Characteristics of online graduate students who utilize LRC services are described to understand the extent to which they may be at risk of not persisting. Further, these students are compared to students in the same course with the same faculty member at the same time to understand relationships between LRC service utilization and academic outcomes.
Student enrollment in online higher education continues to grow (Britto & Rush, 2013). However, higher failure and dropout rates have been associated with online higher education as compared to brick-and-mortar institutions (Britto & Rush, 2013). There are a number of differences between online and brick-and-mortar institutions that might explain why additional academic support services may be needed for students at the former compared to the latter. For example, faculty members at brick-and-mortar institutions have greater and more predictable availability (e.g., class time, scheduled office hours) than those at online institutions, whereas the availability of online faculty members to interact with students is impacted by numerous factors (e.g., time zone differences) (Felder-Strauss et al., 2015). In addition, brick-and-mortar institutions are more likely to have math and writing tutoring centers than online institutions; however, the inability of online students to seek help from faculty members in their specific program or discipline has led online institutions to develop additional specialized services for students (Felder-Strauss et al., 2015). Nevertheless, students must be willing to access these additional services, instead of acting as “lone wolves” (Brown, Hughes, Keppell, Hard, & Smith, 2015). Further, these services need to be perceived as helpful by students (Price, Richardson, & Jelfs, 2007).
Regardless of institution type, LRCs may offer support services that provide students with opportunities to engage with their course curriculum with different media, relearn concepts, and ask for further explanations (Fullmer, 2012). However, these centers appear to be experiencing an “identity crisis” (Truschel & Reedy, 2009). The literature is mixed in regards to what LRCs do and how they function to best support students (Truschel & Reedy, 2009). This is of particular concern in a time when such supports are needed most, as more students enroll in online programs, but continue to struggle to a greater extent than their peers at brick-and-mortar institutions (Britto & Rush, 2013).
This exploratory study was designed to serve as a foundation for the research on LRCs in the online graduate educational context. Specifically, the purpose of this research was to (1) describe the characteristics of online graduate students who utilize coaching services at an LRC to understand the extent to which they may be at risk of not persisting and (2) compare these students to students in the same course with the same faculty member at the same time to understand relationships between LRC service utilization and academic outcomes. The corresponding research questions were (1) what are the characteristics of online graduate students who utilize coaching services at an LRC? and (2) to what extent does use of LRC services relate to persistence?
To answer these research questions, a quantitative methodology and causal-comparative design were employed, as the interest was in explaining differences in persistence based on utilization of LRC services (dichotomous variable). All students who visited the LRC during the selected 3-month period had an equal chance of being randomly selected to be in the LRC sample. First, data were put in numerical order by student ID. Then, the manager of the LRC was instructed to select every third student to create the random LRC sample dataset. To develop the matched sample, a request was submitted to Data Operations with detailed instructions. Specifically, the LRC dataset was sent to Data Operations, and the team member was instructed to locate students who were in the same course with the same faculty at the same time, regardless of current status (active, inactive). If more than one student was located, the Data Operations team member was instructed to find the closest match in terms of school and program. If more than one student was located, the Data Operations team member was instructed to find the closest match in terms of age, sex, and racial/ethnic background. As with typical institutional research data requests, the Data Operations team member was not informed of the purpose of this study. The only goal was to locate a student who was as similar as possible to each student in the random LRC sample, except that the matched student did not visit the LRC. Following this procedure, the resulting dataset included information for 163 students who visited the LRC. Once the matched sample was created, and one student was paired with each student in the LRC sample, the total sample was 326 students (163 students who visited the LRC, 163 students who did not visit the LRC).
The requested data included demographic (e.g., age, sex, racial/ethnic background) and academic (e.g., school, program, time since enrollment, GPA, course grades) data both for students who used the LRC and students in the matched sample who did not use the LRC. In addition, data requested for the LRC sample only included reason for contacting the LRC, current course, and current faculty member. As outlined in the approved IRB application, no identifying student or faculty member data are reported.
The existing student data that were collected for institutional research purposes by the LRC as well as the university data operations team were then analyzed. Descriptive statistics for the larger online university population as well as students using LRC services and those in the matched sample who did not use the services are presented to promote understanding of which students tend to use academic coaching services. In addition, numerous preliminary analyses were conducted to examine the extent to which the LRC sample and the matched sample differed. For the main analysis, results of a McNemar’s test (given the matched pairs of participants and nominal dependent variable) are presented. For this analysis, there were three levels of the outcome variable (persistence): enrolled in next course and vested; enrolled in next course, but did not vest; did not enroll in next course. This is a commonly employed operational definition of persistence. These findings are presented in the context of existing literature. In addition, implications for future research and practice are discussed.