The Journal of College and University Student Housing - Volume 36, No. 2 - 73

Exploring the Effect of a Residential Partnership

(76) were equally qualified engineering and computer science students who did not apply to or participate in the LLC. To ensure that students in both groups were as similar as possible, caliper matching (Anderson et al., 1980) was utilized to match each of the ECS-LLC students with a non-ECSLLC student. Students were matched on variables that included race, gender, classification, major, and the academic index score assigned to each student in the university admissions process. Academic index is calculated by converting the SAT score to a 100-point scale and adding the class rank percentile, for a maximum score of 200. At the conclusion of the matching process, a match was achieved between ECS-LLC and non-ECS-LLC students in the categories of race, gender, and major. Since finding an identical match in an academic index score was unlikely, differences within .15 of a standard deviation were considered acceptable. To verify that the matching process had been successful, the sample of non-LLC student academic indexes was compared to the population of non-LLC student academic indexes. No differences were found (t = 1.639, p-value = 0.1054). This suggested that students selected from the non-LLC pool were a representative sample based on these variables. Next, a paired t-test was performed and revealed no differences in the averages between LLC and non-LLC academic index scores (t = 0.22, p-value = 0.8297). Of the 152 students in the population used for comparison, 63.8% returned usable data from the survey, resulting in a sample of 97 students for this study. Table 1 displays the gender breakdown of the sample by community.

Table 1
Characteristics of Survey Participants
LLC Frequency Row Pct Col Pct GENDER

Female

Male

Total

LLC

10 17.54 58.82 7 17.50 41.18 17

47 82.46 58.75 33 82.50 41.25 80

non-LLC

Total

* Correlation is significant at the .05 level (2-tail).

INSTRUMENTATION
Ordinal data were collected using the LLC Experiences Questionnaire, a self-designed 34-item survey adapted from Pace and Kuh (1998) and created to measure the type and frequency of student interaction with faculty members (6 questions); the type and frequency of participation in specified activities (15 questions); the level of student satisfaction with specified components of the learning environment (7 questions); and, finally, student estimates of gains in specified areas (6 questions). The ordinal responses represented opinions on a scale of 1-2-3-4. The LLC Experiences Questionnaire was reviewed and modified numerous times by a steering committee composed of staff, faculty, and students on the researchers’ campus. Based on the focus group administration of the instrument, test-retest correlation coefficients range between .72 and .89 while internal conV o l u m e 3 6 , No. 2 • 2 0 1 0

The Journal of College and University Student Housing - Volume 36, No. 2

Table of Contents for the Digital Edition of The Journal of College and University Student Housing - Volume 36, No. 2