Att_TCS Teacher Compensation Survey SS Part B 129

Part B. Collections of Information Employing Statistical Methods

The Teacher Compensation Survey (TCS) does not use statistical methods. The survey is intended to provide a limited amount of information on each public school teacher in the nation. The data are collected from state education agencies. Although the respondents are experts in their states, there are opportunities for potential error, either through technical mishap or misinterpretation of the intent of a survey item. NCES has developed edit routines to identify out of scope and duplicate records. These edits rely on internal logic checks; consistency within specified tolerances over time; and consistency within a given state. NCES does not plan to impute for missing data for certain items in this survey.

B.1. Describe the potential respondent universe.

The universe includes every public school teacher in the United States. The 2008 collection (of 2006-07 data) was the first time NCES has collected these data. This is a voluntary data collection, and we expect an average of 40 states per year to report over the 3-year period. The average response rate will be about 80 percent.

B.2. Describe the procedures for the collection of information.

None of the procedures described under this item (stratification, estimation, etc.) are used in the TCS collection.

B.3. Describe methods to maximize response and deal with the issues of nonresponse.

NCES will provide guidance to state education agencies for creating a data file for reporting these data. NCES will only request TCS data from those states that collect these data in their data system and are willing to report the data to us. We expect very few states to be able to report benefit data at this time, but we know of at least 2 states that have benefit data. NCES will not impute for missing items.

B.4. Describe any tests of procedures to be undertaken.

No such tests will be undertaken. This is a new, voluntary, survey and we expect an average of 30 states per year to participate over the 3-year period. However, state representatives at the National Forum on Education Statistics have indicated that there is a need for these data at their states, and they have begun the process to collect these data from their school districts for their own purposes, as well as to respond to this survey.

B.5. Provide the name and telephone number of individuals consulted on statistical aspects of the design.

There are no statistical aspects to the design of the CCD collections and thus no individuals were contacted.