Using responses from a survey of elementary teachers from the state of Ohio, this study analyzed the influences of teachers' professional training (both pre-service training and in-service professional development), teachers' attitude towards ELLs, teachers' bilingualism, schools' percent of ELLs, and schools' resources (both socioeconomic status and per pupil expenditure) on the instructional strategies that teachers reportedly use with English language learners in content area classes. The researcher mailed packets of questionnaires to a random sample of schools in Ohio with ELL enrollments of 8.9% or higher and asked the principals to distribute questionnaire booklets to each content-area teacher in their schools and to collect and return the completed booklets.
Altogether the principals distributed questionnaires to 960 teachers, and of those teachers, 419 (i.e., 44%) returned the instruments. Data analysis first involved calculation of a variety of descriptive statistics useful for examining the values obtained in response to each item on the questionnaire. Next, bi-variate correlation coefficients provided information about relationships between variables. Finally, multiple regression analyses assisted in answering research questions about the separate and combined associations between the independent variables (as listed above) and teachers' reported use of research-based strategies for teaching ELLs. An ancillary analysis, also using multiple regression analysis, examined the association between several independent variables and teachers' attitude toward ELLs. Because the data were nested, intra-class correlations were calculated to determine whether or not follow up studies using multi-level models would be beneficial.
The study's major findings were as follows:
· Teachers' attitude appeared to be a relatively strong predictor of teachers' reported use of research-based strategies with ELLs.
· The percentage of ELLs enrolled in a school appeared to be a predictor of teachers' use of a set of research-based strategies with these students. The higher the percentage of ELLs in the school, the lower the reported use of these strategies.
· The ability to speak more than one language appeared to be a significant predictor of teachers' attitudes toward ELLs.
· Some evidence pointed to the possibility that amount of professional development is a predictor of teachers' reported use of research-based strategies for teaching ELLs.
· District per pupil expenditure appeared to be a significant predictor of teachers' attitude toward ELLs.
· Undergraduate preparation, school socioeconomic status (i.e. percentage of students on free and reduced lunch), district per pupil expenditure, teachers' gender, teachers' years of experience, school size, and district size had no apparent association with teachers' reported use of a set of research-based strategies.
· Teachers' level of experience appeared to be a significant predictor of teachers' attitude toward ELLs. The longer a respondent had been a teacher, the more negative was his or her attitude toward ELLs.
Review of these findings in light of earlier related research revealed that the study contributed new insights as well as confirming some findings from previous studies. Disclosure of the study's limitations provided justification for additional analyses using data from the survey and replication studies using the same research design with larger, more representative samples of teachers. These limitations also supported the need for interpreting findings from the study with caution. Nevertheless, the findings seemed sufficiently compelling to support policy recommendations relating to (1) the use of professional development to influence teachers' attitudes toward ELLs and the instructional strategies they use with these learners and (2) the requirement that pre-service (and possibly also in-service) teachers acquire some degree of fluency in a second language.