This study investigates teacher content knowledge of candidates enrolled in an online graduate teacher certification programme. Descriptive data and linear regression were used to draw conclusions about the content area knowledge of individuals in the sample and the significance of the predictors examined. Descriptive data show 1/3 of the 8-12 life science candidates and 2/5 of the 8-12 mathematics candidates fail the state content examination although a transcript analysis indicates candidates have strong content preparation, high grade point averages and recent coursework. Results of a linear regression analysis indicate, upper-level content area coursework, grade point average and the time elapsed between the initial attempt on the Texas Examinations of Educator Standards content test and the completion of the last upper-level course in the content area were statistically significant predictors of success on the Texas Examinations for Educator Standards (TExES) for 8-12 life science and 8-12 mathematics teacher candidates; F (3, 81) = 3.076, p = 0.032. However, only grade point average was statistically significant accounting for 10% of the variance in Texas teacher content examination scores. Policy-makers may want to reexamine how high stakes teacher testing relates to content knowledge associated with degrees obtained from accredited colleges and universities and how high stakes teacher testing may exacerbate the teacher shortage problem in high need fields such as science and mathematics. (Contains 5 tables and 8 figures.)