排序方式： 共有5条查询结果，搜索用时 8 毫秒
1
1.
2.
3.
Academic careers of the 123 students admitted to a doctoral program in psychology between 1963–1967 were reviewed. The attrition rate was 35%, while 29% completed a Ph.D. within four years. While there was no evidence that women were discriminated against in acceptance, attrition was significantly higher among them than among men. Fifteen traditional predictors including GRE scores, undergraduate GPA, letters of recommendation, and prior graduate work were correlated with 13 criteria such as first and later years' graduate GPA, first-year faculty evaluation, withdrawal from program, and completion of Ph.D. Correlations of predictors with the four most important criteria of graduate success revealed that undergraduate GPA, letters of recommendation, major, and quality of undergraduate department were correlated positively with first-year performance but not with eventual success. The best predictors of the ultimate criterion, the Ph.D., were having a master's degree at entrance, age and marital status (the latter two highly correlated with the first), and first-year faculty evaluation. Inasmuch as the last was the best single predictor of the long-term criteria of years of successful study and obtaining the Ph.D., it is suggested that more attention be given by departments to the systematic development and use of early faculty evaluations of the performance of selected students rather than continuing to be concerned with perfecting selection.Carol Lillie (Rater 1) and Virginia de Wolf (Rater 2) assisted in data collection and development of letter rating schemes referred to. These may be requested from the Bureau of Testing, University of Washington, Seattle 98195. 相似文献
4.
The role of testing in determining college admissions and the impact of that testing on high school students are commanding increasing attention. There is emerging evidence that admissions decisions made solely on the basis of high school record can serve the college as well or better than decisions based, at least in part, on scores obtained on admissions tests. Further, though there is a recognized need for guidance, the tests in current use for admission provide limited information for helping students in making course selections that would make the best use of their abilities. This article reviews an alternative, long neglected and little known, for testing at the high school-college interface that would meet this need. Testing to provide differential prediction of college performance, as proposed by Paul Horst, focuses on allowing students to make the best match between their achievements, interests, and goals with the undergraduate programs of study available to them. 相似文献
5.
1