Statement of Work Postsecondary Education Descriptive Analysis Reports I. INTRODUCTION I.A. Purpose of Project The United States Department of Education has a requirement to provide a series of descriptive reports and data tables on postsecondary education topics. These reports and tabular analyses will draw on several large and complex data collections: national longitudinal surveys; linked cross-sectional surveys; census data; and administrative record surveys. In addition, the Department of Education has a requirement to provide a data library and a data analysis system that will encourage further analyses and facilitate access to the reports and tables. The contractor shall conduct data analyses, produce reports, prepare graphic materials, create and maintain a data library, and convene meetings and panels of experts. This statement of work contemplates ten analytic reports in each year, depending on the availability of funds. It is anticipated that the analysis and reporting activities shall be an incrementally funded, 5-year contract. The work will be monitored by the National Center for Education Statistics (NCES) within the U.S. Department of Education (ED). The activity will be referred to hereafter as the Postsecondary Education Descriptive Analysis Reports (PEDAR). I.B. Current Status NCES currently produces descriptive analysis reports, data libraries, and data analysis systems as an integral part of each postsecondary longitudinal data collection. The data libraries and data analysis systems that are produced with each of the surveys are very similar to the requirement specified herein for PEDAR. Current versions of data libraries and data analysis systems may be retrieved from the NCES home page (www.ed.gov/NCES). The descriptive analysis reports that are initially produced for each of the surveys differ, however, from the requirement for PEDAR. Analysis reports for data collections focus on the specific details for that survey and do not attempt the integration across different surveys and survey waves that is required by PEDAR. Nor do the analysis reports typically include a multivariate section to provide statistical controls for background differences and multiple subgroups. Most importantly, the analysis reports are not designed to focus on policy issues and discuss them using a variety of statistics and multiple sources of data. I.C. Program Overview The Postsecondary Longitudinal Studies program at NCES consists of two major data collections: high school cohort studies and postsecondary cohort studies. The former include: The National Longitudinal Study of the High School Class of 1972 (NLS:72) - followed through 1986; High School and Beyond, 1980 (HS&B) - followed through 1986 (senior cohort) and 1992 (sophomore cohort); and National Education Longitudinal Study, 1988 (NELS:88) - starting with the third follow-up in 1994. The high school cohort studies provide information about access and entry into postsecondary education (PSE). These surveys study single-grade cohorts as they complete high school and enter PSE or the workforce. Students are periodically surveyed thereafter as they complete their education and pursue their occupations. The high school cohort surveys are particularly well suited to study the traditional student who enters PSE soon after leaving high school. Because of attrition (not entering PSE, stopping to work, or leaving school), these surveys cannot provide detailed information about PSE experiences at all levels and in all types of institutions. In particular, the high school cohort surveys provide valuable comparative data for those students who never enrolled in postsecondary education. Important comparisons include labor force participation data as well as information about family formation. To complement the high school cohort surveys, and compensate for attrition, NCES provides postsecondary cohort surveys: National Postsecondary Student Aid Studies (NPSAS:87,90,93,96) cross-sectional surveys; Beginning Postsecondary Students (BPS:90/92/94 and BPS:96/98); and Baccalaureate and Beyond (B&B:93/94/97). These surveys are based on PSE populations, so they represent multiple age cohorts, non-traditional students, as well as many types of postsecondary institutions (public, private not-for-profit, private for-profit, less than 2-year, 2-year and 4-year). They complement the high school cohort surveys by providing adequate sample sizes to follow students from their PSE institutions into the workforce or graduate education. The National Postsecondary Student Aid Study (NPSAS--1987, 1990, 1993, 1996) is a cross-sectional national survey of postsecondary students that provides baseline data as well as continuing information about program participation and finance. NCES also gathers administrative records from postsecondary institutions every year in the Integrated Postsecondary Education Data System (IPEDS) collection that provides background and contextual information. Both the BPS and B&B surveys are drawn from postsecondary cohort populations sampled in the National Postsecondary Student Aid Studies (NPSAS). BPS collects first-time, beginning student data from the 1990 and 1996 base years of NPSAS to study patterns of postsecondary enrollment, persistence in school, progress and changes in curriculum, and award of degree or certificate. In addition, BPS provides information about workforce participation, family formation, graduate school enrollment, costs of attendance, and financial assistance (scholarships, grants, loans, and other forms of support). B&B began with the 1993 NPSAS sample of PSE students. The timing allowed NPSAS students who just completed their bachelors degree to be sampled prior to leaving school. B&B provides information about transitions at the end of PSE and supplements and replaces the NCES survey of Recent College Graduates. In addition to these NCES national surveys, there are other data sources that may provide information for PEDAR. Data from the Bureau of the Census 1990 decennial census and Current Population Surveys may be useful to provide a context for analyses. The major policy topics covered by the NCES data, include but are not limited to the following: Access and choice in undergraduate PSE Persistence Progress Degree attainment Access and choice in graduate/first-professional PSE Rates of return for PSE investments Student financial aid effects Teacher training and careers. II. SCOPE OF WORK Independently and not as an agent of the United States Government, the contractor shall provide all personnel, materials, services, and facilities necessary for the project and perform the tasks as described below. With the exception of an on-going management task and the on-going task of servicing the data library, the work required under this contract will consist of producing a series of descriptive reports. Although each report will address different topics and may use varying data sets, each will involve comparable steps and procedures and will result in 50- to 75-page documents that are similar in scope and complexity. The topics for each of these reports shall be drawn from a menu and prioritized by a panel comprised of experts and others familiar with policy issues in postsecondary education. Annually, the contractor shall ask the panel to evaluate and establish priorities for the proposed topics. As topics are selected and while the reports are being prepared, some of the panelists shall continue to serve as reviewers. Each descriptive analysis report shall begin with a brief survey of the current literature to ensure adequate coverage of issues and avoid duplication of efforts. Once data analysis begins, the relevant data sets will be obtained. The contractor shall then merge these data files and select variables as necessary to extract working data sets. The contractor shall create composite or derived variables and construct data analysis system input files. NCES will create the data analysis system using the input files. Tables and figures produced using the data analysis system will constitute the main body of each descriptive report, along with discussion of the findings and their pertinence to the selected topic. Examples of descriptive reports can be found on the NCES home page (www.ed.gov/NCES/). Some NCES descriptive reports have not attempted to provide control for confounding variables (covariates). Cross-tabulations have provided limited statistical control, but some survey samples have not been sufficiently large to use multi-way cross-tabulations and maintain sufficiently large cell sizes for analysis. The reports required under PEDAR will provide statistical control for confounding variables. The reports shall include a section using multivariate methods to control for differences between multiple subgroups and between multiple background characteristics. In general, it appears that linear regression models provide adequate statistical control, but the contractor is encouraged to use any approach that will achieve the objective of clear and understandable descriptive analysis. All reports shall contain technical appendices that describe new variables and composites created for tables and figures in the report and describe design effects for all variables and standard errors for all table cells. All reports shall include an abstract of no more than two pages. As reports are developed, all variables, composites, and tables used in the report shall be accumulated in a data library. This library shall allow access to machine-readable tables for data users. In addition, it shall contain documentation for all constructed variables. The data analysis system shall be part of the data library and shall enable users to reproduce tables used in the report, modify variables or apply filters to construct tables that were not in the report, and access variables and composites developed during the course of preparing the report. Overall, PEDAR shall have three major components. The first encompasses all management tasks, including overall project quality control and development of the annual analysis plan (Task 1). The second encompasses all library tasks, including development and maintenance (Task 2). The third component is to develop a plan for a particular analysis topic and an outline for a report, develop a data analysis system to produce descriptive tables on the topic, and produce a descriptive analysis report on the assigned topic, complete with tables, figures, and appendices as appropriate (Task 3). III. OBJECTIVE The data collected from ED postsecondary education surveys shall be analyzed and ED's topics addressed. A report shall be written on each topic that summarizes the results of the data analysis. The data analysis system used in each analysis and the report shall be accumulated in a data library. This project will aid ED in describing the meaning and significance of statistical data and in disseminating this information. IV. DESCRIPTION OF TASKS The following tasks describe the work required to adequately meet the objectives of the PEDAR project. These descriptions shall be read in conjunction with the detailed specifications for the deliverables (see section on Deliverable Specifications) to indicate the complexity of the overall project and provide guidance in its conduct. Task 1. Management The contractor shall manage the analysis and reporting of data in a manner that establishes quality control at all levels and fosters communications between the contractor's staff, the Contracting Officer's Technical Representative (COTR), and potential users of the data. This task continues for the entire contract period. Task 1 includes the following requirements: Subtask 1.1 Post-Award Conference Within the first week after contract award, the contractor's project director and other key project staff shall hold a meeting at NCES with the COTR and other appropriate NCES staff to review the overall contract tasks, to identify potential problems and possible solutions, and to discuss areas of concern related to the proposed project staffing plan and other management requirements. The primary purpose of this meeting is to refine the management, staffing, and scheduling plans. These refinements are not to modify or alter the specifications of the contract, but to provide management information for use by both the contractor and the Government in monitoring the work to be performed. This conference is also to help the contractor to make use of the experience and materials that NCES staff have gained over the years. Subtask 1.2 Monthly Progress Reports The contractor shall report monthly on the progress made in accomplishing the project tasks, problems encountered, and plans for the next month. The contractor shall attach an updated schedule, a financial statement showing consumption of funds, and a staffing plan showing labor usage to each monthly report. Deliverables: Monthly progress reports Subtask 1.2.1 Schedules Due to the complexities of large data analysis projects, the contractor shall develop and maintain a detailed schedule for all activities of the project. The schedule shall be updated monthly and attached to the monthly progress report (subtask 1.2 Subtask 1.2.2 Financial Statements Due to the complexities of large data analysis projects, the contractor shall develop and maintain a detailed financial statement for all activities of the project, including budgeted and actual costs for each month and cumulatively. The financial statement shall be updated monthly and attached to the monthly progress report (subtask 1.2). Subtask 1.2.3 Staffing Plan Due to the complexities of large data analysis projects, the contractor shall develop and maintain a detailed staffing plan for all activities of the project. The staffing plan shall be updated monthly and attached to the monthly progress report (subtask 1.2). Subtask 1.3 Confidentiality Reports The contractor shall file reports as needed with the NCES COTR that identify staff with access to restricted data. Deliverables: Confidentiality documents Subtask 1.4 Technical Review Panel To obtain peer review of proposed topics, and to foster communications with potential users of the data, the contractor shall establish a Technical Review Panel (TRP) of experts in the field of postsecondary education. The PEDAR TRP shall consist of all TRP members for NPSAS, BPS, B&B, NELS and any other NCES studies that produce data used by PEDAR. The contractor shall work with the TRP to study feasibility and importance of proposed topics and to improve plans and analysis products; however, the TRP does not report to or advise NCES. The contractor shall also attend TRP meetings for other NCES data collections to ensure coordination between data development and analysis projects. Subtask 1.5. Annual Analysis Menu The contractor shall prepare annually a prioritized set of 15 or more topics for descriptive analysis reports. Deliverable: Annual analysis menu Subtask 1.5.1 Annual Draft Analysis Menu Each year, the contractor shall develop and distribute to TRP members information about proposed topics for analysis reports. The contractor shall also inform the TRP of related information, such as: the availability of related data sets; on-going analyses or related studies; and influential events, such as pending legislation related to the topic. Subtask 1.5.2 TRP Review of the Draft Analysis Menu The contractor shall have the draft analysis menu reviewed by the TRP. The reviews shall include discussions to identify problems, suggest solutions, highlight important factors to be considered, and raise relevant issues that shall be reflected in the annual analysis plan. The TRP may suggest additions or deletions to the analysis menu. After consultations with the COTR, the contractor shall develop a revised analysis menu. Subtask 1.5.3 TRP Prioritization of the Revised Analysis Menu The contractor shall submit the revised analysis menu to all members of the TRP. Each TRP member shall be instructed to select the five best topics. The contractor shall analyze TRP selections and develop the prioritized annual analysis menu. Task 2 Data Library The contractor shall develop and manage a linked set of files into a data library. Linkages to PEDAR files shall be established with a home page on the World Wide Web. In addition to welcoming descriptions of PEDAR and topics (completed and in progress), linkages shall be established to data analysis systems, computation platforms, completed reports (as word processor documents and in html or similar format), and staff (including TRP members) directories. The contractor shall have the accumulated files of the data library pressed onto a compact disk. Deliverable: Data library Task 3 Topic Analysis and Reporting For each topic selected from the annual analysis menu, the contractor shall prepare an analysis plan, data analysis system input files, and the associated descriptive analysis report. Subtask 3.1 Review Literature The contractor shall review the topic in the current literature to ensure completeness, avoid duplication, address TRP concerns, and suggest additional factors, variables, and background contexts that must be taken into account in preparing an analysis report on the topic. Subtask 3.2 Prepare Draft Analysis Plan The contractor shall identify data elements for the analyses. The plan shall include specifications for all variables and composite variables to be constructed and all tables to be generated. The purpose of this draft is to ensure that proposed table shells address the topic in the most complete and compelling way possible. Subtask 3.3 TRP Review of on the Analysis Plan The contractor shall share the draft analysis plan and literature review with the TRP by e-mail. Based on TRP comments, the contractor shall revise the analysis plan. Subtask 3.4 Construct Input Files for Data Analysis System The contractor shall extract data files, define data elements, and construct composite variables as needed based on the analysis plan. The contractor shall add the constructed variables to the existing data analysis system's set of variables and prepare input files for the construction of a new data analysis system by NCES. Deliverable: DAS input files Subtask 3.5 Development of Tables and Bullets Using the data analysis system, the contractor shall prepare tables of estimates according to the analysis plan. The contractor shall develop bullets describing significant findings. Subtask 3.6 TRP Review of the Tables and Bullets At a meeting in Washington, DC, or by e-mail the contractor shall obtain the comments of interested TRP members concerning the quality of the tables and bullets. Subtask 3.7 Develop Multivariate Model The contractor shall develop a multivariate model to provide statistical control for varying background characteristics, multiple subgroups, and other factors that affect relationships in the data that cannot be adequately reflected in cross-tabulations or paired comparisons. All variables specified in multivariate models shall be included within the variable sets of the associated data analysis systems. It is not the purpose of this subtask to advance methodological state-of-the-art. However, advances that enhance understanding of the data are within the scope of this task. Subtask 3.8 Develop Technical Appendices The contractor shall append all technical material that is required for a full and unambiguous understanding of the report's conclusions in technical appendices. These appendices shall include, at a minimum: bibliography for cited works; a glossary listing all variables used in the report; specifications for the multivariate model; illustrative standard errors for all parameters and comparisons that are cited in the text; and design effects for variables that are likely to be useful to readers. Subtask 3.9 Descriptive Analysis Report The contractor shall prepare an analysis report in accordance with the analysis plan, as amended by the TRP. The format and contents of this report are subject to publications standards and to a formal review procedure at NCES. The report shall comply with these standards and the consensus of the reviewers. Deliverable(s): Descriptive Analysis Report Subtask 3.9.1 COTR Review The contractor shall revise the report based on a review by the COTR. Subtask 3.9.2 Program Director Review The contractor shall revise the report based on a review by the NCES Program Director. Subtask 3.9.3 Senior Statistician Review The contractor shall revise the report based on reviews by the NCES Senior Statistician. Subtask 3.9.4 Adjudication Panel Review The contractor shall revise the report based on a review by the NCES adjudication panel. V. DELIVERABLE SPECIFICATIONS The specifications for all deliverables to be generated by the tasks described in the Description of Tasks section are presented below. Unless otherwise specified, deliverables shall not be physically delivered to the NCES COTR. Instead, the NCES COTR and CO shall be notified of the deliverable's placement within the data library for retrieval. V.A. Monthly Progress Reports (Subtask 1.2) The monthly progress report shall detail the progress made in accomplishing the project tasks, problems encountered, and plans for the next month. A financial statement showing consumption of funds, a staffing plan showing labor usage, and an updated schedule shall be attached to each monthly report. Delivery specifications: Monthly (on or before the 5th day) progress reports including attachements shall be delivered as Word files. Monthly reports and attachments shall be placed in the data library. V.B. Confidentiality Documents (Subtask 1.3) Confidentiality reports shall be in two (2) parts. Part 1 shall be a list of all project staff with access to restricted NCES data. Part 2 shall be a completed (signed and notarized) set of NCES Affidavits of Nondisclosure for all project staff who are listed in part 1. Delivery specifications: Within one (1) week of contract award, Part 1 shall be included within the data library as a Word file. Each NCES Affidavit of Nondisclosure shall be completed prior to access to restricted NCES data. The original NCES Affidavits of Nondisclosure shall be mailed to the NCES COTR in monthly batches attached to the monthly report. Part 1 shall be updated on a flow basis. V.C. Annual Analysis Menu (Subtask 1.5) The contractor shall schedule a series of analysis reports to be completed within fixed time periods for each annual cycle of the contract. There will be ten (10) reports in each year. The annual analysis menu shall describe the topics and issues to be addressed by 15 or more proposed reports. In addition, the menu shall briefly discuss the data sets proposed for use and outline general specifications for variables and describe limitations (such as sample size). Delivery specifications: The annual analysis menu for the first year of the contract shall be due within six (6) weeks of the contract award. All subsequent annual analysis menus shall be due in October (of 1998 through 2002). Menus shall be delivered as Word files. V.D. Data Library (Task 2) All data file extracts drawn from NCES or Census or other data collections shall be stored and documented. The extract files shall be in a form compatible with other NCES data libraries. All data analysis systems shall be stored and documented. All analysis menus and completed reports shall be stored and documented. Delivery specifications: Within six (6) weeks of contract award, the files and data library structure shall be available. Two weeks prior to the end of the contract, the contractor shall press all data library files onto a CD-rom for delivery to NCES. NCES will provide existing DAS input files within four weeks of contract award. NCES will construct new DAS system files within two weeks of receipt of contractor prepared input files. V.E. Data Analysis System Input Files (Subtask 3.4) Based on the approximately 50 derived variables, and up to 200 other extract file variables, a data analysis system shall be developed for the estimation of table values for each report. The contractor shall construct the five data analysis system input files as needed. Delivery specification: As needed for each analysis report, the five (5) ASCII input files for the data analysis system shall be delivered. The specifications for each of the files is listed in the attachement. V.F. Descriptive Analysis Reports (Subtask 3.9) As specified in the accepted topic analysis menu and using the accepted data analysis system to produce tables specified in the accepted analysis plan, the descriptive analysis report shall be developed. In addition to text describing group differences, each report shall include approximately 15 graphs and/or charts for the most interesting findings and a 2-page executive summary. The final draft version of each analysis report shall follow the overall format as specified in the OERI format standards (see Mandatory Contract Requirements section). Front cover to contain: the report number and date; title and period covered; contract number; contractor's name and address; and contact within NCES and address. Abstract to contain: the object of the report and a summary of findings. The abstract shall be in format suitable for use as a statistics in brief. Introduction to contain: an overview of current literature that identifies issues, sources of data, and relevant findings to date. Main body to contain: general descriptive findings bearing on the topic of the report using the items and constructed variables from the analysis plan. Additionally, the main body shall contain three or four chapters that treat the overall topic in functional subsets. Each chapter shall have tables and figures along with a discussion of findings that highlights trends and patterns in the data. These chapters shall include a detailed narrative or analysis of the work performed and phases of study. Any innovations and general advances of special interest shall be identified. The draft final report shall be delivered to the NCES COTR for the publication adjudication process meeting(s). The purpose of the adjudication meeting is to set final editing and technical specifications for the analysis report and to provide a final review for the document. Delivery specification: Twelve (12) weeks prior to the date for delivery of the analysis report to the OC, the report shall be delivered as a Word file. During the 12 week period, the report shall be revised to reflect the concerns of the COTR, the concerns of the program director, and the concerns of the senior statistician and Associate Commissioner (NCES). The report shall then be delivered for review by the Adjudication panel and the chief statistician. Finally, the report shall be delivered to Information Services. VI. INSPECTION AND ACCEPTANCE PROCEDURES The contractor shall consult with the NCES COTR before making any major decisions or developing the required deliverables. The contractor shall make frequent phone calls and visits to NCES to discuss potential project problems and progress. The contractor shall set up an electronic system for transferring information via electronic mail and microcomputers. In addition, the contractor shall provide access to files and systems within the data library and remote access for monitoring operations. The contractor shall prepare all reports following the guidelines as specified in NCES's standards (see Mandatory Contract Requirements section). The NCES review process may take 2-6 months, depending on the type of report. It is of paramount importance that schedules for publication releases shall be kept and that NCES standards for all phases of report production shall be followed. All materials must be approved by NCES before release. The contractor shall not release any data in raw or derived form to anyone without NCES's review and approval. Acceptance of deliverables is the responsibility of the Contracting Officer. The NCES COTR will prepare a memorandum to the Contracting Officer concerning the technical aspects of each deliverable and provide a copy of this memorandum to the contractor. The NCES COTR will review each deliverable within two (2) weeks of contractor delivery or notify both the contractor and the Contracting Officer of the expected date of review completion. VII. MANDATORY CONTRACT REQUIREMENTS VII.A Quality Control NCES is firmly committed to determining how well each project meets the project goals, the quality of data reported, and implications for future projects. The contractor shall integrate quality control into every phase of the project. VII.A.1. NCES Standards for Data Analysis All work conducted under this contract for NCES shall at a minimum meet the standards and guidelines set forth in the publication Standards and Policies, NCES Standard 87-01-02, March 16, 1987, as amended by standards implemented since that time. This publication is available on request from Dr. Susan Ahmed at 202/219-1781, or by mail: National Center for Education Statistics, 555 New Jersey Ave., NW, Washington DC 20208-5654. VII.A.2. NCES Standards for Publications All final reports for major products shall at a minimum meet the standards and guidelines set forth in the OERI Publication Guide. This publication is available on request from Dr. Susan Ahmed at 202/219-1781, or by mail: National Center for Education Statistics, 555 New Jersey Ave., NW, Washington DC 20208-5654. VII.A.3. Exemplary Publications Existing analysis reports illustrate the general format and use of standards as well as indicate the general level of complexity and depth of coverage. Recent reports that addresses postsecondary issues are available on NCES' home page (www.ed.gov/NCES/). Attachment PEDAR Statment of Work Technical Review Panelists The members of the Technical Review Panels that review NCES's longitudinal data collections are thoroughly familiar with the data sets that will be of primary use in this project. Moreover, they are familiar with many of the policy issues that will be the topics for the descriptive analysis reports. The names and addresses of current TRP members are listed below. Dr. Cliff Adelman U.S. Department of Education Office of Education Research (ORERI) Room 611A 555 New Jersey Avenue, NW Washington, DC 20208-5578 202/219-2251 CADELMAN@INET.ED.GOV Dr. Nabeel Alsalam National Center for Education Statistics Room 518, Capitol Place 555 New Jersey Avenue, NW Washington, DC 20208-5650 202/219-2252 NABEEL_ALSALAM@ED.GOV Dr. Rick Apling Congressional Research Service CRS/EPW Library of Congress Room LM 320 Washington, DC 20540-7440 202/707-5860 RAPLING@CRS.LOC.GOV Mr. David Bergeron U.S. Department of Education Office of Postsecondary Education Room 4060, ROB-3 7th & D Streets SW Washington, DC 20202 202/708-9069 DAVID_BERGERON@ED.GOV Ms. Sue Betka U.S. Department of Education Room 4251, FB-10B 600 Independence Avenue, SW Washington, DC 20202 202/401-3939 SUE_BETKA@ED.GOV Mr. Dan Chenok Human Resources & Housing Branch Office of Management and Budget Room 310235, New Executive Office Bldg. 725 17th Street, NW Washington, DC 20503 202/395-7316 DANIEL.J.CHENOK@EOP.SPRINT.COM Mr. Bob Davidson U.S. Department of Education Room 4039, FB-10A 600 Independence Avenue, SW Washington, DC 20202 202/401-0281 Dr. Daniel Davis U.S. Department of Education Room 4100, ROB-3 600 Independence Avenue, SW Washington, DC 20202 202/708-7867 Mr. Dan Goldenberg U.S. Department of Education Room 416, FB-10B 600 Independence Avenue, SW Washington, DC 20202 202/401-0182 DANIEL_GOLDENBERG@ED.GOV Dr. Mary Golladay National Science Foundation Room 965, 4201 Wilson Blvd. Arlington, VA 22230 703/306-1774 MGOLLADA@NSF.GOV Ms. Aromie Noe Office of Management and Budget Room 8236, NEOB 725 17th Street Washington, DC 20503 202/395-7773 Noe_A@A1.EOP.GOV Ms. Sharon Loper U.S. Department of Education Room 5624, ROB-3 7th and D Street, SW Washington, DC 20202-4651 202/708-8962 SHARON_LOPER@ED.GOV Mr. Dan Madzelan U.S. Department of Education Room 4060, ROB-3 7th & D Street, SW Washington, DC 20202 202/708-9069 DAN_MADZELAN@ED.GOV Dr. Jay Noell Congressional Budget Office Room 423 Ford 2nd & D Street, SW Washington, DC 20515 202/226-2672 JAYN.HRCD@CBO.GOV Ms. Constance Rhind Congressional Budget Office Room 423 Ford 2nd & D Street, SW Washington, DC 20515 202/226-2672 CONNIER.HRCD@CBO.GOV Ms. Blanca Rosa Rodriguez U.S. Department of Education Room 4004 EPML-21, ROB-3 7th & D Street, SW Washington, DC 20202 202/708-8963 Dr. Mark Schwartz U.S. Department of Education, Room 4325 FOB-10 600 Independence Avenue, SW Washington, DC 20202-7322 202/205-9422 MARK_SCHWARTZ@ED.GOV Ms. Patricia Smith Office of Management and Budget Room 8236, New Executive Office Bldg. 725 17th Street, NW Washington, DC 20503 202/395-5880 SMITH_P@A1.EOP.GOV Dr. Duc-Le To U.S. Department of Education Room 615B 555 New Jersey Avenue, NW Washington, DC 20208 202/219-2248 Dr. May K.-C. Chen Director of Research and Planning Los Angeles Community College District 770 Wilshire Blvd. Los Angeles, CA 90017 213/891-2052 CHENM@LACCD.CC.CA.US Mr. Robert T. Collins Director of Operations Technical Trades Institute, Inc. 1955 North Union Colorado Springs, CO 80909 719/632-8116 RCOLLINS@USA.NET Mr. John K. Curtice Director of Financial Aid Services State University of New York Central Administration, SUNY Plaza Albany, NY 12246 518/443-5474 Ms. Betty L. Davis CC of Allegheny County 1750 Clairton Blvd. West Mifflin, PA 15122 412/469-6241 BDAVIS@CCAC.EDU Dr. Thomas Flint Vice President Robert Morris College 180 North LaSalle Street Chicago, IL 60601 312/836-4888 Dr. Elinor Miller Greenberg EMG and Associates 6725 South Adams Way Littleton, CO 80122 303/771-3560 Dr. Janet C. Hunter Knox College Box K-138 Galesburg, IL 61401 309/343-0112 Dr. Ellen Leininger University of Missouri-Rolla Rolla, MO 65401 314/341-4954 ELLENL@SHUTTLE.CC.UMR.EDU Dr. Michael Nettles University of Michigan 2018 SEB 610 East University Avenue Ann Arbor, MI 48109-1259 313/764-9499 MICHAEL.NETTLES@UM.CC.UMICH.EDU Dr. Jim Palmer Illinois State University 328 DeGarmo Hall Normal, IL 61790-5900 309/438-2041 JCPALMER@R6000.CMP.ILSTU.EDU Dr. Caroline Hodges Persell New York University Room 408 269 Mercer Street New York, NY 10003 212/998-8350 PERSELL@ACFCLUSTER.NYU.EDU Ms. Harriet L. Rojas University of Idaho Moscow, ID 83844-4291 208/885-6312 HROJAS@IDUI1.CSRV.UIDAHO.EDU Dr. Karen Spahn Arizona Board of Regents 2020 North Central Avenue, Suite 230 Phoenix, AZ 85004 602/229-2525 IADKKS@ASU.VM.INRE.ASU.EDU Dr. Jacob Stampen University of Wisconsin, Madison 1186A Educational Sciences Bldg. 1025 W Johnson Street Madison, WI 53706-1796 608/263-4485 STAMPEN@MACCWISC.EDU Dr. Dawn Geronimo Terkla Tufts University 28 Sawyer Avenue Medford, MA 02155 617/627-3274 DTERKLA@PEARL.TUFTS.EDU Dr. Vincent Tinto Syracuse University 259 Huntington Hall Syracuse, NY 13244-2340 315/443-3343 Mr. Charles Treadwell NY State Higher Ed Service Corp. 99 Washington Avenue Albany, NY 12255 518/474-1549 Bitnet: RSH7200@NYSHESCV Ms. Robin Etter Zuniga WICHE P.O.Drawer P Boulder, CO 80301-9752 303/541-0224 Dr. Frank Balz NIICU 1025 Connecticut Avenue, NW Suite 700 Washington, DC 20036 202/785-8866 FRANK@NAICU.EDU Mr. Wayne Becraft AACRAO Suite 330 One Dupont Circle, NW Washington, DC 20036-1171 202/293-9161 BECRAFT@AACRAO.NCHE.EDU Ms. Deborah Carter American Council on Education One Dupont Circle Washington, DC 20036 202/939-9395 Mr. Timothy Christensen NASFAA 1920 L Street NW Suite 200 Washington, DC 20036-5020 202/785-0453 73033.502@COMPUSERVE.COM Dr. Jerry Davis Student Loan Marketing Association 1050 Thomas Jefferson Street, NW Washington, DC 20007 202/298-3911 Dr. Edward Elmendorf AAASCU Suite 700 One Dupont Circle, NW Washington, DC 20036-1192 202/857-1825 ELMENDORFE@AASCU.NCHE.EDU Dr. Jackie King American Council on Education Room 832 One Dupont Circle, NW Washington, DC 20036-1193 202/939-9354 Dr. Janet Hansen National Academy of Sciences HA-178, 2101 Constitution Avenue, NW Washington, DC 20418 202/334-3026 JHANSEN@NAS.EDU Ms. Kay King APSIA 2400 N Street, NW Washington, DC 20037 202/862-7989 APSIA@GWUVM.GWU.EDU Vacancy Research Associate The College Board Washington, DC 20003 202/332-7134 Mr. Jamie P. Merisotis Inst for Higher Ed Policy 1930 18th Street, NW, Suite 24 Washington, DC 20009 202/588-8383 PP000149@INTERRAMP.COM Mr. Kent Phillippe AACC One Dupont Circle, NW, Suite 410 Washington, DC 20036 202/728-0200 KPHILLIPPE@AACC.NCHE.EDU Mr. Peter Syverson Council of Graduate Schools Suite 430 One Dupont Circle, NW Washington, DC 20036 202/223-3791 CGS@@VTVM1.CC.VT.EDU Ms. Laura Greene Knapp 511 Carriage Lane Cary, NC 27511 919/460-0489 Data Analysis Input File Specifications NCES will create DATA ANALYSIS SYSTEMS (DAS) for projects upon receipt of five (5) files with the following specifications. The DAS will create tables and correlation matrices with Taylor standard errors (or replicate standard error approaches if necessary). To prevent disclosure of individual data, the DAS will compress the data. The DAS incorporates a variety of tamper-resistant, tamper-responding, and disclosure avoidance techniques to suppress decompression and data identification. The five files are: (1) DATA.POS Input Data (2) FMT.PRN Describes the format of data.pos (3) MASTER.TXT Detailed information on the variables (4) INPUT.TXT General system identification and inputs (5) README.TXT Readme file for users DAS systems always are designed for and operate in only one of five possible modes of calculation, based entirely on the value of CalcMode as given in the file INPUT.TXT. DATA.POS This is the data file. The data file must contain integers greater than or equal to -1. Absolutely no real values or decimal points are allowed. No values less than -1 are allowed. It includes categorical (contiguous values 1,2,3,4,..., and possibly zero) and continuous (zero and positive integer values) with uniform missing values of -1. See MASTER.TXT below for a description of when zeros are legitimate for categorical variables. For CalcMode=1 or 2, the variables needed for Taylor standard errors (primary sampling unit or PSU, Stratum, and Weight -- also positive integers) must be included on this file and the file should be sorted by Stratum and PSU. For CalcMode=3,4, or 5, the variables for the replicate weights must be included on this file. The maximum record length within DATA.POS is 255. The only formats that are acceptable are F2.0, F4.0, F6.0, and F8.0. Formats should be selected with the smallest fieldwidth possible. Examples of formats which are NOT acceptable include F3.0, F10.0, F12.0, and F3.0. All records must be positional fixed format (see FMT.PRN below) and of fixed length (max = 255). Finally, a check-sum variable (whose definition depends on the CalcMode and is explained under the discussion about INPUT.TXT below) must be included on this file. Weights should be rounded to integers. Observations with zero weights should be excluded. If the rounding of weights causes too much imprecision, mulitply the weights by factors of 10, 100, ... and include the factor under WtFactor in INPUT.TXT. Using a weight factor other than 1.0 reduces the speed of the final table generator and is severely discouraged. Before using a weight factor other than 1.0 please discuss alternatives with NCES. FMT.PRN This is a positional file that documents the format of DATA.POS. In column form, the following information must be included (one line for each variable in DATA.POS): Name Start Width Description varname 1 10 Variable name recnum 11 5 Record number in file startpos 16 5 Starting position for len 21 5 Number of positions used varlabel 26 70 Variable label Variable names should be alphanumeric and begin with an alpha. (In particular this means that underscores should NOT be used in variable names.) Numerics (recnum,startpos,len) should be right-justified. The character items (varname,varlabel) should be left-justified. The widths have been selected so that this file is easy to create using a spreadsheet with the default column width of five. However, a spreadsheet need not necessarily be used. Variable labels should be a composite of SECTION: LABEL, where LABEL is the label suitable for tables and SECTION: is a set of characters to group variables. {ONLY ONE BLANK SHOULD FOLLOW THE COLON:}. LABELS should be ready for tabling--no abbreviations, capitalization for first letter of first word and proper nouns ONLY, no single quotes or double quotes should appear, and all words must be spelled correctly! Try to eliminate NUMBER OF, AMOUNT OF, and RESPONDENT from all labels. Record numbers (Recnum) must be in the range 1,2,3,...,99. The number of positions used (Len) must be even. The only legitimate values for Len are 2, 4, 6, and 8. Only the variables which are listed on FMT.PRN are included in the final DAS, whether or not additional variables are included on the DATA.POS file. For QC, respondent tracking, or other purposes, other variables may be included on DATA.POS as long as they are NOT on the FMT.PRN file. For example, we strongly encourage that an ID variable be included in the DATA.POS file at the beginning of each observation. However, this ID variable should never be included in FMT.PRN. For CalcMode=2,3,4,5 WEIGHTS should have varlabels that begin with WEIGHT: . MASTER.TXT This is a text file that documents the data in DATA.POS. It is structured in 4 parts: 1. Label line Col 1 = \ Col 2-9 = VarName Col 11-80 = VarLabel (SECTION: Label) 2. Percentages & code labels block (Consists of an appropriate number of lines in the form:) Col 1-3 = code value Col 5-9 = weighted percentage Col 11-49 = code label Missing values (-1) are not labeled, but the -1 code and weighted percentage are included. Continuous variables have their code labels enclosed in {}. {Zero} and {min-max,mean/std} are used for continuous variables. If the first code label in a percentage & code labels block is enclosed in {}, then the variable is marked as continuous, otherwise the variable is assumed to be categorical. For categorical variables, second and later code labels may be enclosed in {}, in which case these code values are ignored for R, C, and B tags. In particular, zero is valid as a code value for categorical variables if and only if (1) the associated code label is enclosed in {} so that the zero category is ignored for R, C, and B tags, and (2) it is not the first code in the code label block. For categorical variables, code labels not enclosed within {} are assumed to be categories, and the code values must be consecutive integers starting with one (ie. 1,2,3,4,5,...). The lines with code labels enclosed within {} must all appear after the lines without {}. The lines without {} must be sorted ascending by the code values. Do NOT put PSU, STRATA, or CHECKDIGIT into MASTER.TXT. Do NOT put WEIGHT in MASTER.TXT when CalcMode=1. Do NOT under any circumstances place the actual weighted frequency for a category with an unweighted frequency of less than thirty (30) observations in MASTER.TXT. Categories such as these may exist in the data file but their frequencies must be set to zero in MASTER.TXT. 3. Divider line. Contains a \ in column 1. 4. Description block Contains multiple lines that document item wording, recoding, imputation, source, and analytic cautions. When sources are used, they should begin on a separate line with the introductory word "Source: ", and should be placed after any other text in the description block. The description block may refer to VarNames within this DAS, but not to electronic codebook or other system variables. Maximum line length is 70. INPUT.TXT The input.txt file organizes the input for various miscellaneous input items, such as the system title, names of directories, etc. This file is formatted by lines--our program expects certain input items to be in certain line numbers. The lines with # as the first character are comments. We usually just modify an existing copy, such as the one below which was taken from NPSAS:90 Graduate/First Professionals. Do NOT add additional comment lines or remove existing ones, since then the items will be on the wrong line numbers. # This is the main input file for setting up system # System Name NPSAS:90 Graduate/First Professionals # FPrefix = the prefix for this system, exactly three chars N0G # Name of Master.txt file is masterg.txt # Name of the map file Fmt.prn is n0gmap.prn # Name of the raw data file is n0gdata.raw # The directory name on the CD is n0g # the system name is (used for many purposes) n0g # variable name for weight pskeepwt # variable name for strata ofcon1 # variable name for PSU psu # variable name for check digit chkdgt # Number of cases on data file 52000 # password (used for developmental testing only) npsas90g # CalcMode for Calculation Mode (See Documentation) 1 # NumWts - Number of weights 1 # WtFactor (weights have been multiplied by this factor) 1.0 # WtList-If necessary, List of variable names for add'tl weights The most important feature of this file is CalcMode, which specifies what kind of system we are building. The interpretation of some of the other input values depends on the value for CalcMode. The possible values for CalcMode are: 1 One weight, Taylor method Handles case where there is only one weight. There are no additional weights listed under WtList, and the one weight is listed under "Variable name for weight". NumWts is ignored. The check digit variable on DATA.POS must be the sum of Stratum, PSU, and weight. 2 Multiple weights, Taylor method Handles case where more than one weight may be used for estimation. The user will be able to select one of several possible weights to be used for estimation. NumWts is the number of available weights. The variable names for the availabe weights must be listed one per line under WTList. The line labelled "Variable Name for weight" in the input file is ignored. However, for documentation purposes this weight should be included and should be the weight used for computation of the frequencies in MASTER.TXT. The weight which was used to compute the frequencies in the frequency windows should also be documented in the README.TXT file. The check digit variable on DATA.POS must be the sum of Stratum and PSU only. For all the following replicate methods, the weight used to produce estimates is listed after the line "Variable Name for Weight", the number of replicates is listed in NumWts, and the variable names for the replicate weights are listed one line per weight under WtList. The check digit variable on DATA.POS must be the sum of the replicate weights plus the estimate weight. The estimate weight used for the check digit variable is the same used to produce estimates, listed after the line "Variable Name for Weight". 3 BRR 4 Jacknife I 5 Jacknife II The WtFactor is the factor that all the weights on DATA.POS have been mutliplied by. In other words, estimates using the weights on DATA.POS would need to be divided by WtFactor. Before changing the WtFactor to something other than 1.0, see the discussion about weights under DATA.POS above. README.TXT This file is copied unchanged to the final DAS system. It is used to provide the user with general information, warnings about tampering with the DAS system, etc. The install program requires that it be available, and automatically displays it on the screen during the install process. The format of the file is ascii text, with lines of hyphens ("-") indicating pagination. The DAS install program uses the hyphens to determine when to pause, so lines of hyphens must appear in the file at the end of each screen of information. There is no software limit to the number of lines or screens of information, however please remember that you force the users to read the entire file during the install process (there is a timed pause for each screen of information). An example, taken from NPSAS:90 Graduates, follows. We usually edit this file to produce new readme.txt files. U.S. Department of Education Office of Educational Research and Improvement National Center for Education Statistics WELCOME You may install the Data Analysis System for the NATIONAL POSTSECONDARY STUDENT AID STUDY OF 1989-90: GRADUATE AND FIRST PROFESSIONAL STUDENTS This system is protected by Federal Law. Please carefully read the information screens that follows. The NPSAS:90 G1P Data Analysis System (DAS) will NOT display individual records. Tampering with the files in this system to attempt access to individual records is a violation of Federal Law and can result in a fine of $250,000 and imprisonment of up to 5 years. ----------------------------------------------------------------- By installing the NPSAS:90 G1P DAS you accept the legal conditions of use and agree to protect the DAS from misuse by others. Do NOT tamper with any of the DAS files. Do NOT share the files with other users. The Federal Law protecting the individual records embedded within the DAS are as follows: The Privacy Act of 1974 (as amended) [5.U.S.C.552a] Augustus F. Hawkins - Robert T. Stafford Elementary and Secondary School Improvement Amendment of 1988, Public Law 100-297 [20.U.S.C.1221e] General Education Provisions Act (GEPA) [Section 406d] ----------------------------------------------------------------- To obtain access to individual records, contact DATA SECURITY OFFICER STATISTICAL STANDARDS AND METHODOLOGY DIVISION NCES/OERI - ROOM 408 US DEPARTMENT OF EDUCATION 555 NEW JERSEY AVENUE, NW WASHINGTON DC 20208-5654 (202) 219-1831 ----------------------------------------------------------------- VIII. ADDITIONAL COMPLIANCE REQUIREMENTS (Section H) VIII.A. Legislation and Confidentiality of Data The Postsecondary Longitudinal Studies program (PLS) at NCES is authorized by law under the General Education Provisions Act (GEPA) section 406(b), as amended (20 USC, 1221e-1): "The purpose of the Center shall be to collect, analyze, and disseminate statistics and other data related to education in the United States and in other nations. The Center shall: collect, collate, and from time to time report full and complete statistics on the condition of education in the United States; conduct and publish reports on specialized analyses of the meaning and significance of such statistics; ..." The Center assures participating individuals and institutions that any data collected shall be in conformity with NCES's standards for protecting the privacy of individuals. Paragraph (4)(A) of section 406(d) of the GEPA as amended by Hawkins-Stafford states that: "The Center shall develop and enforce standards designed to protect the confidentiality of persons in the collection, reporting, and publication of data under this section ... Except as provided in this section, no person may -- (i) use any individually identifiable information furnished under the provisions of this section for any purpose other than statistical purposes for which it is supplied; (ii) make any publication whereby the data furnished by any particular person under this subsection can be identified; or (iii) permit anyone other that the individuals authorized by the Commissioner to examine the individual reports." This paragraph further requires the Commissioner to limit such access to those who have sworn to observe the limitations imposed by subparagraph (A) and makes them liable for fines and imprisonment for disclosure. This paragraph was further amended by Part E of the Excellence in Mathematics, Science, and Engineering Education Act of 1990 which excludes the National Postsecondary Student Aid Study (NPSAS) and all postsecondary longitudinal surveys from paragraph (4), but further restricted use such that: "(ii) Any person, except those sworn to observe the limitations of this subsection, who uses any data as described in clause (i) [NPSAS and postsecondary longitudinal studies] provided by the Center, in conjunction with any other information or technique (including de-encryption), to identify any individual student, teacher, administrator, or other person and who discloses, publishes, or uses for a purpose other than that for which it was collected, or who otherwise violates clause (i) or (ii) of subparagraph (A), shall be fined ... or imprisoned ... or both." Thus, while restrictions on the distribution of individual data are eased somewhat for these studies, all users who attempt to disclose are subject to penalty, especially those who have not so sworn but attempt to disclose.