Volume 1 - Issue 1 - DBU Journal for K-12 Educational Leadership - Page 44

42 non-PBL group on the basis of the outcome measures of mathematics, reading, and social studies in terms of the re- porting categories for each of the assessments. The factors of comparison, or dependent variables, were the STAAR re - porting categories for each subject. The independent vari- able was whether PBL was the platform for instruction. Social Studies MANOVA Results In Table 1, the results for the multivariate tests of signifi - cance indicate whether “there are statistically significant differences among the groups on a linear combination of the dependent variables” (Pallant, 2013, p. 304). Using the Wilk’s Lambda Test and the criteria that alpha must be less than .05, if the p-value is less than .05, there is a difference among the groups (Pallant, 2013). There was a statistically significant difference between PBL and nonPBL groups on the combined dependent variables, F (4, 358) = 4.42, p = .002; Wilks’Lamda = .95; η2 = .05). When the results for the de - pendent variables were considered separately using a Bon - ferri adjusted alpha level of .0125, all four objectives reached statistical significance: Social Studies Objective 1, F (1, 364) = 10.42, p = .002, η2 = .027; Social Studies Objective 2, F (1, 364) = 13.19, p = .000, η2 = .035; Social Studies Objective 3, F (1, 364) = 11.85, p = .001, η2 = .032; and Social Studies Objective 4, F (1, 364) = 8.83, p = .003, η2 = .024. The effect size was represented by eta squared (η2) for the MANOVA. Miles and Shelvin (2001) re - ported the thresholds for small effect to be .01, for mediumef - fect tobe .06, and largeeffect tobe .14. All of theη2 represented small effects for the four social studies objectives as indicated in Table 1. (See Table 1 and table 2 on page 45.) Reading MANOVA Results The multivariate tests of significance indicated no statistical - ly significant difference in the PBL and nonPBL groups in terms of academic achievement on any of the three reading objectives. Since the multivariate tests of significance did not indicate a statistically significant difference on any of the three reading objectives between the PBL and nonPBL groups for the null hypothesis, no further investigation of three reporting categories for reading occurred. Mathematics MANOVA Results The multivariate tests of significance indicates no statistical - ly significant difference in the PBL and nonPBL groups in terms of academic achievement on any of the five mathe - matics objectives. Since the multivariate tests of significance did not indicate a statistically significant difference for any of the five mathematics objectives between the PBL and non - PBL groups, no further investigation of the five reporting categories for mathematics occurred. Research Question 2 The one-tailed t-test for independent samples was per - formed to compare the attendance of the characteristic-pres - ent and comparison groups at a significance level of .05. The t-test was used to compare two groups in terms of outcomes (Creswell, 2014). The one-tailed test was used to determine whether one mean was higher than the other. The one-tailed t-test for independent samples of the hypothesis for Re- search Question 2 addressed the impact of PBL on student attendance. The t-test produced t (364) = .528 with p = .598. There was no significant difference on student attendance as measured by 2013-2014 PEIMS data of students taught with PBL (M = 6.24, SD = 5.385) and students taught without PBL (M =6.70, SD = 7.364). Research Question 3 The one-tailed t test for independent samples was performed to compare the discipline referrals of the characteristic-pres- ent and comparison groups at a significance level of .05. An independent t-test was used to test the hypothesis for Re - search Question 3 addressing the impact of PBL on student discipline referrals. The t-test produced t (103.297) = -1.223 with p = .224. There was no significant difference on student discipline referrals as measured by 2013-2014 PEIMS data between students taught with PBL (M = 1.05, SD = 2.399) and students taught without PBL (M = .71, SD = 1.509). Findings This study compared the standardized test results for eighth grade STAAR social studies, reading, and mathematics be- Cynthia Anne Mika, Ed.D.