Skip Navigation
Funding Opportunities | Search Funded Research Grants and Contracts

IES Grant

Title: Mapping Barriers to Community College Completion Among Older Learners: Identifying Malleable Factors to Improve Student Outcomes
Center: NCER Year: 2016
Principal Investigator: Cummins, Phyllis Awardee: Miami University
Program: Postsecondary and Adult Education      [Program Details]
Award Period: 3 years (7/1/2016-6/30/2019) Award Amount: $1,390,425
Goal: Exploration Award Number: R305A160156
Description:

Co-Principal Investigator: Peter Bahr (University of Michigan)

Purpose: The goal of this exploratory project is to identify malleable factors at the student, instructor/classroom, and institution levels that can lead to improved education and labor market outcomes for older workers. Identifying malleable factors for adult students ages 40 to 64 has the potential to help them remain competitive in the labor market as they seek to upgrade their skills and credentials in response to demands within the increasingly technological and information-based economy. This research will contribute to the limited current base of knowledge on factors that lead to college persistence and degree completion for older adults, and the ways in which community colleges can structure instruction, support, and programs to meet their needs.

Project Activities: This mixed-methods project includes quantitative and qualitative analyses, as well as triangulation of findings from the two analyses. The early stage of the project involves acquiring and coding a large amount of quantitative administrative data from the Ohio Longitudinal Data Archive. In addition, the research team will interview educators, support staff, and students at community colleges across Ohio about their interactions with older students. The middle project stage includes extensive analysis of the quantitative and qualitative data to identify strategies for supporting older students. In the final stage of the project, researchers will triangulate findings from the two analyses and conclude by proposing testable intervention strategies for increasing degree completion among older students.

Products: Researchers will produce preliminary evidence of potentially promising practices and strategies for improving postsecondary persistence and degree completion for older students. The researchers will generate policy briefs and targeted topical reports and disseminate them to broad audiences of policymakers and educators, and will also produce peer-reviewed publications.

Structured Abstract

Setting: The study will include all 23 community colleges in Ohio. The colleges are dispersed across the state in urban, suburban, and rural settings.

Sample: The quantitative sample will include all adult students (ages 18–64) in 15 cohorts who entered an Ohio community between 2000 and 2014. Mean enrollment of first-time students over the period was approximately 48,000.

Intervention: The researchers will focus on intervention strategies at the individual, classroom and institution levels that can improve the academic performance and postsecondary attainment of older college students. The strategies will include changes to standard classroom instruction, standalone student success courses, enhanced support services for older students, and institutional practices (e.g. flexible class scheduling) that accommodate the schedules of employed students. The researchers are seeking strategies that improve students' study skills, reduce their anxiety about tests and schooling in general, and promote their engagement with peers, instructors, and support staff.

Research Design and Methods: For the quantitative analysis, researchers will use a correlational design for this exploratory project that distinguishes factors operating at the individual, classroom, and institution levels. Although the modeling strategy clearly distinguishes between student characteristics, student engagement, and features of the learning context, it does not rule out the possibility that unobserved student characteristics (e.g. motivation and academic skill) determine patterns of engagement or choices of learning context. Despite these potential confounds, relationships observed between student engagement, learning contexts, and the outcomes of interest should suggest promising intervention strategies. For the qualitative analysis, researchers are using a stratified purposeful sampling strategy for choosing administrators, faculty, and staff at each of the 23 colleges for semi-structured interviews. After the first round of interviews, researchers will expand the sample by identifying respondents in key units capable of answering questions pertinent to the research aims of the project. Stratification within units will guarantee that researchers obtain insights from key subgroups of administrators and educators.

Control Condition: Due to the exploratory nature of the research design, there is no control condition.

Key Measures: Researchers are focusing on education as well as labor market outcomes. The main distal education outcomes are completion of a credential (i.e. an associate's degree or certificate), or transfer to a 4-year institution within 4 years of college entry. The main labor market outcomes include earnings, number of weeks worked, and employment status within each quarter. Proximal measures of progression through college include completion of a college-level course in math and/or English, GPA, and degree-applicable credits in the first year following initial enrollment in college. Quantitative measures of malleable factors will include participation in a student success course, receipt of financial aid, and classroom instruction by a full-time faculty member. Qualitative measures of malleable factors will include the structure of student orientation sessions, delivery of tutoring, and characteristics of job placement services.

Data Analytic Strategy: For the quantitative analysis, researchers will use multi-level models with students at level one, classrooms at level two, and institutions at level three. The modeling strategy will accurately decompose variance across these three levels, allowing the researchers to identify the most promising level(s) to intervene for the benefit of students. Random error terms at levels two and three will allow researchers to distinguish variation that can be explained by classroom and institution-level factors in the model from variation that remains unexplained. By interacting a measure for each student's age with measures of his/her engagement and learning context, the researchers will identify factors that appear particularly important for the success of older students.

For the qualitative analysis, researchers will use a constant comparative methodology in which themes from early interviews and focus groups will inform questions asked in subsequent interviews and focus groups. The coding process will also be iterative, beginning with open coding to generate a large number of constructs from within the data, and then moving to axial coding in which a core category is chosen and analyzed in relation to all other categories.


Back