Volume 18 Fall 2016 Table of Contents

Independent paired samples t-tests were run using SPSS. The SPSS output tables are found in Table 7. On average, teacher candidates received higher RTOP scores from the university supervisors (M=70.11, SE= 5.08), than from the researcher (M=44.26, SE=4.26). This difference was statistically significant t(18)=5.79, p<.05; it represents a large sized effect r = .65.

The university supervisors consistently used the RTOP as a reference for high quality mathematics instruction. They repeatedly stated in interviews and follow-up conversations that they expected more rigor and higher level thinking as a result of the focus on the RTOP rather than the open-ended general observation form used in the past. This intensified level of expectations in the mathematics instruction included the requirement of K-5 students justifying and sharing their strategies. Interviews highlighted an increase in the expectancy of real world and hands-on learning. They wanted to see the teacher candidates actively involving students. Even though two university supervisors did not assess the teacher candidates themselves, but instead used the teacher candidates’ self-assessment on the RTOP, the use of the RTOP did increase the emphasis on the mathematics content knowledge and the pedagogical content knowledge.

The university supervisors approach to the post-conference changed between the baseline semester and the study. They described that they listened more carefully to the teacher candidates’ reflections and put an emphasis on teacher candidates’ reflections. They allowed the teacher candidates to problem solve and come up with their own strategies and ideas for improving their instructional practice versus telling them their strengths and weaknesses. Comparing the interviews from the baseline to the study, teacher candidates did experience an increase in mathematics support from the university supervisors.

Observations revealed that in 15 out of 19 post observation conferences, paraphrasing was used. Closed and probing questions were the most common types of questions asked versus the more reflective mediating and open-ended questions. The data revealed that the coaching practices of the university supervisors are in the novice stage. In reviewing the post-conferences, the limited use of paraphrasing with the teacher candidates displays a need for more professional development, modeling and practice. In order to facilitate change in beliefs and practices, university supervisors need to practice active listening, which is demonstrated through the use of paraphrasing. Some university supervisors need to examine the type of questions they are asking. Closed questions dominated their conferences and that type of question should be used sparingly as they often require a single answer and don’t foster reflection. Questions should also be connected and based on the teacher candidates’ response to the paraphrase or previous question.

The findings revealed subtle changes in beliefs on the part of the university supervisors and the teacher candidates. Forty-five percent of the university supervisors did have teacher candidates whom experienced a significant change in beliefs according to the results of the paired t-test on the MBI scores. Fifty-five percent of university supervisors had a belief change in the learning construct. The belief construct curriculum was an area that teacher candidates’ exhibited a significant change under the supervision of thirty-six percent of the university supervisors. One university supervisor had a significant change in efficacy beliefs for her teacher candidates. The results of the teacher candidates’ MBI scores are found in the Table 8.

Three of six university supervisors who did not experience change in their teacher candidates’ beliefs exhibited some negative behaviors or comments during the study. These negative factors could have attributed to the insignificant results of their teacher candidates. Another supervisor was new to the position with the university, so her getting acclimated to the role could have impacted her influence on the teacher candidates. The fifth supervisor had three years of experience. Examination of her two post-observation conferences found that she did not pose any questions related to mathematics in the lesson during one conference and the second conference included only two questions out of the twelve she asked that were related to mathematics. So the inconsistent focus on mathematics content and instruction could be a contributing factor to her insignificant results. Likewise, in interviews with the teacher candidates it was noted that the university supervisors with no change in beliefs were the ones who weren’t as strong supporting mathematics teaching.

In order to examine the impact of the methods course on the teacher candidates’ beliefs, the data were also analyzed by methods instructor. Two instructors had significant change in all three of the belief constructs (curriculum, learning, and efficacy). A third instructor had a significant change in two of the constructs, and the fourth instructor did not have any significant change in the beliefs of her teacher candidates. This instructor was a part-time adjunct instructor who was teaching the course for the first time. Also seven of the sixteen teacher candidates in her class were not observed by a university supervisor which may have been due to a lack of accurate communication of requirements. This means that a university supervisor did not have a post-conferences with these teacher candidates. Therefore, these teacher candidates did not have the same opportunity to be coached and reflect on their teaching of mathematics as other teacher candidates in the program. The results for the MBI were also grouped by instructor and are found in Table 9. Regrettably, the data could not be analyzed to compare mentor teachers.