Page 9

ACEF Journal Vol 3 Issue 1 December 2012

Morris Slate Table 2 Descriptive Statistics for the Difference in Students Scoring Above the Criterion on AP and IB Exams by Gender and School Size Variable n of schools M SD Boys Large Schools 365 20.49 9.97 Medium Schools 90 16.08 11.94 Small Schools 446 9.78 14.64 Girls Large Schools 365 25.69 11.23 Medium Schools 90 20.50 14.50 Small Schools 446 12.37 16.82 Instrumentation and Procedures Archival AEIS data, obtained from the TEA website for the 2008-2009 school year, were downloaded from the website as .dat files and contained student passing rates for AP and IB exams separated by gender and school size. The .dat files were merged using the Statistical Package for Social Sciences (SPSS). Because the data from the TEA website were reported by school district, errors are assumed to be minimal or few. Readers are referred to the technical manuals on the TEA website for information regarding score reliability and score validity. Included in the AEIS report are three values that are calculated for AP and IB exams: (a) percentage of students in grades 11 and 12 who took at least one AP or IB test; (b) percentage of students who scored at or above criterion score (i.e., a score of 3 on the AP test or a score of 4 on the IB test); and (c) percentage of scores at or above the criterion score (Texas Education Agency, 2011). Results The standardized skewness coefficients (i.e., the skewness value divided by the standard error of skewness) and the standardized kurtosis coefficients (i.e., the kurtosis value divided by the standard error of kurtosis) were calculated to determine normality. The majority (22 out of 24) of the standardized skewness and kurtosis coefficients were outside the range of normality (i.e., +/- 3, Onwuegbuzie & Daniel, 2002). When the statistical analyses were performed, the assumptions for the Box’s Test of Equality of Covariance and Levene’s Test of Equality of Error Variances wereviolated. However, the robustness of the Multivariate Analysis of Variance (MANOVA) made it the appropriate statistical procedure to be utilized even though the assumptions were violated (Field, 2009). Therefore, a MANOVA was performed to determine if the percentage of boys and the percentage of girls taking AP or IB exams and the percentage of boys and girls above the criterion on AP or IB exams differed as a function of school size. 9 Vol. 3, No. 1, 2012


ACEF Journal Vol 3 Issue 1 December 2012
To see the actual publication please follow the link above