08 September 2011 10:58AM
When I signed up for Dr. Michelle Connolly’s honors thesis seminar in the spring of my junior year, I knew only broadly that I was interested in the subfield of the economics of education, an interest that developed out of being a tutor for America Reads and Counts in Durham and a teaching assistant for the Duke Talent Identification Program. The thesis seminar was the perfect place for me to get an introduction to the field, narrow my interests to a specific issue, and then bounce ideas for further exploration off of my fellow classmates. By looking through past honors theses, I was able to get a feel for what students have been able to do in the past and learn about the resources available to Duke students.
I came across a past thesis where a student was able to work with the North Carolina Education Research Data Center (NCERDC), an incredible resource housed at the Center for Child and Family Policy at the Sanford School of Public Policy. To my knowledge, this is the only statewide panel data set in existence that uniquely identifies and matches students and teachers. It’s the kind of data that draws researchers from all over the country and I feel very fortunate to have access right here in Durham.
Following that discovery, I started searching for papers that had utilized the NCERDC data set to see what has already been done and then find a way to make a meaningful contribution of my own. What excited me the most was a recent paper that attempted to quantify the peer effects that exist amongst teachers. That is, how the quality of the teachers in your peer group affects the outcomes of your own students. Peer effects have been studied in a variety of contexts, from berry pickers to super market clerks to freshman roommates, but this was the first time they had been empirically tested for in teachers.
Where I am trying to make my mark is on the introduction of Teach for America status into this equation. Teach for America teachers constitute a very unique subset of educators with regard to variables such as age, education, experience, and licensure and labor attachment. Therefore, it is possible that TFA teachers’ interactions with peer teachers may be significantly different from traditionally certified teachers. Given the anecdotally terse relationships between TFA and the traditional teaching community, one can imagine a scenario in which the introduction of TFA teachers into a school may breed hostility or competition that could be a driver of peer effects. Alternatively, TFA teachers may bring new pedagogical techniques into a school system given their unique training. I am testing for the existence of this influence, be it positive or negative.
Having the summer to research has given me the time necessary to really understand a complex data set like the NCERDC’s. Teach for America was kind enough to provide me with match information for their corps members, and the NCERDC has been extremely gracious in their customization of the data using these matches. In addition, being in Durham has allowed me to remain in contact with my faculty advisor Dr. Thomas Nechyba, who has been a great resource for questions both technical and theoretical. Finally, access to the computing resources at the Social Science Research Network has provided the power and data security that this kind of undertaking demands.
Economics Department