Environments that encourage mathematics graduate teaching assistants : the effects of institution type and availability of training

Abstract

This dissertation examined factors which are related to mathematics graduate teaching assistants' (GTAs') teaching attitudes and perceptions of the support for good teaching in their respective math departments. The research questions addressed differences between math GTAs at four groups of universities, categorized using an institution's Carnegie Classification and the availability of teacher-training for GTAs, in regard to math GTA perceptions of the support for good teaching in their department and in regard to math GTA attitudes toward teaching. Additionally, correlations between math GTA perceptions for support and math GTA attitudes toward teaching were examined. Finally, through use of multiple data sources, a qualitative analysis of the primary teaching support structures available to math GTAs at the participating universities in the four groups was conducted.

The majority of data were collected through the Teaching Assistant Survey which was distributed in Fall 2005. Participants were 252 math GTAs representing 13 Research Universities - Extensive (R/E), and 16 Research Universities - Intensive (R/I). A brief synopsis of the study results is as follows: Math GTAs at R/I Universities perceived more support for good teaching in their departments than math GTAs at R/E Universities. Math GTAs at R/I Universities reported more positive attitudes toward teaching than their counterparts at R/E Universities. There existed a weak positive correlation between perceived departmental support for good teaching and GTA attitudes toward teaching. The effects of training on GTA attitudes toward teaching and perceptions of support for good teaching were inconsistent. The qualitative analysis of a particular open-ended survey item about support for good teaching suggests that the construct of departmental culture and the ways in which a department can support good teaching are far more complex than anticipated.