259-
268). Boston: Pearson Publishing.

208

Cruce, T., Wolniak, G. C., Seifert, T. A., Pascarella, E. T., & Blaich, C. (2006). Impacts
of good practices on cognitive development, learning orientations, and graduate
degree plans during the first year of college. Journal of College Student
Development, 47(4), 365-3 83.

Dave, R. H. (1976). Foundations of lifelong education. Hamburg: UNESCO Institute for
Education, and Oxford: Pergamon Press.

Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in human
behavior. New York: Plenum.

Deci, E. L., & Ryan, R. M. (2000). The “what” and “why” of goal pursuits: Human deeds
and the self-determination of behavior. Psychological Inquiry, I I, 227-268.

Deci, E. L., & Ryan, R. M. (2002). Hancflrook of self-determination research Rochester,
NY: University of Rochester Press.

De Leeuw, E. D. Hox, J. J ., & Dillman, D. A. (2008) Cornerstones of survey research. In
E. D. De Leeuw, J. J. Hox, & D. A. Dillman (Eds.), International handbook of
survey methodology (pp. 1-16). New York: Taylor & Francis.

Duderstadt, J. J. (2000). A university for the 21st century. Ann Arbor, MI: The University
of Michigan Press.

Duke, A. (1991). Importing Oxbridge: English residential colleges and American
universities. New Haven, CT: Yale University Press.

Dweck, C. S., & Leggett, E. L. (1988). A social/cognitive approach to motivation and
personality. Psychological Review, 95, 256-273.

Gaff, J. G. (1970). The cluster college. San Francisco: J ossey-Bass.

Gabelnick, F. MacGregor, J. Matthews, R. S., & Smith, B. L. (1990). Learning
communities: Creating connections among students, faculty, and disciplines. (New
Directions for Teaching and Learning No 41). San Francisco: Jossey-Bass.

Gagne’, M. (2003). The role of autonomy support and autonomy orientation in prosocial
behavior engagement. Motivation and Emotion, 27, 199—223.

Gamson, Z. F. (2000). The origins of contemporary learning communities: Residential
colleges, experimental colleges, and living-learning communities. In D. DeZure
(Ed.) Learning from change: Landmarks in teaching and learning from Change
magazine 1969-1999. Sterling, VA: Stylus.

209

George Washington University Academic Success Center, (11. d.). Make a to-do list.
Retrieved on April 8, 2009 fiom
http://gwired.gwu.edu/counse1/asc/index.gw/Site__ID/46/Page_ID/14542/

Gilligan, C. (1993). In a ditferent voice: Psychological theory and women ’s ’
development. Cambridge, MA: Harvard University Press.

Gonyea, RM. (2005). Self-reported data in institutional research. In P. Umbach (Ed),
Survey Research: Emerging Issues (New Directions for Institutional Research
Monograph No. 127; pp. 73-89). San Francisco: Jossey-Bass.

Gorard, S., & Selwyn, N. (2005). What makes a lifelong learner. Teachers College
Record, 107, 1193-1216.

Guardia, J. R. & Evans, N. J. (2008). Factors influencing the ethnic identity development
of Latino fi'atemity members at a Hispanic serving Institution. Journal of College
Student Development, 49. 163-181.

Guskin, A. E. (1994). Reducing student costs & enhancing student learning: Part II-
Restructm'ing the role of faculty. Change, 26, (5), l6.

Hart, D., & Smith, T. B. (1993). Residential colleges: Vestige or model for improving
college residence halls. In T. S. Smith (Ed.), Gateways: Residential colleges and
the freshman year experience, 14, (pp. 2531). Columbia, SC: The University of
South Carolina, National Resource Center for the First-year Experience and
Students in Transition.

Hawkins, H. (1999, Winter). The making of the liberal arts college identity. Daedalus,
128,1-2.

Hayek, J. C., & Kuh, G. D. (1998). The capacity for life-long learning of college seniors
in the mid-19808 to the mid-19908. Paper presented at the Annual Meeting of the
Association for the Study of Higher Education. Miami, FL.

Hayek, J. C., & Kuh, G. D. (1999). College activities and environmental factors
associated with the development of life-long learning competencies of college
seniors. Presented at the Annual Meeting of the Association for the Study of
Higher Education. San Antonio, TX.

Herzog, S. (2007). The ecology of learning: The impact of classroom features and
utilization on student academic success. (New Directions for Institutional
Research No 135). New York: Wiley Periodicals.

Hirt, J. B. (2006). Where you work matters. Lanham, Maryland. University Press of
America.

210

Hohenbary, J ., Bryan, R. A., Cross, M. B., & Cruse, L. S. (1993). Us and them. In T. S.
Smith (Ed), Gateways: Residential colleges and the fieshrnan year experience,
14. (pp. 19-23). Columbia, SC: The University of South Carolina, National
Resource Center for the First-year Experience and Students in Transition.

Hox J. J. (2002). Multilevel analysis: Techniques and applications. Mahwah, NJ:
Lawrence Erlbaum Associates.

Ilardi, B. C., Leone, D., Kasser, T., & Ryan, R. M. (1993). Employee and supervisor
ratings of motivation: Main effects and discrepancies associated with job
satisfaction and adjustment in a factory setting. Journal of Applied Social
Psychology, 23, 1789-1805.

Inkelas, K. K., Brower, A. M., Crawford, 8., Hummel, M., Pope, D., & Zeller, W. J.
(2004). National study of living-learning programs: 2004 report of findings.
Retrieved November 12, 2007, from
http://www.livelearnstudy.net/studyresults.hnnl

Inkelas, K. K., Longerbeam, 8., Leonard, J. B., & Soldner, M. (2005, November).
Understanding differences in student outcomes by types of living-learning
programs: The development of two typologies. Paper presented at the annual
meeting of the Association for the Study of Higher Education, Philadelphia, PA.

Inkelas, K.K., & Weisman, J. L. (2003). Different by design: An examination of student
outcomes among participants in three types of living-learning programs. Journal
of College Student Development, 44. 335-368.

Inkelas, K. K., Zeller, W. J ., Murphy, R. K., & Hummel, M. L. (2006). Learning moves
home. About Campus, 10(6): 10—16.

International Labour Organization. (2000). Lifelong learning in the twenty-first century:
The changing roles of educational personnel. Geneva: International Labour Ofice.

Jerome, J. (2000). The living-learning community. In D. DeZure (Bi), Learningfiom
change: Landmarks in teaching and learning fiom change magazine 1969-1999
(pp. 117-118). Sterling, VA: Stylus. (Original work published in 1971)

Johnson, R. M. (2005). Minority stress and the well-being of sexual minority college

students viewed through the lens of a (bio)ecological model of Man
development. Unpublished doctoral dissertation, Iowa State University.

J osselson, R. (1987). Finding herself: Pathways to identity development in women. San
Francisco: Jossey-Bass.

J osselson, R. (1996). Revising herself: The story of women’s identity from college to
midlife. New York: Oxford University Press.

211

Knapper, C. K., & Cropley, A. J. (2000). Lifelong Learning in Higher Education (3” e09.
Sterling, VA: Stylus.

Kodarna, C. M., McEwen, M. K., Liang, C. T., & Lee, S. (2001). A theoretical
examination of psychosocial issues for Asian Pacific American Students. NASPA
Journal, 38.411-437.

Krefi, I., & De Leeuw, J. (1998). Introduction to multilevel modeling. Thousand Oaks,
CA: Sage.

Kuh, G. D. (1995). Cultivating “high-stakes” student culture research. Research in
Higher Education, 36. 563-576.

Kuh, G. D. (1999). Setting the bar high to promote student learning. In G. S. Blimling &
E. J. Whitt (Eds.), Good practice in student aflairs: Principles to foster student
learning (pp. 67-90). San Francisco: Jossey-Bass.

Kuh, G. D., Kinzie, J ., Schuh, J. H., Whitt, E. J. and Associates. (2005). Student success
in college: Creating conditions that matter. San Francisco: Jossey-Bass.

Kuh, G. 1)., Schuh, J. H., Whitt, E. J., & Associates. (1991). Involving colleges:
Successfitl approaches to fostering student learning and personal development.
San Francisco: Jossey-Bass.

Kuh, G. D., Vesper, N., Connolly, M., & Pace, C. R. (1997). College Student
Experiences Questionnaire: Revised norms for the third edition. Bloomington, IN:
Center for Postsecondary Research and Planning, Indiana, University.

Lenning, O. T. & Ebbers, L. H. (1999). The Powerful Potential of Learning
Communities: Improving Education for the Future. ASHE-ERIC Higher Education
Reports, 26, 1-137.

Lewin, K. (1936). Principals of topological psychology. New York: McGraw.

Lyman Briggs College Website. (n.d.). About Lyman Briggs College. Retrieved on
September 5, 2008 fi'om http://www.lymanbriggs.msu.edu/about/about_lbc.cfin

Macy, T. J. (1993). Meeting the needs of today’s students: the evolution of a residential
academic program. In T. S. Smith (Ed), Gateways: Residential colleges and the
fieshrnan year experience, 14. (pp. 35-40). Columbia, SC: The University of
South Carolina, National Resource Center for the First-year Experience and
Students in Transition.

Magolda, P. M. (1994). A quest for community: An ethnographic study of a residential
college. Unpublished doctoral dissertation, Indiana University.

212

Marton, F ., & Saljo, R. (1976). On qualitative differences in learning: 1]. Outcome as a
function of the learner’s conception of the task. British Journal of Educational
Psychology. 46. 115- 127.

Marton, F ., & Stiljo, R. (1984). Approaches to learning. In F. Marton, D, Hounsell & N
Entwistle, (Eds). The experience of learning (pp. 36-55). Edinburgh: Scottish
Academic Press:

Matthews, R. S., Smith, 3., L., MacGregor, J., & Gabelnick, F. (1997). Creating learning
communities. In J. Grafi‘, J. Ratclifl‘, and Associates (Eds.), The handbook of the
undergraduate curriculum (pp. 457-475), San Francisco: Jossey-Bass.

Mayhew, M. J ., Wolniak, G. C., & Pascarella, E. T. (2008). “How educational practices
affect the development of life-long learning orientations in traditionally-aged
undergraduate students. Research in Higher Education, 49, pp. 337-35 6.

McCaslin, M., & Good, T. (1992). Compliant cognition: The misalliance of management
and instructional goals in current school reform. Educational Researcher, 21 (3),
4-1 7.

McCombs, B. L. (1991). Motivation and lifelong learning. Educational Psychologist, 26,
1 17-127. ‘

Michalak, S. J ., & Robert, J. F. (1981). Research productivity and teaching effectiveness
at a small liberal arts college. Journal of Higher Education, 52, 6. 578-597.

Moos, R. H. (1976). The human context: Environmental determinants of behavior. New
York: Wiley.

Moos, R. H. (1979). Evaluating educational environments. San Francisco: J ossey-Bass.

Moos, R. H. (1986). The human context. Environmental determinants of behavior.
Malabar, FL: Krieger.

Moos, R. H., & Insel, P. M. (1974). Issues in social ecolog: Human milieus. Palo Alto,
CA: National Press Books.

National Center for Educational Statistics. (2004). 2003-2004 National Postsecondary
Student Aid Study. (NPSAS:04). U. S Department of Education. Washington, DC:
National Center for Education Statistics.

National Center for Educational Statistics. (2006). Profile of undergraduates in US.
postsecondary education institutions: 2003—04: With a special analysis of
community college students (N CES 2006-184). US. Department of Education.
Washington, DC: National Center for Education Statistics.

213

OECD. 1996. Lifelong Learning for all: Meeting of the education committee at
ministerial level, 16-1 7 January 1996. Paris: OECD.

O’Hara, R. J. (n.d). Making the most of college: A message for parents and students from
the collegiate way website. Retrieved April 15, 2008 from

http://collegiateway.org/parents-students/

O’Hara, R. J. (2001). How to build a residential college. Planning for Higher Education,
30(2), 52-57.

Pace, C. R. (1985). The credibility of student self-reports. Los Angeles: University of
California, Center for the Study of Evaluation, Graduate School of Education.

Pascarella, E. T., & Terenzini, P. T. (1991). How college afibcts students. San Francisco:
Jossey-Bass.

Pascarella, E. T., & Terenzini, P. T. (2005). How college affects students. San Francisco:
Jossey-Bass.

Pascarella, E., Wolniak, G., Seifert, T., Cruce, T., & Blaich, C. (2005). Liberal arts
colleges and liberal arts education: New evidence on impacts. San Francisco:
Jossey-Bass/ASHE.

Pasque, P. A., & Murphy, R. (2005). The intersections of living-learning programs and
social identity as factors of academic achievement and intellectual engagement.
Journal of College Student Development, 46(4), 429-440.

Peterson, R. D. (2006). White, male and “minority”: A stuay of racial consciousness
development among White male undergraduate students at a public historically
Black university. Unpublished doctoral dissertation, Iowa State University.

Pike, G. R. (1995). The relationships between self-reports of college experiences and
achievement test scores. Research in Higher Education, 36, 1-22.

Pike, G. R. (1996). Limitations of students’ self-reports of academic development as
proxies for traditional achievement measures. Research in Higher Education,
37(1), 89-114.

Pike, G. R. (1999). The efl’ects of residential learning communities and traditional
residential living arrangements on educational gains during the first year of
college. Journal of College Student Development. 40(3), 269-284.

Pike, G. R., Schroeder, C.C., & Berry, T. R. (1997). Enhancing the educational impact of
residence balls: The relationship between residential learning communities and
first-year college experiences and persistence. Journal of College Student
Development. 38(6), 609-621.

2l4

Pizzolato, J. E. (2003). Developing self-authorship: Exploring the experiences of high-
risk college students. Journal of College Student Development, 44. 797-812.

Porter, S. R., & Swing, R. (2006). Understanding how first-year seminars affect
persistence. Research in Higher Education, 4 7. 89-109

Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and
data analysis methods. Thousand Oaks, CA: Sage.

Renn, K. A. (2003). Understanding the identities of mixed-race college students through
a developmental ecology lens. Journal of College Student Development, 44. 383-
403.

Renn, K. A. (2004). Mixed race students in college: The ecology or race, identity, and
community. Albany, NY: SUNY Press.

Renn, K. A., & Arnold, K. D. (2003). Reconoeptualizing research on college student peer
culture. Journal of Higher Education, 74. 261-291.

Ryan, M. B. (1992). Residential colleges. Change, 24. 26-35.

Ryan, M. B. (1993). Residential colleges: A historical context. In T. S. Smith (Ed),
Gateways: Residential colleges and the freshman year experience, 14. (pp. 11-18).
Columbia, SC: The University of South Carolina, National Resource Center for
the First-year

Salant, P., & Dillman, D. A., (1994). How to conduct your own survey. New York: John
Wiley & Sons.

Schuh, J. H. (1990). Streamlining the ecosystems approach to residence hall environment
assessment. NASPA Journal, 27, 185-191.

Schuh, J. H., & Veltman, G. C. (1991). Application of an ecosystem model to an oflice of
handicapped services. Journal of College Student Development, 32. 236-240.

Schuman, S. (2005). Old main: Small colleges in twenty-first century America.
Baltimore: Johns Hopkins University Press.

Seifert, T. A. (2006). Effect of college mq’or and its context on 21" century knowledge
economy competencies. Unpublished doctoral dissertation, University of Iowa.

Seifert, T. A., Goodman, K. M., Lindsay, N., Jorgensen, J. D., Wolniak, G. C., Pascarella,

E. T., & Blaich, C. (2008). The effects of liberal arts experiences on liberal arts
outcomes. Research in Higher Education, 49, 107-125.

215

Shapiro, N. S., & Levine, J. H. (1999). Creating learning communities: A practical guide
to winning support, organizing for change, and implementing programs. San
Francisco: Jossey-Bass.

Smith, B. L. (2001). The challenge of learning communities as a growing national
movement. Peer Review, 3. 2-4.

Smith, T. (1994). Integrating living and learning in residential colleges. In C. C.
Schroeder & P. Mable (Eds.), Realizing the educational potential of residence
halls (pp. 241-265). San Francisco: Jossey-Bass.

Smith, J. & Spurling, A. (2001). Understanding Motivation for Lifelong Learning.
Leicester, UK: National Institute of Adult Continuing Education (N IACE).

Snider, K. J. G., and Venable, A. M. (2000). Assessing learning community
efl‘ectiveness: A multi-campus approach. Presented at the Association for
Institutional Research Annual Meeting, Cincinnati, OH.

Stassen, M. (2003). Student outcomes: The impact of varying living-learning community
models. Research in Higher Education, 44, 581-613.

Strange, C. C., & Banning, J. H. (2001). Educating by design: Creating campus learning
environments that work. San Francisco: Jossey-Bass.

Svinicki, M. D. (2004). Learning and motivation in the postsecondary classroom. Bolton,
MA: Anker.

Thelin, J. (2004). A history of American higher education. Baltimore, MD: Johns
Hopkins University Press.

Torres, V. (2003). Influences on ethnic identity development of Latino college students in
the first two years of college. Journal of College Student Development, 44. 532-
547.

Tuijnman, A. C., & Bostrom, A. (2002). Changing notions of lifelong education and
lifelong learning. International Review of Education, 48, 93-110.

Tuijnman, A. C., Kirsch, 1., & Wagner, D.A. (1997). Adult basic skills: Innovations in
measurement and policy analysis. Cresskill, NJ: Hampton.

Umbach, P. D. (2004). Web surveys: Best practices. (New Directions for Institutional
Research N0121). New York: Wiley Periodicals.

University of Kansas Counseling and Psychological Services website (n. d.) 10 steps to

organization at college. Retrieved on April 8, 2009 from
http://www.caps.ku.edu/selflrelp/organization2.shtrnl

216

University of Michigan Residential College Website (n. d.)What is the RC? University of
Michigan Residential College. Retrieved on April 15, 2008 from
http://www.rc.lsa.umich.edu/html/l__lwhat__is_rc.htm

US. Department of Education. (2006). A test of leadership: Charting the fitture of [1.8.
higher education. Washington, DC.

Venegas, K. M. (2005). Aid and admission: Financial aid narratives of eight pre-college
urban Latinas. Unpublished doctoral dissertation, University of Southern
California.

Wabash National Study of Liberal Arts Education. (2008). Effective practices and
experiences fiom the Wabash National Study. Retrieved on September 7, 2008
from:
http://www.wabash.edu/cila/docsl08.l9.08GP°/o203cales%20summary%20All%20
FinalAdobe5.pdf

Wawrzynski, M. R. & Jessup-Anger, J. E. (in press). From Expectations to Experiences:
Using a Structural Typology to Understand First-year Student Outcomes in
Academically-based Living-learning Communities. Journal of College Student
Development. '

Wawrzynski, M. R., Jessup-Anger, J. E., Helman, C., Stolz, K., & Beaulieu, J. (in press).
Exploring students’ perceptions of academically-based living-learning
communities. Journal of College Student Affairs.

Wigfield, A., & Eccles, J. (2000). Expectancy-value theory of achievement motivation.
Contemporary Educational Psychology, 25, 68-81.

Zhao, C. D., & Kuh, G. D. (2004). Adding value: Learning communities and student
engagement. Research in Higher Education, 45. 115-138.

217