Luisa Guillemard’s research while affiliated with Texas A&M University and other places

The purpose of this investigation was to develop and pilot-test maze-like semantic maps for assessing reading comprehension of content-area information. Semantic map tests reflect students' understanding of relationships among key passage concepts, as an integrated whole. The production, administration, and scoring of semantic map-tests was standardized, using undergraduate college students in education. Next, inter-scorer reliability, scaling and interpretation of four types of map-scores were investigated. The scores were also validated against three criteria: standardized test scores, science report card grades, and teacher ratings. These psychometric properties were investigated through pilot-testing with a heterogeneous group of 144 Grade 7 and 8 students in a low-achieving school. Results showed that good quality maps could be produced and scored with high reliability among preservice teachers. Criterion related validity based on a standardized test was weak, but moderate-to-strong validity was obtained against criteria of teacher ratings and science grades. Interestingly, teacher ratings and science grades supported different types of map-test scores. Semantic map tests merit further research, with emphasis in three areas: (a) retest and alternate form reliabilities, (b) relative sensitivity to background knowledge versus text-based knowledge, (c) usefulness in an instructional context.

A multistate survey of 859 school psychologists who indicated prior experience conducting bilingual psycho-educational assessment found that over half had used interpreters. Seventy-seven percent of the school psychologists who reported using interpreters had received no or very little training to do so. Only 37% of the school psychologists reported that their interpreter had received formal training. In only 7% of the cases reported were both school psychologists and interpreters trained in the interpretation process. The data illustrate the need to increase training and recruitment of minority and/or non-minority graduate students proficient in two or more languages for the purpose of more accurately assessing limited-English proficient students. Additional areas for further research involving interpreters are suggested.