Student Scores on Advanced Placement Exams: Gender Variables

The results of the Advanced Placement exams given to students in 2006 were disaggregated according to gender. The level of performance was compared between males and females using Cohen's d. The standardized differences between male and female performance group levels were compared to previous results for the 1992 Advanced Placement (AP) exams. One purpose of this study was to determine whether or not the male-favored results that existed in 1992 still existed in the 2006. This study found that differences still exist in results based on gender and no real progress has been made in reducing the gap in achievement between males and females. A second purpose of this study was to compare and contrast the local data to the national data set to see if scores in Brazosport were similar to scores collected at the national level. To determine if similar results would be obtained at the local level the results of 267 Advanced Placement exams taken by 190 students from Brazsosport Independent School District in 2006 were disaggregated according to gender. The level of performance was compared between males and females using Hedge's g. Because of limited sample size, only 9 Advanced Placement exams were reviewed at the local level. This study found that the results from Brazosport were, in many cases, quite different from those found on the national level and there was no pattern to explain the variation among the differences. This study supports the collection of local data for monitoring gender bias that might exist on Advanced Placement exams. The data collected in the current study indicates that individual district progress in overcoming gender differences that historically have existed in specific scores on the AP exam might be overlooked if only national data is reviewed