Micaela Mercado, Ph.D. Candidate
School of Social Work
University of North Carolina at Chapel Hill
October 9, 2011
1
Examining the Effects of Learning
Communities on Students’ Engagement in
Community Colleges
Micaela Mercado, Ph.D. Candidate
School of Social Work
University of North Carolina at Chapel Hill
Contact email:
[email protected]
18th National Conference on Students in Transition, St. Louis, MO
October 9, 2011
2
Introduction
Theory
Methods
Results
Implications
3
Compared to 4-year institutions…
4
Parental education level by race/ethnicity, 2003-04 (n=7,900)
SOURCE: U.S. Department of Education, National Center for Education Statistics, 2003-04 Beginning Postsecondary Students Longitudinal
Study, First Follow-up (BPS:04/06).
5
3-years
6-years
6
Percent Change
In Employment,
by education
2008-2018 (projected)
7
The Community College Environment:
Factors conducive to promoting student engagement, persistence, and degree attainment
Challenge
Educational Institution
Supports
At-risk students
lack academic &/or
social preparedness
Informal and formal
environmental structures:
Programs, services, courses,
study groups, professional
organizations, learning
communities, study skill
courses, mentoring
Assistance navigating the
system, information, resources,
guidance, opportunities
8
academic courses integrated within
a social environment designed to
promote learning and engage
students in class/college
assumption that students who
participate build relationships
with faculty & peers are more
likely to succeed in college
9
Research Questions
1. Which students participated in LCs?
2. What was the impact of LCs on active-collaborative learning?
Student-faculty interactions? And, students’ academic
performances?
3. Did the impact of LCs vary by minority and first-generation
status?
10
Data Source: Community College Survey of Student Engagement
Intervention
Sample
Completed
Plan to Enroll
Learning Community
35,761
12,816 (36%)
22,945 (64%)
Completed Learning Community:
Variable
Female
Frequency (%)
8,248 (64%)
Race:
Variable
Frequency (%)
Started here
9,435 (74%)
Full-time
10,441 (82%)
White
8,301 (65%)
Black
1,645 (13%)
Certificate
1,034 (8%)
Hispanic
1,512 (12%)
AA
5,449 (43%)
Traditional
8,601 (67%)
Both
3,411 (27%)
First-generation College
3,375 (33%)
Work: 20 hrs < LT
7,467 (59%)
HS diploma/GED
9,893(77%)
Dependents: 1+hrs
7,645 (60%)
Degree goal:
11
Analytical Model
Independent Variables
Pre-College Characteristics:
• Gender
• Race
• Age
• Generational status
• Marital status
• Academic credential
• College status
• Enrollment status
• Degree goal
• Work hours
• Income
• Dependents
Intervention
(Treatment)
Learning community
Outcome
Behavior: Engagement
Active-collaborative learning
Student-faculty interactions
Academic performance
Persistence
Degree attainment
College Experience
• Perceived college support
• Social service participation
• Peer relations
• Instructor relations
• Administrative relations
• Intervention control
12
Dependent Variables
Active-Collaborative Learning scale
Student-Faculty Interactions scale
Asked questions in class or contributed to
class discussions
Used email to communicate with an
instructor
Made a class presentation
Discussed grades or assignments with an
instructor
Worked with other students on projects
during class
Talked about career plans with an instructor
or advisor
Worked with classmates outside of class to
prepare class assignments
Discussed ideas from your readings or
classes with instructors outside of class
Tutored or taught other students (paid or
voluntary)
Worked with instructors on activities other
than coursework
Participated in a community-based project as
a part of a regular course
Discussed ideas from your readings or
classes with others outside of class (students,
family members, co-workers, etc.)
Scale: 0=Never 1=Sometimes 2=Often 3=Very often
13
Propensity Score Weighting
Neyman-Rubin counterfactual framework
• observed outcome
• unobserved outcome
Yi = WiY1i + (1 – Wi) Y0i
Y0i, Y1i two potential outcomes;
Wi=1 denotes receipt of treatment,
Wi=0 denotes no receipt of treatment; and
Yi is the measured outcome variable.
Estimates of counterfactual are examined by:
E(Y0|W=0) comprises the mean outcome of the nontreatment group; &
E(Y1|W=1) comprises the mean outcome of the treatment group.
τ= E(Y1|W=1) – E(Y0|W=0).
14
15
RQ1: Which students participate in LCs?
Logistic regression predicting participation in LCs (n= 35,761)
Variable
Log Odds
S.E.
Gender: Men
0.77***
0.02
Race: White
African American
0.57***
0.02
Hispanic
0.65***
0.03
Other
0.82***
0.04
Age: Traditional age student
0.97
0.03
First-generation status
0.85***
0.02
Academic credential
1.65***
0.06
Started at college
1.01
0.03
Enrollment: Full-time
1.54***
0.05
p<.05; * p<.01; *** p<.001***
16
RQ1: Which students participate in LCs?
Lower odds of participating in LCs
• Men
• African American students
• Hispanic students
• First-generation college students
Higher odds of participating in LCs
• Students with a certificate or college degree
• Full-time students
17
RQ 2 & 3: The Effect of LCs on Outcomes
ActiveCollaborative
Learning
Student-Faculty
Interactions
Academic
performance
(College GPA)
Avg.
Treatment Effect
0.83 (.11)***
0.31 (.09)***
0.04 ( .02)
LC*Gender
-0.26 (.09)**
LC*African American
LC*Enrollment
0.28 (.12)*
0.25(.11)*
*Hispanic and First-generation status
18
0.38(.09)***
0.05(.02)*
• Clustering
• Covariate structure
• Unobservables
• Implementation
• CC population
• Institutional/formal programs
19
• Methodology
• Treatment effects
Implications for Practice
Implications
Identify and strengthen key programmatic features of LCs to better promote
active-collaborative learning, student-faculty interactions, and academic
performance
Assess students’ academic and social progress over time
Build mechanisms within curricula to promote meaningful learning activities,
collaborative learning between students, and student-faculty interactions
Assist Men, First-Generation, Hispanic and African American students
actualize their educational objectives
20
Implications for Policy
Study
Implications
Not all CCs offer LCs
Create opportunities for academic success
LCs effectiveness
Complement LCs with academic and social
services that target engagement, and academic
performance
Long-term effects of LCs
Sustain efforts that increase students
academic and social involvement at the
institution (i.e., data-driven systems)
21
22