Teacher certification content area tests: Predictors of teacher knowledge for post-baccalaureate secondary candidates.

In response to a growing teacher shortage, increasing numbers of secondary teachers are prepared through streamlined certification programs. For this reason, assessing candidates' content area knowledge gained from institutions of higher education across the United States is an important program admission factor as candidates must demonstrate content area knowledge by passing a Texas content area certification test (TExES). This study examines content knowledge for candidates enrolled in an online post­-baccalaureate program from September 1, 2002 through April 30, 2005. Academic transcript analysis and grades 8-­12 subject tests of the TExES were used as a proxy for subject matter knowledge for a sample of individuals seeking initial teacher certification in a post­-baccalaureate teacher certification at the University of North Texas. Descriptive data,linear regression, and logistic regression analyses were used to draw conclusions about the content area knowledge of the individuals in the sample. Scores on the TExES were used to determine the relationships between the content area knowledge of initial certification students and the number of content area courses completed, the grade point averages, and time elapsed between the completion of the last content area course and the student's initial attempt on the TExES. Results differed by the content area of the candidates. Analysis of variance results indicate significant differences between the five test groups with regard to number of courses taken F(4,139) = 9.334, p < .001 grade point average F(4,139) = 5.733, p < .001 and time between the last course taken F(4,139) = 6.135, p < .001. The three­-predictor model was statistically significant F(3,32) = 3.753, p = .02 for the History test group. The variable, upper-­level grade point average accounted for approximately 12% variance among scores within the History test group, and the variable months of time elapsed between last content area course work and the initial state content examination accounted for approximately 13% of variance among scores