This study assessed the types of research and the statistical methods used in three representative health education journals from 1994 through 2003. Editorials, commentaries, program/practice notes, and perspectives represent 17.6% of the journals' content. The most common types of articles are cross-sectional studies (27.5%), reviews (23.2%), and analytic studies (i.e., case-control, cohort, and experimental studies) (18.4%). The estimated annual percentage change across the study period in these types of articles was 3.3, -9.3, and 5.5, respectively. A significant increase was observed in use of descriptive statistics (estimated annual percentage change = 2.4), parametric test statistics (4.4), nonparametric test statistics (3.5), epidemiologic statistics (10.3), generalized linear models (6.8), validation statistics (6.7), and other statistics (8.2). Movement toward increasing use of cross-sectional studies, analytic study designs, and statistical methods--representing greater emphasis on needs assessment for health education, health education program development, and program evaluation--indicates the need for better quantitatively trained health educators. (Contains 2 figures and 6 tables.)