TEACHER FOLLOW-UP SURVEY
2009
OMB SUPPORTING STATEMENT
September 23, 2008
National Center for Education Statistics
A. Justification 3
1. Necessity of Information Collection 3
a. Purpose of this Submission 3
b. Legislative Authorization 4
c. Prior TFS Studies 4
d. Study Design for the TFS 09 and the TFS 10 4
2. Needs and Uses 5
3. Use of Information Technology 6
4. Efforts to Identify Duplication 7
5. Minimizing Burden 7
6. Consequences of Less Frequent Collection 7
7. Special Circumstances 8
8. Consultants Outside the Agency 8
9. Provision of Payments or Gifts to Respondents 10
10. Assurance of Confidentiality 10
11. Need for the Use of Sensitive Questions 11
12. Estimates of Hour Burden 12
13. Estimates of Cost Burden 12
14. Costs to Federal Government 13
15. Reason for Change in Burden 13
16. Project Schedule 13
17. Request Not to Display the Expiration Date 14
18. Exceptions to the Certification 14
B. Collection of Information Employing Statistical Methods 15
1. Respondent Universe 15
2. Procedures for Collection of Information 18
3. Methods for Maximizing Response Rates 19
4. Tests of Procedures and Methods 21
5. Reviewing Statisticians 22
C. Item Justification 23
PART A. JUSTIFICATION
1. Necessity of Information Collection
a. Purpose of this Submission
This document is a request from the National Center for Education Statistics (NCES) of the U.S. Department of Education (ED) for renewal of clearance of the activities occurring in calendar year 2009 related to the 2008-09 Teacher Follow-up Survey (TFS 09) to the 2007-08 Schools and Staffing Survey (SASS 08). All prior TFS collections have been cleared under OMB No. 1850-0617. Since the TFS now occurs on a four-year cycle, the prior clearance expired on 09/30/2007. As a follow-up to SASS, the TFS is a survey of teachers, and its main purpose is to provide a one-year teacher attrition rate.
Early activities for TFS 09, those that take place from September through December 2008, were cleared as an addendum to the SASS 08 clearance (OMB No. 1850-0598: Expires 02/28/2010)). Those activities include re-contact of SASS schools to determine the schools’ best estimates of the status of the teachers sampled for SASS. The design allows for differential sampling of teachers believed to be “stayers”, “movers”, and “leavers”.
Similar to earlier TFS collections, the TFS 09 sample is a sub-sample of the teachers who responded to SASS 08. This TFS will include all public school teachers who were in their first year of teaching in 2007-08, estimated to be about 2,000 teachers, and a sample of 3,500 teachers drawn using a sampling design similar to the one used in earlier TFS collections. Thus, the total sample size for TFS 09 will be about 5,500 teachers. Prior TFS samples included about 7,500 teachers. NCES has decreased the sample size in order to preserve resources for additional collections on the cohort of SASS first-year teachers. At a minimum, current plans include following the cohort of first-year teachers for at least one additional collection—in the 2009-10 school year. A clearance request for that collection will be submitted in the spring of 2009. NCES will be putting in for a budget increase to follow this cohort for as many as 10 years to examine the various career paths of teachers.
To preserve resources, the content of the TFS survey instruments has been cut. The only new items will go to first-year teachers. The new items cover such topics as teacher induction and mentoring. Again, NCES is scaling back in order to preserve resources for at least one additional collection on the first-year teachers.
Per earlier discussions with OMB, the TFS 09 will primarily be a web-based collection. This suits the TFS well because of the complicated branching within the questionnaire.
The U.S. Bureau of the Census will be conducting the collection under interagency agreements with NCES. Because the proposed TFS 09 is similar in content to prior TFS data cycles, NCES is requesting an expedited clearance for this survey.
b. Legislative Authority
The TFS 09 is sponsored by NCES, within the Institute of Education Sciences (IES). The Census will collect the data for NCES by authority of Public Law 107-279, Title I, Part E, Section 183, The Federal Statistical Confidentiality Order of 1997, and the U.S.A. Patriot Act of 2001, Public Law 107-56.
c. Prior TFS Studies
To date, NCES has completed five data cycles for SASS (1987-88, 1990-91, 1993-94, 1999-2000, and 2003-04) and five corresponding data cycles for TFS (1988-89, 1991-92, 1994-95, 2000-01, and 2004-05). The data from the TFS are linked to characteristics of those same teachers who participated in the prior year’s SASS.
The TFS 09, like earlier TFS collections, is designed to measure the one-year attrition rates of teachers who leave the profession and permit comparisons of stayers, movers, and leavers. “Stayers” are teachers who remain in the same school between the SASS year of data collection and the follow-up year. “Movers” are teachers who stay in the teaching profession but change schools between the SASS year and the follow-up year. “Leavers” are respondents who leave the teaching profession between the SASS year and the follow-up year.
The sample design of the TFS 09 will permit comparison of public versus private school teachers, new versus experienced teachers, and elementary versus secondary school teachers. Survey data collected in the SASS 08 will be used as crosscutting variables in analyzing TFS data. Prior SASS and TFS data have played an important role in improving our understanding of teacher supply and demand and the conditions that affect its balance. SASS and TFS provide national data on turnover in the teacher workforce, including rates of entry and attrition from teaching, sources and characteristics of newly hired teachers, and characteristics and destinations of leavers. These data help shift the debate from the issue of teacher quantity to teacher quality; that is, from its focus on teacher shortages measured in terms of the numbers of teaching positions left vacant to the qualifications of teachers who are hired and retained to fill teaching positions.
d. Study Design for the TFS 09
The TFS 09 will be conducted in 2008-09, one year after the SASS 08. The main difference between TFS 09 and previous TFSs is the composition of the TFS sample. The sampling frame for the TFS consists of all teachers who responded to SASS 08. The TFS 09 sample will consist of all public school teachers in their first year of teaching who responded to SASS 08. NCES expects there to be about 2,000 such teachers. Another 3,500 teachers will be selected according to the same design as earlier TFS samples. That is, the sample will be designed to allow for comparisons between public school and private school teachers and new and experienced teachers (less than 3 years v. more than 3 years of experience).
The major objectives of the TFS 09 are detailed below.
To determine the attrition rate for teachers.
To determine the characteristics of those who stay in the teaching profession, those who move from one school to another, and those who leave the profession.
To obtain major activity/occupation data for those who leave the teaching profession and career pattern data for those who remain in the profession.
To obtain data on educational activities and future plans.
To obtain data on attitudes about the teaching profession and job satisfaction.
The TFS 09 will primarily be a web-based collection, which is different from the mail/telephone/personal visit procedure of earlier TFS collections. Teachers will be encouraged to complete their questionnaires online. Paper versions will only be provided as a final resort to teachers who do not complete the web-based instruments on their own or by having a Census Bureau employee taking information by telephone and entering it in the web-based instrument.
For ease of administration (i.e., to eliminate complicated skip patterns), the TFS will have four separate hardcopy questionnaires. They are:
Teacher Follow-up Survey Questionnaire for Current Teachers, used to survey the stayers and movers.
Version for other-than-first-year teachers
Version for first-year teachers
Teacher Follow-up Survey Questionnaire for Former Teachers, used to survey the leavers.
Version for other-than-first-year teachers
Version for first-year teachers
2. Needs and Uses
The data will be used by Congress, the Department of Education, and other Federal agencies, State Departments of Education (also known as State Education Agencies or SEAs), private educational and other associations concerned with elementary and secondary education, and education research organizations. Data from previous surveys have been used by all of these sectors, and many requests have been made for NCES to update these data.1
As noted earlier, one of the major objectives of the SASS is to collect national data for estimating teacher turnover and for understanding attrition patterns. The TFS 09 is designed to be used in conjunction with the SASS to model the attrition behavior of public and private school teachers. The SASS and the TFS, conducted on a regular basis, provide necessary information for estimating and analyzing teacher turnover and for updating the turnover estimates used in projections of teacher demand.
There are a number of reasons to believe that teacher attrition rates are not static. First, the age composition of the teaching force changes over time, so the proportion of the teaching workforce nearing retirement also changes. The median age of the public school teacher workforce in 2003-04 was 43 years. Policymakers are concerned about the pending retirement wave that will occur this coming decade. In addition, recent national data, as well as data from a number of states and school districts suggest that annual attrition rates are especially high for inexperienced teachers during their first few years.2 The experience composition of the teaching force -- also related to the age distribution -- may be an important (and changing) variable. Third, labor market forces in teaching and in the general economy undoubtedly influence attrition. When teaching positions are scarce, temporary exits may be fewer due to expected difficulty in reentering; when other opportunities are plentiful, career changes are more likely. Finally, policy variables may also influence attrition rates. Incentives for early retirement, for example, became widespread in school districts during the 1990’s, as costs for experienced teachers rose. Such incentives may work, ironically, to produce shortages. Current policy initiatives, such as mentoring, offering student loan forgiveness, or additional bonuses for teachers who gain national certification, are intended to reduce attrition rates.
The U.S. Department of Education must be prepared to regularly estimate attrition rates for understanding the nature, composition, and dynamics of the teacher workforce. The continuing administration of the TFS is the key to producing these important data on a regular basis.
3. Use of Information Technology
For the TFS 09 data collection, teachers will be asked to respond through a web-based instrument. Data collection will occur in phases; the first phase will allow sample members to complete a self-administered web-based questionnaire. During the second phase, Census Bureau call center staff will complete telephone interviews and enter data using the same web-based instrument.
All members will be mailed a letter requesting participation in the survey through the on-line instrument. Sample members who have provided email addresses will also be contacted about the survey via two emails - one with the username, the other with the password. Email will also be used during various reminder phases throughout data collection.
Information technology will also be used throughout data processing and analysis.
4. Efforts to Identify Duplication
All available data sources were examined to determine that the data were not available elsewhere. Continuing discussions with state education agencies, private school associations, and other data providers and data users, as well as continuing review of other data sources within NCES and other Federal agencies and programs, indicate that similar information is not available. This dataset will permit analysis within each of the components of TFS (i.e., within the current teacher data file and the former teacher data file). This linkage across the different respondent groups makes the TFS dataset unique. TFS staff continues to monitor the field of teacher attrition and retention in preparation for subsequent administrations of TFS. As of this submission, no duplicate efforts have been identified.
5. Minimizing Burden
NCES has attempted to minimize burden on the TFS 09 teachers through (1) use of a web-based instrument, and (2) fewer items on the instrument. The web-based instrument will efficiently take respondents through appropriate skip patterns. At the same time, NCES has eliminated items on the TFS.
6. Consequences of Less Frequent Collection
While the periodicity of the SASS has varied over time (3 three-year intervals, followed by a six-year interval, followed by 2 four-year intervals), a TFS collection has always occurred one year after each SASS collection. If the TFS 09 collection is not conducted in 2008-09, there will be a minimum eight-year interval between the last TFS done in 2004-05 and the next, planned for 2012-13. That kind of interval will make it difficult for policymakers and practitioners to have useful information on teacher attrition.
7. Special Circumstances
There are no special circumstances that will require special data collection efforts.
8. Consultations Outside the Agency
Since its initial conception, the development of the TFS has relied on the substantive and technical review and comment of people both inside and outside the Department of Education. Through an established series of meetings held with various groups, the plans for content, design, analysis, and reporting of TFS data have been shared with researchers, policymakers, data providers at all levels, and data users. Their input has been important in redesigning the content and direction of the TFS. Below are the lists of those who have helped shape the TFS from outside of the Department of Education.
Current Advisors are:
Prof. Richard Ingersoll
Graduate School of Education
University of Pennsylvania
Philadelphia, Pennsylvania
[email protected]
Michael Strong
Director of Research
New Teacher Center, UC Santa Cruz
Santa Cruz, CA 95060
Howard Nelson
Senior Associate Director
American Federation of Teachers
555 New Jersey Avenue, NW
Washington, DC 20001
Dr. Martin West
Department of Education
Brown University
Providence, RI 02912
and for cognitive laboratory or focus group research:
Mr. Michael Long
Senior Manager
ORC/Macro
Calverton, Maryland 20705
[email protected]
Advisors in earlier rounds of the TFS include:
Prof. Michael Podgursky
Department of Economics
University
of Missouri at Columbia,
Columbia, Missouri 65211
[email protected]
Prof. Tim Sass
Department of Economics
Florida State
University
Tallahassee, Florida
[email protected]
Prof. Jianping Shen
Department of Teaching, Learning, and
Leadership
Western Michigan University
Kalamazoo,
Michigan
[email protected]
Prof. Edward Liu
Graduate School of Educaton
Rutgers
University
New Brunswick, New Jersey
[email protected]
Prof. David Figlio
Warrington College of Business
Administration
University of Florida
Gainesville, Florida
[email protected]
Prof. Susanna Loeb
School of Education
Stanford
University
Stanford, California
[email protected]
Prof. Susan Moore Johnson
Graduate School of Education
Harvard
University
Cambridge, MA
[email protected]
Prof. Jennifer Imazeki
Department of Economics
San Diego
State University
San Diego, California 92812
[email protected]
Prof. Daniel Goldhaber
Evans School of Public Affairs
University of Washington
Seattle, Washington. 98103
[email protected]
Prof. Steve Rivkin
Department of
Economics
Amherst College
Amherst, Massachusetts 01002
[email protected]
NCES makes semi-annual presentations to, and solicits input from, the Federal Statistics Cooperative System and its state members at the NCES annual Data Conference. NCES also holds annual meetings with the NCES Private School Group, consisting of representatives of private school associations. Key NCES staff members involved in these meetings are Mark Schneider, Jeff Owings, Kathryn Chandler, Kerry Gruber, and Steve Broughman.
9. Provision of Payments or Gifts to Respondents
There will be no provision of payments or gifts to respondents for the TFS 09.
10. Assurance of Confidentiality
From the initial contact with the participants in this survey through all the follow-up efforts, careful attention will be paid to informing potential survey respondents that NCES and the Bureau of the Census will protect their personal data. The respondents will be informed that the study is authorized by law under the Education Sciences Reform Act (Public Law 107-279) and their participation is voluntary and that their responses are protected from disclosure by federal statute (P.L. 107-279, Title 1, Part E, Sec. 183). All responses that relate to or describe identifiable characteristics of individuals may be used only for statistical purposes and may not be disclosed, or used, in identifiable form for any other purpose, unless otherwise compelled by law.
Some identifying information must be collected from teachers to: (a) permit follow-up of respondents who return incomplete surveys so that missing and inconsistent data can be corrected, and (b) permit at least one additional collection from the first-year public school teachers in the TFS sample. To accomplish these objectives, the instruments ask teachers to verify or update their names, home addresses, home telephone numbers, and emails. First-year teachers will also be asked to provide information about family members who would be aware of their contact information over time.
The web-based collection will be housed on NCES servers and data will be exported to the Census Bureau on a regular basis through a secure file transfer operation. Hardcopy returns will go directly to the Census Bureau for data capture and machine processing. Only authorized individuals will have access to the TFS data.
11. Need For the Use of Sensitive Questions
TFS 09 contains sensitive questions that pertain to personal or family income, the number of people that the teacher supports, and how many of these are dependents. The respondents are asked questions about their salary for the year, other types of income during the school year and the summer. Those who left teaching (and are working) are asked about their yearly salary and the combined family (or household) income. Data on income and compensation are important variables for analyses of average teacher salaries, overall teacher compensation, comparative income between current and former teachers, and teacher job satisfaction. Teachers are also asked for the number of people that they support and the number of dependents in the household. These variables will help NCES analyze financial burden of the teacher. It is important to have comparative income data and financial responsibility data between leavers and stayers or movers, as salary and financial burden may be explanatory variables in the decision to leave teaching or change schools. The family income and financial burden gives a broader look at the decisions made by teachers to stay in or quit teaching.
In this day and age, items requesting personally identifiable information, such as name, telephone number, address, and email, may also be deemed sensitive. Such items are never released, but are needed in the event that the respondent needs to be contacted to clarify some of his/her responses. The first-year teachers will also be asked to provide contact information for a friend or relative who does not live with them. These data are needed so that respondents can be found in future rounds of the TFS. NCES’ policy on the protection of personal information is discussed above in Section 10.
12. Estimates of Hour Burden
The projected number of sample units and the total estimated respondent burden for the TFS 09 are as follows:
Table A1 DETAILS OF INFORMATION COLLECTION BURDEN FOR TFS 09 |
|||||
Respondent Type |
Survey |
Sample Size |
Number of Respondents |
Estimated average response time per respondent |
Total Hours |
Stayers and Movers
|
Current Teacher Questionnaire |
2,000 |
1,830 |
17 minutes |
519 |
Stayers and Movers (First Year Teachers) |
Current Teacher Questionnaire |
1,700 |
1,500 |
34 minutes |
850 |
Leavers |
Former Teacher Questionnaire |
1,500 |
1,320 |
20 minutes |
440 |
Leavers (First Year Teachers) |
Former Teacher Questionnaire |
300 |
260 |
36 minutes |
156 |
TOTAL |
1,965 |
||||
These estimates are based on the average response times to the TFS 05, adjusted by the number of items on the TFS 09. The math for each of the rows is as follows:
Current teacher (other than first-year): (114 items in TFS 09/230 items in TFS 05) x 35min = 17 minutes.
Former teacher (other than first-year): (104 items in TFS 09/130 items in TFS 05) x 25 min = 20 minutes
Current teacher (first-year): (222 items in TFS 09/230 items in TFS 05) x 35 min = 34 minutes
Former teacher (first-year): (188 items in TFS 09/130 items in TFS 05) x 25 min = 36 minutes
NCES’s standard procedure for estimating cost to is to multiply the estimated total survey reporting hours (amount of time it takes to complete the survey) by the average salary of school employees (assumed to be $26.00 per hour). Following these assumptions, the total respondent dollar cost is estimated to be about $51,090 (based on responses from about 3,330 stayers and movers, and 1,580 leavers) for the TFS 09. Respondents for this survey will not incur any cost other than the time it takes to respond.
14. Cost to the Federal Government
The cost to the federal government for the Teacher Follow-up Survey is estimated to be $1.9 million for the work conducted by the Census Bureau. The Census Bureau estimates were compiled from individual estimates developed within each Census Bureau division involved in the survey. Estimates were based on the sample sizes, the length of the questionnaires, and the data processing requirements. Administrative overhead, forms design, printing, and mailing costs are included. NCES has also contracted with KForce Government Solutions, Inc., to produce the on-line instrument for the TFS 09. That work is estimated at $160,000. Finally, TFS is done with the support of the Education Statistical Services Institute (ESSI). Approximately $500,000 has been/will be spent on ESSI support for the TFS 09. Altogether, then, the TFS will cost the Federal Government about $2.56 million.
15. Reason for Change in Burden
The burden reported in section A12 above of 1,965 hours is less than the TFS 05 of 2,657 hours because of the reduction in sample size and survey length.
NCES anticipates that a set of reports similar to those produced on other NCES national surveys will be prepared, including:
A Survey Documentation report that will summarize the procedures for sampling, data collection, data control, and data processing; and
A “First Look” report that contains tabular summaries of basic data for dissemination to a broad audience.
Early TFS activities through December 2008, including teacher status operations and sample selection, were approved by OMB as an amendment to the clearance for SASS 08. The remaining TFS 09 activities will be conducted according to the following time schedule:
Teacher Data Collection |
|
Teachers with Email Addresses |
|
Send email and letter with Internet questionnaire link |
2/18/2009 |
Send reminder with Internet questionnaire link |
3/4/2009 |
Teachers without Email Addresses |
|
Send letter with Internet questionnaire link |
2/18/2009 |
Send reminder with Internet questionnaire link |
3/4/2009 |
Telephone data collection with all nonrespondents |
3/12 – 4/17/2009 |
Mail hardcopy questionnaires to nonrespondents |
4/20/2009 |
Telephone reminders/data collection |
5/4 – 5/29/2009 |
Edit and Key Questionnaires |
3/2/09 - 6/12/09 |
Process Data |
6/15/09 - 12/31/09 |
NCES Reports Results |
5/29/2010 |
NCES is not seeking approval not to display the expiration date of OMB approval.
There are no exceptions to the topics in Item 19 of Form OMB 83-1.
1 See Wiley, Reynolds, Cobb, and Luekens, Secondary Use of Schools and Staffing Survey Data, National Center for Educational Statistics, 1999. Also Rouk, Weiner, and Riley, What Users Say About Schools and Staffing Survey Publications, National Center for Education Statistics, 1999.
2 ?See Bobbitt, Faupel, and Burns, Characteristics of Stayers, Movers, and Leavers: Results from the Teacher Followup Survey, 1988-89, National Center for Education Statistics, 1991; Boe and Gilford, Teacher Supply, Demand, and Quality, National Academy Press, 1992; Bobbitt, Leich, Whitener and Lynch, Characteristics of Stayers, Movers, and Leavers: Results from the Teacher Followup Survey, 1991-92, National Center for Education Statistics, 1994; Whitener, Gruber, Lynch, Tiagos, Perona, and Fondelier, Characteristics of Stayers, Movers, and Leavers: Results from the Teacher Followup Survey: 1994-95, National Center for Education Statistics, 1997.
| File Type | application/msword |
| File Title | TEACHER FOLLOW-UP SURVEY |
| Author | #Administrator |
| Last Modified By | #Administrator |
| File Modified | 2009-02-11 |
| File Created | 2009-02-11 |