Evaluating the Teacher-Intern-Professor Model in a Professional Development School Partnership Setting using a Bayesian Approach to Mix Methods
Two needs of Georgia State University Professional Development School Partnerships are to show increases in both student academic achievement and teacher efficacy. The Teacher-Intern-Professor (TIP) Model was designed to address these needs. The TIP model focuses on using the university and school p...
| Main Author: | |
|---|---|
| Format: | Others |
| Published: |
Digital Archive @ GSU
2009
|
| Subjects: | |
| Online Access: | http://digitalarchive.gsu.edu/eps_diss/46 http://digitalarchive.gsu.edu/cgi/viewcontent.cgi?article=1045&context=eps_diss |
| Summary: | Two needs of Georgia State University Professional Development School Partnerships are to show increases in both student academic achievement and teacher efficacy. The Teacher-Intern-Professor (TIP) Model was designed to address these needs. The TIP model focuses on using the university and school partnership to support Georgia State University student intern preparedness and student academic achievement for those participating in the program. TIP Model outcomes were analyzed using a quasi-experimental design for achievement data and a Bayesian approach to mix methods for efficacy data. Quantitative data, in the form of test scores, were analyzed to compare mean student academic achievement at the classroom level. Mean differences between treatment and comparison groups were not significant for the TIP treatment factor (F(1, 60) = .248, p =.620) as measured by a benchmark test. Results favored the treatment group over control group for the TIP treatment factor (F(1, 56) = 17.967, p < .001) on a geometry test. A methodological contribution is the exploration and development of an approach to mix methods using Bayesian statistics to combine quantitative and qualitative data. Bayesian statistics allows for incorporation of the researcher’s prior belief into the data analysis. Narrative Inquiry was the qualitative framework employed to gain understanding of the participants’ qualitative data, thus providing a particular way of prior belief elicitation. More specifically, a content analysis of the qualitative data, which included interviews, observations, and artifacts, was used in conjunction with quantitative historical data to elicit prior beliefs. The Bayesian approach to mix methods combined prior beliefs from the teacher efficacy qualitative data with the quantitative data from Gibson’s and Dembo’s Teacher Efficacy Scale to obtain posterior distributions, which summarized beliefs for the themes of teacher efficacy and personal efficacy. |
|---|