RESUMO
Statistical methods have become an essential component of all empirical biomedical research. Science requires that these methods arefully reported with complete accuracy so that the evidence base could be fully appraised for validity, reliability, and generalizability. Tomeet this objective, Statistical Analyses and Methods in Published Literature (SAMPL) guidelines have been prepared for statisticalreporting in biomedical publications. This communication proposes substantial improvement of these guidelines to make them morecomprehensive, organized, compact, and easier to adopt.
RESUMO
The first trial of common achievement test-computer-based testing was held from January through August 2002. The number of examinees was 5, 693, of whom 5, 676 were analyzed. Single-best-answer, five-choice questions were used. The highest score was 92 points, the lowest score was 19 points, and the average score was 55.9±10.2 points (standard deviation). Scores were distributed normally. The test sets did not differ significantly in difficulty, although test-set items differed for each student. The percentage of correct answers, the ∅-coefficient, and the point-biserial correlation coefficient were calculated for each category of the model core curriculum. The percentage of correct answers was highest in category A of the model core curriculum, and percentages of correct answers were similar in categories B, C, D, E, and F. The ∅-coefficient and the correlation coefficient were low in categories A and F and were highest in category C. Although the percentage of correct answers in this trial was lower than expected, many test items had discriminatory power. The Test Items Evaluation Subcommittee is now evaluating test items, determining pool items, and revising new test items for the second trial and expect to compile a useful item bank.