TCS 2010-2013 Supporting Statement Part A

The National Center for Education Statistics (NCES) is requesting OMB approval for the Teacher Compensation Survey (TCS) data collections for school years 2009–10 through 2012-2013. The TCS is a new annual collection of basic information pertaining to the compensation of public school teachers and is part of the Common Core of Data (CCD) survey system. The TCS information is drawn entirely from existing state education agencies’ administrative records systems. The state education agencies (SEAs) providing the administrative records for the TCS also provide responses to the CCD school finance surveys. The TCS data is smaller in magnitude than the other CCD surveys and augments the collection.

Six surveys now comprise the CCD survey system. These are:

State Nonfiscal Survey of Public Elementary and Secondary Education;
Public Elementary/Secondary Local Education Agency Universe;
Public Elementary/Secondary School Universe;
Local Education Agency Financial Survey;
National Public Education Financial Survey (NPEFS);
Teacher Compensation Survey (TCS).

The State Nonfiscal Survey of Public Elementary and Secondary Education, Public Elementary/Secondary Local Education Agency Universe, and Public Elementary/Secondary School Universe data will be collected by EDFacts.

The Local Education Agency Financial Survey is co-sponsored with the U.S. Census Bureau, Governments Division, which is responsible for securing clearance for that survey. It is discussed in this request only as it relates to the National Public Education Financial Survey.

The National Public Education Financial Survey (NPEFS) is conducted by NCES and received a 3-year OMB clearance on December 10, 2009 (OMB Control No. 1850-0067 v.7).

The TCS is a new survey conducted by NCES, which began as a pilot study. The TCS pilot was initiated in 2007 with data collection for the 2005-2006 school year from 7 states, followed in 2008 by data collection for the 2006-2007 school year from 17 states, and in 2009 by data collection for the 2007-2008 school year from 17 states (approved under OMB#1850-0803 v.8). In 2010, the TCS has at least 24 states volunteering data for the 2008-2009 school year. NCES will continue to recruit new states to the survey and ultimately build a complete universe of teacher compensation data. In accomplishing this goal, NCES will work with SEAs to ensure accuracy and comparability, as well as confidentiality of personally sensitive data. Simultaneously, the U.S. Department of Education is providing the Statewide Longitudinal Data Systems (SLDS) grants that support SEAs’ ability to efficiently and accurately collect, mange, analyze, and use education data, including individual student and staff records. The SLDS grants support the inclusion of education-related data from preschool through postsecondary and workforce, including employment, wage, and earnings information. As new SEAs participate in the collection, the TCS will grow from a cross-sectional to a longitudinal dataset, expanding the possibilities for time series research.

The America COMPETES Act provides in part that with respect to preschool through grade 12 the SEAs should develop a teacher identifier system with the ability to match teachers to students. These teacher identification systems are directly analogous to the unique personal identifiers for teachers that are already part of the TCS collection. Finally, NCES anticipates that compensation data related to benefits, which currently remain elusive, will become increasingly available as SEAs continue to coordinate their data systems.

NCES has published a Research and Development Report from the first pilot collection, An Exploratory Evaluation of the Data from the Pilot Teacher Compensation Survey: School Year 2005–06, available at http://nces.ed.gov/pubs2008/tcspilot/, and the Research and Development Report for the second pilot year, An Evaluation of the Data From the Teacher Compensation Survey: School Year 2006–07, is currently undergoing review (the link will be provided once the report is published).

A.1 Importance of Information

The TCS is the first attempt to collect salary, benefits, teaching experience, and other data on public school teachers across the country. The authorization to collect this information is encompassed by NCES’s duty to “collect, report, analyze, and disseminate statistical data related to education in the United States” (P.L. 107-279, Part C, Sec 153). Participation in the TCS is voluntary, and an average of 31 SEAs per year are expected to report data over the next 3 years.

A.2 Purposes and Uses of the Data

The TCS collects a limited amount of information on individual public school teachers. The data can be used by researchers to study the distribution of teachers across schools and to make comparisons of teacher salary and benefits. Because the TCS can be linked to other CCD surveys, researchers can also explore the relation between school characteristics and teacher characteristics such as compensation or level of experience. Administrators can use these data to get a more complete picture of teacher compensation in various jurisdictions, as well as information on teachers’ education, years of experience, age, race/ethnicity, and gender. The data can also be used to calculate average teacher salaries at the school and district levels, and for calculating the average total compensation teachers receive.

A.3 Improved Information Technology (Reduction of Burden)

The TCS data are collected electronically from SEAs. A record layout is sent to state coordinators, who create a file and then upload the file through a secure web site. The SEAs are requested to fill out a data plan that provides further information about the data, such as any exceptions to TCS definitions. Passwords are used to limit access to the data. The Census Bureau assigns an NCES teacher ID based on the statewide teacher ID and maintains a crosswalk of statewide and NCES Teacher IDs. NCES never sees the statewide teacher ID, which is considered sensitive information in some jurisdictions.

A.4 Efforts to Identify Duplication

NCES has established that there are no other surveys designed to collect data on salaries for the universe of public school teachers that have the capacity to link teacher salary data to individual schools. The salary data that are included in the CCD School District Finance Survey and National Public Education Financial Survey are totals at the district and state levels, and do not include information about benefits. The NCES Schools and Staffing Survey collects salary and benefits data for a sample of teachers every 3 to 4 years. The Bureau of Labor Statistics collects salary and benefits data on teachers at the state and labor market level. The most commonly reported teacher salary data come from the National Education Association (NEA). The NEA data are at the state level only.

A.5 Minimizing Burden for Small Institutions

No small businesses or other small entities are surveyed in the TCS collection. All of the TCS data are reported to NCES by state education agencies (SEAs). NCES does not specify how these data are collected. The responding agencies draw upon their existing administrative records to provide the data.

A.6 Frequency of Data Collection

Consequences if not collected. This survey provides researchers, policy makers, and administrators with teacher level data on salary, years of experience, highest degree earned, age, race/ethnicity, and gender. These data can be used to compute average salaries, average years of teaching experience, staffing needs, and other indicators for each school and school district in states participating in the survey. Although no Federal program currently requires these data, they are the single source of comparable teacher- and school-level information that can be used in policy analyses of the association between teacher salaries and other characteristics and school and student characteristics; and between teacher characteristics and various measures of student outcomes aggregated at the school-level. Further, the TCS is the only national source of longitudinal data about teachers as it links teachers to specific schools and school districts and tracks teachers over time within states. Though data from this survey are not used in any Federal grant programs, researchers and administrators have requested that NCES collect these data to enable research on the important aspect of education that relates to teacher compensation.

The statewide administrative record systems that supply these data collect the information annually for the SEAs’ own purposes. The TCS brings the data together in a common format to allow comparability with only a minimum added burden.

A.7 Special Circumstances

There are no special circumstances associated with the TCS.

A.8 Consultations Outside NCES

The 60-day Federal Register notice was published on May 6, 2010 (75 FR, No. 87, p. 24934). We have received no public comments in response to this notice.

NCES routinely confers with the jurisdictions that provide the data and with the researchers and policy analysts interested in analyzing teacher compensation information. In June of 2009, NCES conducted a TCS workshop with participating and prospective SEAs in order to identify problems that SEAs encounter in collecting the data, clarify data item definitions, discuss data availability and quality, discuss limitations of the TCS, and offer technical advice. Another TCS workshop for participating and prospective SEAs is scheduled for May 14, 2010. NCES seeks the advice of researchers by presenting the TCS at such conferences as the American Education Finance Association Conference in 2008 and 2009, at the American Education Research Association in 2009, the Joint Statistical Meetings of the American Statistical Association in 2010, and the Federal Committee on Statistical Methodology conference in 2010.

A.9 Payments or Gifts to Respondents

This collection is from SEAs and therefore does not involve payments or gifts.

A.10 Assurance of Confidentiality

The TCS draws upon existing statewide administrative records, and thus no assurance of confidentiality is provided directly to individual teachers. Also, no teacher names, addresses, phone numbers, or social security numbers are collected. The data collected for each teacher include statewide unique numerical identifiers, base salaries, total salaries, school district (or SEA) expenditures for employee benefits, race, gender, year of birth, years of experience, school where the teacher is assigned, and status (such as full-time or part-time). The data collected on TCS are already available to the public in many jurisdictions, but in some teacher level data are not publically available. Given the confidential nature of these data in some jurisdictions, NCES will collect all TCS data in compliance with section 183 of the 2002 Education Sciences Reform Act (ESRA). Respondents will be assured that the information they provide is protected by federal statute (ESRA; P.L. 107-279, Title I, Part E, Sec. 183) and that their answers may be used only for statistical purposes and may not be disclosed, or used, in identifiable form for any other purpose except as required by law.

Data are collected through a secure web data collection site and are stored on secure servers. NCES and the Census Bureau conduct many confidential surveys and have the experience, protocols, and equipment to keep the data safe. NCES and Census staff who work on this survey are required to take several courses annually on the handling and storage of confidential data. Passwords and secure file transfer protocols will be used in order to limit access to the files. NCES will only make detailed, teacher-level data available to licensed users through a restricted-use data file and these files will not include the teacher IDs provided by the states (NCES assigns new teacher identification numbers). NCES plans to release school-level and district-level summaries of the data. We will employ appropriate disclosure avoidance techniques, such as data perturbation, whole case suppression, and/or multiple case suppression in order to protect the salary and benefit data from disclosure to the fullest extent possible.

A.11 Sensitive Questions

There are no questions of a sensitive nature on the TCS. The TCS will collect data on salaries and year of birth of teachers. The salaries, years of experience, and qualifications of public employees are typically matters of public record. Regardless of whether individual SEAs allow public access to these records, NCES will treat the data as confidential.

A.12 Estimates of Burden

The information reported on the TCS has already been collected by the reporting agencies for their own uses. The added burden for the CCD is limited to the SEAs’ effort to extract data from files, transfer them to NCES, and respond to edit reports.

The estimated hours of burden are based upon feedback provided by the 17 SEAs that participated in the collection of TCS data from March-September 2009.

The 2009 costs were derived from information about the salary and benefits costs of the SEA employees who work with NCES testing programs. These staff persons are comparable in knowledge and experience to the technical staff that report the CCD data. Across 50 states, and the District of Columbia, the average hourly rate estimated for technical staff reporting TCS data is estimated to be $44.25. It was estimated that 20 percent of the time spent with the TCS collections may be contributed by managers, who are estimated to be paid at double the rate of technicians at $88.50 per hour.

Estimated Annual Burden and Cost in Reporting CCD TCS Data

Teacher Compensation Survey	Average Hours	Respondents	Total Hours	Total Cost
SEA Technical Staff - Total	86	31	2,666	$141,291
Technician (80% of burden time at $44.25)	69	31	2,139	$94,651
Manager (20% of burden time at $88.50)	17	31	527	$46,640

Burden hours are estimated to be 86 hours per state. The burden for the anticipated 31 participating states totals 2,666 hours per year. This is a voluntary data collection, and we hope to get more states participating every year. In the collection of the 2007-08 data, we received data from 17 states. We are estimating that an average of 31 states per annum will provide data over the 3-year period (estimating about 27 states in 2010-11, 31 in 2011-12, and 35 in 2012-13), at the total estimated cost to respondents of $432,873.

A.13 Total Annual Cost Burden

There are no additional record keeping costs to the responding SEAs. All information collected on the TCS is from administrative record systems and is already collected by the SEAs for their own purposes.

A.14 Annualized Cost to Federal Government

Annual costs are based on current costs for the CCD. The costs include a 22 percent load on salaries for Department of Education staff. Contractor hours are also loaded costs. Department of Education staff assigned to the CCD NPEFS include 60% of one FTE GS/14 staff. One-sixth FTE of the program manager’s (GS/15) time is included. One full-time research analyst is employed through the Education Statistical Services Institute, a contractor to NCES.

Estimated Annual Cost of CCD to Federal Government

Cost Type	2010
Dept ED Staff	$100,528
Census (IAD)	$747,000
Contracted Staff	$127,890
Training	$250,000
Total	$ 1,225,418

Costs reported for the Census Bureau include staffing and other charges. The cost was estimated by assuming a 5 percent increase from the 2008 collection of 2006-07 school year data. This includes the full cost of salary, benefits, overhead, and fees.

Training costs reflect the cost of general training and professional development for all state TCS coordinators. These costs also include all-day training sessions at NCES for new and existing TCS Coordinators, at an estimated cost of $2,000 for each person trained.

A.15 Program Changes or Adjustments

This is a new data collection so the respondent burden and costs to the federal government associated with it are new.

A.16 Plans for Tabulation and Publication

NCES will publish both Research & Development and First Look reports of analytical findings based on summaries of the data. The Research and Development (R&D) series at NCES has been initiated to share studies and research that are developmental in nature, to disseminate the results of studies that are, to some extent, the “cutting edge” of methodological developments, and participate in discussions of emerging issues of interest to educational researchers, statisticians, and the federal statistical community in general. First Look reports are short publications (typically 30 pages or fewer) that summarize the findings from a data collection.

NCES also plans to release public-use data files that aggregate the data at the school and district levels. Whole case suppression and/or multiple case suppression will be employed to prevent the possible identification of any individual teacher. A restricted-use file of the teacher-level data will be made available to licensed users. Eventually, after study of the initial years’ results, NCES would like to make available through NCES data tools (e.g. ‘Build-a-Table’ and the Peer Search) some of the public use TCS data aggregated to the district level,. These web tools allow users to create their own lists, counts, or cross-tabulations.

The tables below show how the TCS data will be presented in reports as simple cross-tabulations.

Shape3

Shape4

Shape5

Annual Timeline

The TCS is an annual collection, and the schedule is shown below. Note that the files cannot be closed until the last SEA is able to report all requested data.

TCS Survey Collection, Processing, Publication

February Mail instructions to SEA respondents, open website to receive data.

March First reporting date.

September Close TCS file on Tuesday following Labor Day.

March NCES review of prior year files, file documentation, short reports.

May Training for new TCS Coordinators.

July Release of reports and data files.

A.17 Display OMB Expiration Date

NCES displays the OMB expiration date on the instruction manual sent to the SEA CCD Coordinators and on the data collection web site.

A.18 Exceptions to Certification Statement

There are no exceptions to the certification statement.

NOTE: Regarding race/ethnicity data

On October 19, 2007, the U.S. Department of Education posted to the Federal Register the “Final Guidance on Maintaining, Collecting, and Reporting Racial and Ethnic Data to the U.S. Department of Education.” (See the Federal Register, Volume 72, Number 202, pp. 59266-59279: http://a257.g.akamaitech.net/7/257/2422/01jan20071800/edocket.access.gpo.gov/2007/pdf/E7-20613.pdf).

SEAs will be required to implement this guidance in order to report data for the 2010–11 school year. Under the guidance, although not required to do so, SEAs already collecting individual-level data in the manner specified by the notice are encouraged to immediately begin reporting data in this format to the Department. The new race and ethnicity reporting categories will not be available for reporting TCS data until the 2008-09 collection year, which begins in April 2010.

The current and new race/ethnicity categories for reporting TCS data are outlined below.

Current TCS Reporting Categories

1 = American Indian / Alaskan Native

2 = Asian / Pacific Islander

3 = Hispanic

4 = Black, Not Hispanic

5 = White, Not Hispanic

6 = None of the Above

New TCS Reporting Categories

1 = Non-Hispanic American Indian / Alaska Native

2 = Non-Hispanic Asian

3 = Hispanic of Any Race

4 = Non-Hispanic Black or African American

5 = Non-Hispanic White

6 = Non-Hispanic Native Hawaiian or Other Pacific Islander

7 = Non-Hispanic Two or More Races

File Type	application/vnd.openxmlformats-officedocument.wordprocessingml.document
Author	barbara.spinner
File Modified	0000-00-00
File Created	2021-02-02