The economic and employment outlook for individuals without a high school diploma is bleak. For many of these individuals, passing the General Educational Development (GED) Test is the first step in competing in the increasingly demanding job market. GED test-taking policies vary across test centers and jurisdictions, and have the potential to affect several outcomes related to the GED credential, such as passing the test or preparedness for postsecondary education. However, little is known about this relationship. This study examines the relationship between GED policies and performance on the content areas and the GED Test as a whole. The model that best fits the data, including test center- and jurisdiction-level predictors, explained approximately 15 percent of the variability in overall GED standard scores, which varies by content area, ranging from 10 percent for Language Arts, Reading to 17 percent for Science, suggesting that the importance of different variables differs as a function of the content area. The results suggest that candidates of various backgrounds are at a disadvantage even after controlling for relevant candidate factors. The most consistent finding associated with test centers was that the gap in performance between African-American and white candidates was smaller in test centers that required the Official GED Practice Test (OPT). However, this requirement did not decrease the performance gap between Hispanic and white candidates. In summary, although there is variability in overall GED standard scores and content area scores associated with the test center and jurisdiction levels, only one of the predictors at these levels--centers open all months of the year--helped account for this variation. Additional implications are discussed. (Contains 2 figures and 16 tables.)