in participation from 1-16 total weeks of measurement, with a mean of 6.9 occasions, a median of 7 occasions, and a mode of 8 occasions overall. The RISE participants had a mean of 6.6 occasions, a median of 7 occasions, and a mode of 6 occasions. The Comparison participants had a mean of 6.9 occasions, a median of 7 occasions, and a mode of 8 occasions (please see Figure 1).
Figure 1. Number of Weeks Participated (All Participants)
Total Weeks Participated
The result of this analysis was to set the participation cutof in the longitudinal data set to 6+ occasions of measurement. 258 of the 378 total participants (i.e., 68%) were therefore included in the longitudinal analysis (45 RISE participants, and 213 Comparison group participants). Participants ranged from Grades 1-5 (see Appendix 1). The diference in attrition between RISE and the comparison group was quite modest (84% to 83% (i.e., -1%) in the comparison group, and 16-17% (i.e., +1%) in the RISE group). Therefore, our approach met the requirements for minimizing diferential attrition in the RISE and Comparison samples. Missingness by Occasion of Measurement There was signifcant missingness in six of the 18 total weeks; Week 1 (73% missing), Week 10 (100% missing), and Weeks 15-18 (74% missing). Week 1 missingness was due to communication issues at launch, Week 10 was spring break, and Weeks 15-18 were interrupted by state testing. Therefore, Weeks 1, 10, and 15-18 were excluded from analysis. In addition, the participants were asked to respond on a biweekly basis, so missingness at the individual level typically followed a biweekly pattern. Therefore, scores were averaged in two-week intervals (e.g., Weeks 2 and 3 were averaged, and Weeks 4 and 5 were averaged). This averaging resulted in a reduction to six occasions of measurement. After
LXD Research -RISE and RISE UP Winter 2022-Spring 2023 Report
23
Made with FlippingBook - Online catalogs