Talent Flow From Undergraduate to Graduate School: 1982-1993
- Grandy, Jerilee
- Publication Year:
- Report Number:
- Document Type:
- Subject/Key Words:
- Databases graduate study student characteristics talent identification
This report describes the early development, recent revision, and some uses of a special Gradate Record Examinations (GRE) database for studying talent flow between undergraduate and graduate school. More specifically, it discusses the following: (1) development and expansion of a talent flow database containing all variables from the GRE General Test files on all examinees between 1982 and 1993; (2) availability of the database to other researchers; (3) analyses of the database to highlight talent flow patterns; (4) designs of graphic displays that are useful to illustrate talent flow; and (5) suggestions for future uses of the database, including a new approach to studying test validity. Variables in the database include all questions asked on the GRE background questionnaire in addition to GRE scores. The database was constructed so that its format and coding are as comparable as possible across all 12 years, in spite of radical changes made in the background questionnaire during that period. Several formats of the database are available to researchers upon request. The report includes summary statistics for all broad fields of study˛for each gender and all ethnic groups˛including trends in the numbers and percentages of examinees planning graduate study in each area and the academic qualities of those examinees. Talent flow analyses examined the holding power of each broad field of study. Those fields with the greatest holding power appeared to coincide with the fields having the fewest number of examinees seeking a doctorate. For examinees changing fields, we were also able to identify fairly consistent patterns associating test scores with chosen field of graduate study. For example, education majors with relatively high verbal scores switch to humanities, and education majors with relatively high quantitative scores switch to physical sciences. Such patterns also suggest an entirely new way of looking at test validity. Analyses are quite detailed, focusing on each broad field of study and some specific fields, and on each gender and ethnic group. Some graphic methods are developed to illustrate the complexities of talent flow. Finally, the report suggests other ways in which the database may be used by researchers.