Overview of the Current Research on Learning Communities

This is an important time to be conducting this study. For several decades research has been conducted, which demonstrates that learning communities are an important intervention to support students. Learning communities come in many different forms, but most include linking courses together in order for students to form a community around different educational topics in order to support learning. Learning communities also provide students a chance to form stronger relationships with each other and their instructors, engage more deeply with the integrated content of the courses, and access extra support, making it more likely they will pass their courses, persist from semester to semester, and graduate with a credential. There have been several meta-analyses of the research on learning communities, including Learning Community Research and Assessment: What We Know Now and Learning Communities and Student Success in Postsecondary Education (Price, 2005).

In addition, a recent study by MDRC included control groups and studied multiple institutions. The MDRC study found that learning communities in community colleges had a “modest” positive impact. The study offers reliable findings on the effectiveness of learning communities. It examines the long-term impacts of a learning community program at one college as well as looks at the short-term impacts of programs at six community colleges. Beyond these investigations, there still remains a dearth of literature that is rigorous, uses multiple institutional types and compares different types of learning communities among different populations. By looking at different types of institutions and learning communities and having findings that can be generalized to a broader group of settings, our study will have greater external validity than the MDRC study. In summary, the findings from the proposed study will have strong internal and external validity.

Limitation of the Current Learning Community Research

Single Institutions and Lack of Control Group

Overall almost all studies looking at learning communities have been conducted at single institutions and often without control groups. (Price, 2005). In addition past research has helped to describe different types of learning community models (Goodsell-Love & Tokuno, 1999; Shapiro & Levine, 1999). What we don’t know in detail from these descriptions is how these different models affect students’ outcomes generally, or how they may impact different student populations specifically. As Price (2005) notes: “A demonstration could include a differential impact study that compares two different learning community models at a specific site against the control group; for example, the evaluation could compare one-term with two-term learning communities, learning communities with and without student support services, or learning communities with and without a financial voucher for textbooks.” Price notes no such studies exist.

Limited Scope of Outcomes

Most research tends to be descriptive and focuses on describing programmatic elements. Almost all of the impact literature is based on self-reported data and the literature based on satisfaction surveys rely on reflections made long after the students have experienced the various program elements, making the claims vulnerable to concerns about validity. Also, in nearly every study, GPA is used as the measure with which academic success is assessed (Andrade, 2007). Generally, learning communities cohorts’ grades are compared with non-learning communities students’ grades. Any noticeable improvement is then lauded as the positive effect of the learning community (Andrade, 2007). These studies are not randomized controlled trials; they are just comparisons to those not in the learning communities. While the conceptual literature stresses sense of community and motivation or self-confidence as the important indicators of success – these aspects are not studied empirically (Andrade, 2007; Price, 2005). Grades do not help understand the underlying mechanism at play that supports not just student’s marginal improvement but success and thriving that more robust measures like academic self-efficacy and sense of belonging demonstrate.

Cross-sectional Research

A further note, the scholarly research points to the difficulty in studying communities at single points in time or even in intervals as communities are dynamic and they change over time and these changes are important to understanding how they help facilitate student success. More recent scholarship has recommended ways to tap into student experience on an on-going basis versus intermittent focus groups, interviews or surveys through digital diaries, time management surveys or reflection journals (Andrade, 2007; Dodge, 2004; Engstrom, 2008; Hotchkiss et al., 2006). Too many studies only take “point in time” measures, making linkages between dynamic events and status measures difficult.

Our Study Addresses these Limitations

The study presents an opportunity to expand the outcomes examined related to learning communities. In this study we propose to examine self-efficacy and sense of belonging and other psycho-social outcomes which have all been associated with student persistence and success. We recognize that other outcomes may emerge in our qualitative findings. We believe that it is important to more systematically contribute to the literature by demonstrating whether learning communities do indeed shape these important outcomes (academic self-efficacy and sense of belonging) as touted in the literature but not supported by empirical studies (Hoffman, Richmond, Morrow, Salomone, 2002; Hurtado & Carter, 1997; Zhao & Kuh, 2004).

The outcomes we are interested in, particularly persistence outcomes, are dynamic. We intend to increase the frequency of measurement, making it more likely to draw direct connections between attitudes/behaviors and outcomes of interest. We propose to take into account the timing of the events of interest. The questions developed for this project address important areas that are currently not addressed in previous research –even the MDRC study –related to new more robust outcomes, more detailed and nuanced study of program elements, a longitudinal study of outcomes, and disaggregated by student subpopulations, which can make a significant contribution.

Given there is a dearth of information on how different types of learning community models impact or shape students’ experiences and outcomes, we believe it is important that this study examine this gap in our knowledge. We propose to examine the distinctive (living learning, commuter center) and common (mentoring, cohort courses) program elements. For example, we’ll look at the faculty/student interaction and what types of interactions is most important to shaping student experience and success in learning communities. In addition, we are interested in student engagement as an intended goal of the Thompson Scholars program and how student engagement mediates other psycho-social outcomes. Given the importance of student engagement, we will examine different levels of engagement among students to see if it results in different outcomes. Another area of inquiry noted in the past literature is focused on which parts of the learning community mediate the degree to which students feel connected to the community. Since current learning communities research stresses the importance of community and small cohorts as being critical to more effective education, being unable to define successful community building acts as a barrier to replicating successful learning communities (Smith, 2004). Therefore, research that describes the ways that different aspects lead to community building can help future efforts at building strong learning communities.

It is important to acknowledge that the program involves more than learning communities in their most traditional sense-clustering courses. The program includes a scholarship component, social support programs such as mentoring, residential component, and many other component parts that essentially aim for many of the same outcomes the learning communities are designed to achieve such as academic self-efficacy and sense of belonging.

Theoretical Framework

Relying heavily upon Tinto’s (1997) model of institutional departure and research conducted by Hurtado, Han, Saenz, Espinosa, Cabrera, and Cerna (2007), the following conceptual model (see Figure 1) has been adapted to the proposed study. The conceptual model provides an overarching visual representation of theoretical concepts operationalized with either individual survey items or scales. Conceptually, this framework stands upon modifications of Tinto’s departure model where academic and social integration into the campus environment are vital. In other words, critical aspects of the academic and social domains of college must be examined to better understand students’ “academic performance, academic/intellectual development, and non-cognitive gains (in psychosocial domains) as intermediate outcomes, which determine subsequent goals, institutional commitment, and persistence in college” (p. 847). Hurtado et al (2007) adds that students’ characterizations (i.e. student perceptions), social interactions (i.e. frequency and quality), formal memberships (i.e. learning communities), and perceived social cohesion offer other necessary sociological constructs critical for understanding the impact of college on these student-related outcomes. The conceptual relationship as we have proposed for the current project depicts a linear research model (i.e. Input-Environment-Output) and asserts that students’ demography, pre-college experiences, SES (socio-economic status), college readiness, institution location, and expected time use will have moderating effects on the stated outcome variables. We expect that college engagement (defined by peer interaction, student-faculty interaction, time use, campus service use, and engagement in learning community) mediate students’ academic development/ performance, sense of belonging, and academic self-efficacy. Direct correlations between student background characteristics and college engagement offer additional insight and when considered together (i.e. interaction effects) may provide additional explanation value.


Figure 1. Theoretical Framework (Theory of Change)

For a list of the references listed above please see Literature Review References.