Effects of coaching community college Hispanic students on English language skills assessed in a high-stakes academic skills exam

Date of Award




Degree Name

Doctor of Philosophy (Ph.D.)


Higher Education

First Committee Member

John H. Croghan - Committee Chair


Attention has been given to coaching as a method of improving test scores for minorities in high-stakes tests. Many Hispanic students are not passing the College Level Academic Skills Test (CLAST), an exam certifying competence for admission to the upper division level in Florida's state university system, because they have scored low on its English Language Skills (ELS) sub-test.This study focused on the relationship of verbal skills preparation and performance of Hispanic students in a high-stakes exam. The effects coaching had on ELS performance scores of a small sample of Caribbean, Central and South American community college Hispanic students from South Florida were investigated.A quasi-experimental, randomized subject, pretest-posttest comparison group design was implemented. One group was offered a ten contact hours long coaching course over a four weeks period addressing grammatical rules and applications included in the three broad language skill areas (Word Choice; Sentence Structure; Grammar, Spelling, Capitalization and Punctuation) assessed by the ELS subtest.The study showed coaching had no effects on the ELS subtest performance of Hispanic students. Analysis of covariance did not reveal a significant difference between the groups in overall test performance. Their difference in performance on the subtest sections assessing each of the broad skill areas was also insignificant. The influence of grades in first year composition courses, the use of English or Spanish as the subjects' preferred language, and the subjects' placement level was not significantly related to the effect of coaching. Analysis of variance indicated there were no main or interaction effects among any of the above variables in their relationship to the overall test score nor in their relationship to the broad skill area scores of the two groups.Possible reasons for non-significance were discussed and recommendations for further research were presented.


Education, Community College; Education, Bilingual and Multicultural; Education, Tests and Measurements; Sociology, Ethnic and Racial Studies

Link to Full Text


Link to Full Text