2018 Supporting Statement Part B

Department of Commerce

U.S. Census Bureau

OMB Information Collection Request

Annual Survey of School System Finances

OMB Control Number 0607-0700

Part B – Collections of Information Employing Statistical Methods

1. Universe and Respondent Selection

The universe consists of approximately 15,084 school systems as counted in the most recent list of school systems in the Governments Master Address File and distributed as follows:

Independent School Districts 12,478

Dependent Educational Service Agencies (SEAs) 1,198

Dependent School Systems (not SEAs) 1,408

The Department of Education’s National Center for Education Statistics (NCES) has maintained annual reimbursable agreements with the Census Bureau that require the collection and dissemination of universe data. The full universe of school systems are included in the annual survey.

Due to the use of central collection from state education agencies, the approximate response rate for the 2015 school district finance survey cycle was 95 percent.

2. Procedures for Collecting Information

The form, F-33, contains the elementary-secondary education finance items collected by the Census Bureau. In practice, this form serves more as a data-reporting guide for respondents rather than as a data collection instrument. The Census Bureau relies heavily on collecting this public school system finance data centrally at the state education agency. All states provide significant amounts of these data centrally to the Census Bureau via the Internet using File Transfer Protocol (FTP). Supplemental forms are sent to school systems in states where the state education agency cannot provide information on assets (F-33-L1), indebtedness (F-33-L2), or both (F-33-L3).

The Census Bureau facilitates central collection by accepting from states data in one of two formats. Currently, 19 states provide the Census Bureau electronic copies of state-specific detailed education finance data files. The Census Bureau maintains programs for converting these data from the state agency format to the Census Bureau F-33 format. Thirty-two states reformat state-specific data files into the Census Bureau’s format prior to submitting the data electronically to the Census Bureau.

All school districts are canvassed in the survey. As such, there is no sampling error for the school district estimates. However, data are subject to nonsampling errors such as respondent error, coverage error, nonresponse error, and processing error. To mitigate the impact of the nonsampling errors, editing and estimating for missing data items (e.g., prorating state level amounts to school districts) are used.

3. Methods to Maximize Response

Nonresponse does not have a major impact on the data from this survey. State education agencies provide most of the finance data centrally for their school systems and this survey benefits from the multiple methods that states can use to increase response. Fiscal data reported to state education agencies by school systems are important factors in determining how much state money each school system receives, and thus this incentive helps minimize nonresponse. Fiscal data are used to determine Title I and Impact Aid federal grant allocations.

The Census Bureau uses traditional mail canvass techniques to enhance response for the forms it sends to school systems to obtain debt and asset information. Conducting follow-up mailings to nonrespondents, initiating central data collection techniques, soliciting audit reports in lieu of completed questionnaires, and an Internet collection for debt and asset data are several methods the Census Bureau uses to maximize response.

The Annual Survey of School System Finances has unit and item response rate goals of 85 percent, which meet the response rate thresholds specified in OMB, Census Bureau, and NCES Quality Standards.

The Annual Survey of School System Finances has always exceeded these response rate goals due to long-standing central collection agreements with each state and the District of Columbia. State education agencies generally require school districts to submit complete financial data to them. The survey has received a file from every state respondent every year, which ensures a high response rate.

4. Test of Procedures or Methods

The Census Bureau has not recently conducted procedural tests for the Annual Survey of School System Finances. In 2016 we added 2 data items through a nonsubstantive change. The new data items are essentially an aggregate of existing items. Where there was no formal cognitive testing, we were in contact with respondents and NCES through a webinar and our summer data conference to discuss and determine the wording of these data items.

5. Contacts for Statistical Aspects and Data Collection

Statistical Aspects monitored by:

Magdalena Ramos, Assistant Division Chief, Statistical Research and Methodology, Economic Statistical Methods Division, (301) 763-4295.

Data Collection questions to:

David Gromos, Chief, Education Finance Branch, Economic Reimbursable Surveys Division, (301) 763-4659

Stephen Wheeler, Supervisory Survey Statistician, Educational Finance Branch, Economic Reimbursable Surveys Division, (301) 763-9950

Attachments to the Supporting Statement:

Appendix A: Announcement letter, Form F-33, Supplemental forms F-33-L1, F-33-L2, and F-33-L3, and Screenshots from online application.