TEACHERS & PREPARATION INSTITUTIONS - USING DATA

Gathering Data

The data needed to answer these questions requires regular collection of information from individual teachers (and potential teachers) at key points in each teacher’s development. These key points include the following:

• Graduation from high school;
• Admission to an undergraduate institution;
• Graduation from an undergraduate institution;
• Completion of teacher licensing requirements;
• Employment (including ways to identify those who work outside of public schools and outside of a state);
• Employment as a teacher each year after graduation; and
• Participation in programs, including teacher preparation programs, but also recruitment programs such as Teach for America and Troops for Teachers, hiring incentive programs, new teacher induction, and graduate programs.

The key to fully understanding teacher supply is information on individual teachers that can be linked with other data on the teachers across time, with data about the schools that employ the teachers, and with data on the students that teacher serves. That means that a central component to all data collection is the use of common identification (ID) numbers for teachers, students, and schools. School ID numbers are the most simple, since the U.S. Department of Education maintains a public school numbering system in the Common Core of Data. 

Teacher and student ID numbers can be more difficult. Many states are creating unique student ID numbers that allow tracking students through the education system. Massachusetts provides a description of this effort. Some states have unique teacher ID numbers that include the use of certification numbers. However, teacher Social Security numbers (SSNs) play a central role in learning about teachers. By definition, SSNs are unique to each teacher. This is critically important for linking data sets across time, since identifiers such as names can change over time and numerous different institutions such as high schools, colleges, school districts, and other employers typically collect the SSN. In other words, the SSN provides the key link that allows data to be used and shared between government agencies including colleges and universities, state departments of education, different states, and potentially state unemployment agencies over any time period. 

The data that should be collected include the typical demographic data, program data (where people go to school and work and their performance in terms of grades and assessments), and most importantly for teachers, data on the students they teach and information on their achievement. These data are currently collected by some education agencies (i.e., schools, districts, and colleges) as part of normal record keeping. However, fully understanding teacher supply also requires regular collected survey data from teachers and their supervisors. These surveys should be administrated at key junctures of a teacher’s career, including on first entering the  teacher workforce, at tenure, and later in the career. These should be standardized surveys providing opinions about preparation, induction, future plans, and areas of strengths and weaknesses.

Finally, understanding teacher supply across states can require two different types of information. First is the sharing of the above Teacher-level data must be shared between state departments of education and/or with state unemployment insurance agencies. The sharing of data on people who were prepared to teach or have taught within a state provides detailed information on teacher flows. However, these teacher flows are often part of larger interstate migrations, the second type of information, which is described in data provided by the U.S. Census Bureau and the IRS.

Analyzing Data

Simply collecting the data is only the first step in this effort. Once data are collected, they must be analyzed, disseminated, and acted upon. Key to ensuring that the analysis of the data is as simple as possible is the creation of data sets that are as clean and as simple as possible. Data sets that require excessive amounts of data management and manipulation will create a serious impediment to the use of the data. Indeed, the hardest part of learning about teacher supply is often not the analysis, but instead the data management, linking, and “cleaning.” Central to all of these analyses is linking multiple datasets together. Creating data sets that can be linked and ensuring data are linked appropriately are central challenges to conducting this analysis and require time and resources.  

Analysis of the data often involves some relatively simple descriptive statistics. A short description on how to do that analysis (as well as demand analysis) is located at Mid-continent Research for Education and Learning.

However, some questions may require the use of very sophisticated statistical techniques. For example, a paper by Teacher Policy Researchers honed in on the importance of the distance from home when teachers take their first jobs, whereas a paper by Ric Hanushek and colleagues focused on the connection between teachers’ valued-added achievement gains and their mobility.

Last updated: February 6, 2006