1. TEORÍA PARA LA INTEGRIDAD DE LA CONTABILIDAD
2.2 RIESGOS PERCIDOS
2.2.1 En la comunidad aledaña
The data reported in this chapter are organized by grade level with descriptive and inferential analysis for each dependent variable (NRTf CRTf ATT). Each question is answered with the appropriate statistical hypothesis. Cell, counts are shown in parentheses for descriptive data.
Analyses of variance (ANOVA) for pretest data and analyses of covariance (ANOCOVA) for posttest data were used to analyze NRTf CRTf and ATT test results. The ANOVAs and ANOCOVAs were completed by using the General Linear Models (GLM) procedure in SAS (1987). The GLM procedure was used because of unequal cell sizes in the data. Type
III sums of squares were used in all analyses (SASf 1987). All ANOVA and ANOCOVA tables involved were presented and after each table the hypotheses were tested and discussed.
Grade 7
A total of 324 seventh graders were pretested in the Fall of 1987. Because some students moved out of the districts during this studyf only 289 complete data sets
were used in the analysis of the seventh grade data. This is less than the 320 students desired, but within the range of the sample size requirements* Descriptive summary
statistics along with adjusted means from the Analysis of Covariance (ANOCOVA) are presented in Appendix B, Table 27. Analysis of Grade 7 NRT Data
An analysis of variance, as shown in Table 3, was completed to test the assumption of equality of NRT means in the pretest treatment groups.
Table 3. ANOVA of Grade 7 Pretest NRT Means
Source DF Sum of Squares Mean Squares F value P GENDER I 26.76 26.76 .52 .4702 ML I 777.22 777.22 15.18 .0001 WT I 64.25 64.25 1.26 .2635 GENDER * ML I 25.71 25.71 .50 .4791 GENDER * WT I 363.30 363.30 7.10 .0082 ML * WT I 239.21 239.21 4.67 .0315 GENDER * ML * WT I 2.36 2.36 .05 .8303 ERROR 281 7519.54 26.86
There was a significant difference among the
treatments. The analysis of the interactions (refer to Table 4) of mastery learning and wait time was significant
(p = .0315) and disordinal with mastery learning with wait time having the largest NRT mean of 26.67.
Table 4. Interaction of ML and WT NRT Pretest Means
on Grade 7
No Mastery Learning Mastery Learning
No Wait time 22.18 23.71
Wait Time 21.40 26.67
The interaction between GENDER and WT was significant (p = .0082) and is summarized in Table 5. The interaction was disordinal with males with wait time having the largest mean of 25.24 and females had their larger mean in the no wait time group.
Table 5. Interaction of GENDER and WT on Grade 7 NRT Pretest Means
Males Females
No Wait Time 22.12 23.81
Wait Time 25.24 22.31
There was a significant difference between the main effects of mastery learning versus no mastery learning (p =
.0001) . The mean for the mastery learning group was 25.05 and for the no mastery learning group, the mean was 21.75.
These statistically significant results on Grade.7 NRT pretest means indicated that differences among the
experimental groups existed at the start of treatment in
this study. The statistics concerning main effects and interactions for Grade 7 NRT posttest data are summarized in Table 6. The results are presented for each hypothesis. Table 6. Grade 7 Analysis of Posttest NRT Means Using
ANOCOVA with NRT Pretest as Covariate
Source DF Sum of Squares Mean Squares F value P GENDER I 141.60 141.60 5.27 .0224 ML I 10.65 10.65 - .40 .5294 WT I 26.64 26.64 .99 .3201 GENDER * ML I 49.67 49.67 1.85 .1749 GENDER * WT I 3.67 3.67 .14 .7120 ML * WT ’ I 16.60 16.60 .62 .4323 GENDER * ML * WT I 11.98 11.98 .45 .5048 ERROR 280 7519.54 26.86
Ho: There is no interaction among the independent variables: mastery learning versus no mastery learning, wait time versus no wait time and gender on the dependent variable, adjusted NRT test scores.
Decision: Retain the null hypothesis at the alpha = .10 level. There was no three way interaction among mastery learning, gender, and wait time on adjusted NRT posttest means (p = .5048).
Ho: There is no interaction between the independent variables: mastery learning versus no mastery learning, and wait time versus no wait time on the dependent
Decision: Retain the null hypothesis at the alpha = „10 level. There was no two way interaction between mastery learning and wait time on adjusted NRT posttest means (p = .4323).
Ho: There is no interaction between the independent variables: wait time versus no wait time and gender on the dependent variable, adjusted NRT test scores.
Decision: Retain the null hypothesis at the alpha = .10 level. There was no two way interaction between gender and wait time on adjusted NRT posttest means (p = .7120).
Ho: There is no interaction between the independent variables: mastery learning versus no mastery learning, and gender oh the dependent variable, adjusted NRT test scores.
Decision: Retain the null hypothesis at the alpha = .10 level. There was no two way interaction between gender and mastery learning on adjusted NRT posttest means (p =
.1749).
Ho: There is no statistically significant difference between the means of the adjusted NRT test scores for wait time versus no wait time.
Decision: Retain the null hypothesis at the alpha = .10 level. There was no significant difference in adjusted
\
Ho: There is no statistically significant difference between the means of the adjusted NRT test scores for mastery learning versus no mastery learning.
Decision: Retain the null hypothesis at the alpha = .10 level. There was no significant difference in adjusted NRT posttest mastery learning means (p = .5294).
Ho: There is no statistically significant difference between the means of the adjusted NRT test scores for gender.
Decision: Reject the null hypothesis at the alpha = .10 level. There was a significant difference in adjusted NRT posttest gender means Cp = .0224) with females having an adjusted NRT posttest mean of 25.75 and males having an adjusted NRT posttest mean of 24.34. In looking at actual meansr the females began with an actual mean of 23.14
versus 23.71 for the males. At posttest the means were