A Comparison of the Performance of Graduate and Undergraduate School Applicants on the Test of Written English™
- Zwick, Rebecca J.; Thayer, Dorothy T.
- Publication Year:
- Report Number:
- RR-95-15, TOEFL-RR-50
- Document Type:
- Subject/Key Words:
- Comparative analysis graduate students scores undergraduate students
The performance of graduate and undergraduate school applicants on the TWE® (Test of Written English™) was compared for each of 66 data sets, dating from 1988 to 1993. The analyses compared the average TWE score for graduates and undergraduates after matching examinees on the TOEFL total score. The main finding was that, for matched examinees, undergraduate TWE means were higher than graduate means in 63 of the 66 data sets. Although these standardized mean differences (SMDs) never exceeded 0.3 of a TWE score point (with standard errors that were typically between 0.01 and 0.02), the results are noteworthy because they give a different picture than do simple comparisons of means for unmatched graduates and undergraduates, which show higher mean TWE scores for graduate applicants in the majority of cases. Of the nine SMDs exceeding 0.2, eight were in Region 1 (Asia and the Pacific) between October 1988 and May 1992. In evaluating these findings, the effects of the examinees' intended field of graduate study were investigated. For groups matched on the TOEFL score, applicants to graduate programs in the physical and biological sciences tended to have lower TWE means than undergraduates and graduates in the social sciences. Graduate students in the physical sciences tended to have lower TWE means than matched graduates in the humanities. Graduate-undergraduate differences in TWE performance may result in part from a greater concentration of graduate applicants in scientific fields. Another hypothesis is that undergraduates are more likely to have recently participated in intense English writing instruction. Region 1 may show larger graduate-undergraduate differences because of greater demographic disparities in that region between these two groups of examinees.