An Investigation Of A Brief Measure Of School Membership

Adolescence, Summer, 1998 by Winston J. Hagborg

Intellectual Achievement Responsibility (IAR). Developed by Crandall, Katkovsky, and Crandall (1965), the IAR is a 34-item forced-choice measure of locus of control. Two alternative attributions, one internal and one external, are presented for each item. An I score (internality of success) represents the number of positive items for which the individual takes responsibility; the I- score (internality of failure) represents the number of negative items for which the individual takes responsibility. The Total IAR score (the sum of 1 and I-) indicates internality of causal attributions. In terms of internal consistency, Crandall et al. reported correlations of .60 for both the IAR and TAR-; test-retest correlations (2-month interval) were .65 for Total IAR, .47 for IAR , and .69 for TAR-.

RESULTS

Table 1 displays the mean PSSM-Brief scores and alphas by grade level and gender. The internal consistency (alphas) of the PSSM-Brief ranged from .79(7th grade) to .94(5th grade). A 2 (gender) X 4 (grade) ANOVA revealed no significant gender, grade, or interaction effects.

The mean PSSM-Brief score was 3.57 (SD = .81), with a median score of 3.55. Using a median split, the sample was divided into a low group (n = 57) and a high group (n = 63). These two groups were compared on several dependent variables using a series of independent sample t tests (see Table 2). Scores for the high group significantly exceeded those for the low group on grades, homework time, LASSIHS motivation, IAR , and JAR Total. To further clarify the strength of the association between the PSSM-Brief and each of the dependent variables, correlations were computed using the entire sample (see Table 2). The groups did not significantly differ on educational aspirations, yet a statistically significant positive correlation was found for the total sample (r = .20, p [less than] .05).

DISCUSSION

This investigation examined the psychometric properties of a shortened version of the PSSM. Guided by previous factor analysis (Hagborg, 1994), 11 of 18 items were selected to create a uniform measure of school belonging (2 positively worded items and all 5 negatively worded items were excluded). In the present study the PSSM-Brief, with 7 fewer items, demonstrated a high degree of reliability (alpha = .90). The mean score for female students was slightly higher than that found in prior PSSM research (Hagborg, 1994), increasing from 3.47 to 3.52, while the mean score for males rose to a far greater extent, from 3.31 to 3.62.

Through the use of high/low contrast groups and correlational analysis, the PSSM-Brief was found to demonstrate criterion validity. The high group had higher grades, spent more time on homework, and indicated greater motivation, which is consistent with previous findings for the PSSM (Goodenow, 1993; Hagborg, 1994).

Based on Wehlage's theoretical model of school membership, it was predicted that students with a greater sense of school membership would also have higher educational aspirations (Wehlage et al., 1989). While contrast-group differences were not found, there was a small but significant positive correlation between the PSSM-Brief and educational aspirations for the entire sample. One difficulty in assessing educational aspirations was that almost all students (81%) reported a desire to complete four years of college or more. In fact, only one student did not want to complete high school. Use of a high-risk sample or older, high school students would probably have offered a more appropriate test of Wehlage's hypothesis.