The General Social Survey (GSS)The Next Decade and Beyond

The General Social Survey (GSS)

The Next Decade and Beyond

National Science Foundation

Workshop on Planning for the Future of the GSS

May 2-3, 2007

National Science Foundation

Arlington, Virginia

Participants and Attendees

Patricia White, National Science Foundation

Workshop Organizer

Frank Scioli, National Science Foundation

Workshop Moderator

Andrew Beveridge, City University of New York Queen’s College

Robert Bell, National Science Foundation

Suzanne Bianchi, University of Maryland

Norman Bradburn, National Opinion

Research Center

Linda Carlson, National Science Foundation

Mark Chaves, University of Arizona

Paul Ciccantell, National Science Foundation

James Davis, National Opinion Research Center

Cheryl Eavey, National Science Foundation

Barbara Entwistle, University of North Carolina - Chapel Hill

Jeremy Freese, University of Wisconsin

Kevin Gotham, National Science Foundation

Peter Granda, Inter-university Consortium for

Political and Social Research (ICPSR) University of Michigan

Edward Hackett, National Science Foundation

Michael Hout, University of California, Berkeley

Brian Humes, National Science Foundation

Ronald Inglehart, University of Michigan

Jon Krosnick, Stanford University

David Lightfoot

, National Science Foundation

John Logan, Brown University

Robert Mare, University of California, Los Angeles

Peter Marsden, Harvard University

Douglas Maynard, University of Wisconsin

Leslie McCall, Northwestern University

Daniel Newlon, National Science Foundation

Steven Nock, University of Virginia

Philip Paolino, National Science Foundation

Gregory Price, Jackson State University

Tom Smith, National Opinion Research Center

Lynn Smith-Lovin, Duke University*

Steven Ruggles, University of Minnesota*

Roberta Spalter-Roth, American Sociological Association

Frank Stafford, University of Michigan

Steven Tuch, George Washington University

* Submitted paper; did not attend meeting.

Executive Summary

The National Science Foundation (NSF) began supporting the General Social Survey in the early

1970’s and has continued to do so with a grant to the National Opinion Research Center (NORC)

in 2005 to complete the 2006 and 2008 surveys. The GSS, along with the Panel Study of

Income Dynamics and the American National Election Studies, is one of the three major

infrastructure projects supported by the Division of Social and Economic Sciences. The GSS is

an important survey data resource used by sociologists and other social scientists for research

and teaching.

On May 2-3, 2007, the Division of Social and Economic Sciences convened a workshop on The

General Social Survey (GSS): The Next Decade and Beyond at the National Science Foundation

(NSF) in Arlington, Virginia. The Sociology Program funded and organized the workshop. The

purpose of the workshop was to hold a conversation with the social science community about

how best to maintain and enhance the GSS as NSF moves forward with plans for a recompetition

for the conduct of the 2010 survey and beyond. Workshop participants were asked to assist the

NSF in completing an assessment of the GSS as a major NSF infrastructure investment and to

“assure the best use of NSF funds for supporting research and education.”

The workshop brought together twenty-one experts in survey research methodology and scholars

with intimate knowledge of the GSS. These participants prepared background papers and

discussed major areas of the GSS, including operations and governance, survey topics and

content, conceptual and methodological innovations, collaboration and integration with other

major national and international survey programs, outreach to the social science and other

scholarly communities, dissemination of data and results dissemination, and other broader

impacts. These background papers are included in the second section of this report. The

workshop discussions and papers provided advice and generated recommendations to NSF on

how to improve and strengthen the GSS as it moves into the next decade.

The report is organized in three major sections – Background of the GSS (and the Workshop); a

descriptive Overview of the GSS; and a list of workshop Recommendations. The first section

provides a brief discussion of the context from which the idea for the workshop emerged, the

purpose of the workshop, and its role in the 2010 GSS recompetition. The second gives an

overview (including a chronology) of the major components of the GSS. This overview covers

the purpose and development of the GSS, current GSS design, survey development, the GSS

core, recent innovations, cross-national research, experimentation and methodological research,

and data dissemination and usage. The final section lists workshop recommendations, organized

around three topics—(1) The Basic GSS, What to Maintain and Continue; (2) Governance,

Survey Administration and Funding; and (3) Data Collection, Dissemination and Outreach.

Additionally, the report Appendix includes the workshop agenda and papers submitted by the

workshop participants.

Summary of Recommendations

Workshop participants provided a number of insightful recommendations to NSF on how best to

improve and strengthen the GSS as it moves into the next decade.

The Basic GSS- What to Maintain and Continue

The GSS should remain a nationally representative survey of attitudes and intergroup relations

and continue to monitor trends in attitudes and behaviors. It should maintain and further

enhance its openness to the addition of new topics and questions by the social science

community.

Continue to administer the GSS on a biennial cycle. Since time is the critical independent

variable in administering the GSS, it should maintain the biennial schedule.

Develop plans to maintain the panel design. The panel component is a valuable addition to the

GSS and allows for the measurement of individual change over time on many variables,

particularly changes in attitudes, allowing the direct study of temporality and causation. The

panel study also allows GSS users to study issues that could not be studied with the earlier,

purely cross-sectional survey design.

Maintain GSS participation in the International Social Survey Program (ISSP) data collection

program. The ISSP is an important vehicle that can be used to study important social processes

in a comparative perspective by both examining differences across societies and changes within

the current 43 nations over time.

Continue a timely and comprehensive release of all data. Data from the GSS, both the core and

topical modules, should continue to be made available to all users at the same time and as soon

as possible after collection and data cleaning.

Governance, Survey Administration and Funding

Strengthen GSS governance and the role of the Board of Overseers. Significantly enhance and

strengthen the role of the GSS Board in evaluating the content of the survey, including the core

and topical modules.

Enable the Board of Overseers to participate more fully in the selection and development of GSS

topical modules. Topical modules provide a vehicle for innovation and should remain part of the

future GSS, but the Board of Overseers should have a stronger voice in evaluating topical

modules for scientific merit.

Provide support for the GSS Board of Overseers to hold a “module” competition. The quality of

the GSS would be further enhanced if funds were provided to the GSS Board of Overseers to

allow them to manage competitions to determine the addition of modules and questions to the

GSS. This competition would be open to social scientists who could propose topics and

questions to add to the GSS.

Explore ways to realize cost savings in survey administration. The GSS should explore

alternative modes of administration in order to reduce per-interview costs.

Seek innovative modes of administering the GSS. There is considerable value in maintaining the

face-to-face CAPI interview process. The future GSS, however, should examine incorporating

other modes of administering the survey, including a leave-behind questionnaire and internet

surveys by providing internet access and even laptop computers to allow those in the sample to

respond to a series of internet surveys over time.

Study the impact of the split-ballot technique on data analyses. The split-ballot administration

that gives each respondent two-thirds of the core may create significant problems for analysis.

The GSS should study the impact of this aspect of survey administration and, in particular,

explore ways to eliminate the split-ballot of the core items.

Share the funding of the GSS across other social science and education fields. The GSS is a

valuable survey data resource for disciplines other than sociology, with less than 50 percent of

GSS users identifying themselves as sociologists. The GSS is heavily used as a teaching and

research tool at both the undergraduate and graduate levels. NSF is encouraged to seek joint

funding of the GSS more broadly across the social sciences at NSF.

Data Collection, Dissemination and Outreach

Make the GSS core more transparent to the user community and allow for changes in core items.

The content of the core and how it has evolved over time should be made clearer to all potential

users of the GSS. The GSS should sketch out the core set of items, but the core should be

allowed to evolve over time via the interaction of the GSS Board, GSS PIs, and the user

community. Also, the possibility of enlarging the core, including bringing back time series

questions that were deleted in the early 1990s, should be explored.

Explore ways to enhance the data collected. The GSS should explore the possibility of

oversampling minority groups in order to allow comparisons between groups. It should also

consider collecting and making available to users much more “paradata/metadata,” including

interviewer characteristics, the mode of administering the interview, the number of calls made to

obtain the interview, characteristics of the home and neighborhood, respondent reaction times for

answering questions, and spatial identifiers. These data would be released subject to full

protection of respondent confidentiality, which in some cases will include special protections and

restricted access as all major data projects now do.

Experiment with digital-recording of interviews. Explore ways to digitally record interviews

using the computers that interviewers use to conduct the interview. Such recordings would

enable studies of “how” the interview is done; these studies would help improve data quality and

encourage the integration of qualitative and quantitative research.

Provide opportunities for experimentation. The GSS should encourage the embedding of both

methodological and substantive experiments in the survey.

Develop “targeted” GSS dissemination activities. The current NSF funding model does not

directly address the issues of data access and dissemination. Targeted NSF funding is needed for

the dissemination of data, including providing web access and increasing ease of access to the

data for research and teaching purposes.

Upgrade and create more user-friendly modes of dissemination. The current modes of

dissemination are confusing and badly outdated. The future GSS should include a state-of-the-

art dissemination system that is updated frequently, including a user-friendly website, a high

quality index, easy to use search tools, thorough documentation about survey procedures,

questions, and data files, simple data downloading in a variety of formats, accessible technical

assistance, and simple on-line analysis tools for users.

Expand the type of data products that are publicly available. In addition to disseminating the

complete dataset, the future GSS should explore making other data products available to users,

such as datasets of only core items, each topical module, and each minority group. Also, any data

linked to GSS data by users should become part of the GSS publicly available data, including the

computer code necessary to replicate analyses.

Secure a separate budget for data dissemination. Proposals for the future GSS should include an

extensive plan for data dissemination with an adequate budget, whether dissemination is to be

done in-house or via a third party. Past NSF grants have not included adequate funds for

outreach and dissemination.

Develop cooperative relationships with other major NSF surveys. Both the principal

investigators and the Boards of other NSF-funded surveys, in particular the Panel Study of

Income Dynamics and American National Election Studies, should work cooperatively to share

information and technologies. They should also consider greater coordination and integration,

such as common data access mechanisms and repositories, as well as using common measures

for demographic variables.

Table of Contents

Executive Summary…………………………………………………………4

An Overview of the General Social Survey………………………………10

Current GSS Design.........................................................................11

Survey Development.........................................................................11

The GSS Core................................................................................... 13

Recent Innovation in the Survey.....................................................13

GSS Panel Component.........................................................14

Spanish Language Translation............................................14

Cross-National Research..................................................................15

Experimentation & Follow-up Studies...........................................15

Data Dissemination and Usage........................................................15

Recommendations.........................................................................................16

The GSS- What to Maintain and Continue....................................17

Governance, Survey Administration and Funding........................17

Data Collection, Dissemination and Outreach...............................18

Appendices

Appendix 1: Workshop Agenda.....................................................20

Appendix 2. Workshop Papers.......................................................23

Andrew Beveridge, The General Social Survey and Its Impact in Sociology

and Other Social Sciences................................................................24

Suzanne Bianchi, The GSS, the ANES and the PSID & GSS Methodology

Comments Prepared for the NSF Workshop on Planning the Future

of the GSS..........................................................................................27

Norman Bradburn, Thoughts on the General Social Survey...................31

Mark Chaves, The General Social Survey: Innovation and Dissemination

Comments for NSF Workshop on the GSS....................................33

Barbara Entwisle, Hard Choices: Reflections on the Design of the

General Social Survey.......................................................................36

Jeremy Freese, Commentary for GSS Workshop, Methodology &

Technological Innovation and Cyberinfrastructure......................40

Peter Granda, Best Practices in the Dissemination of Survey Data.........43

Ronald Inglehart, The GSS and International Surveys: Issues and

Opportunities.....................................................................................47

Jon Krosnick, Thoughts on the GSS Recompetition..................................50

Robert Mare, Operational Aspects of the GSS from the Standpoint

of Board of Overseers........................................................................57

Douglas Maynard, Issues of Data Quality and Data Generation: The

General Social Survey and Ethnomethodology/Conversation

Analysis............................................................................................. 61

Leslie McCall, Review of the Content of the GSS..................................... 65

Steve Nock, Conceptual and Methodological Innovations & Contribution

of the GSS to Sociology and Its Broader Impacts......................... 67

Gregory Price, The General Social Survey: Contributions to Economics

And Recommendations for Future Dissemination........................ 70

Steven Ruggles, Review of Web-Based Dissemination of the General

Social Survey.....................................................................................73

Lynn Smith-Lovin, GSS Content and Innovations...................................77

Background

The National Science Foundation (NSF) began supporting the General Social Survey

(GSS) in the early 1970’s, funding the first survey in 1972. It has continued to do so with a

renewal grant to the National Opinion Research Center (NORC) in 2005 to complete the 2006

and 2008 surveys. The GSS, along with the Panel Study of Income Dynamics (PSID) and the

American National Election Studies (ANES), is one of the three major infrastructure projects

supported by the Division of Social and Economic Sciences. The GSS is funded and managed

by the Sociology Program, which allocates approximately twenty percent of its annual budget to

the support of the GSS. The GSS is a major survey data resource used by sociologists and other

social scientists for research and teaching, and in these roles adds tremendous value to the

conduct of basic social science research.

The National Science Board in a Resolution Concerning Competition, Recompetition

and Renewal of NSF Awards (NSB 97-224) affirmed strong support for the principle that

“expiring awards are to be recompeted unless it is judged to be in the best interest of U.S.

science and engineering not to do so.” The Sociology Program recognizes the important

scholarly and scientific achievements and accomplishment of the GSS, but also its challenges for

the future. To assess how to best maintain and enhance this important data resource, while also

positioning it to take advantage of innovations in survey research and information technology,

the Program convened a workshop—The General Social Survey: The Next Decade and Beyond

Workshop on Planning for the Future of the GSS—on May 2-3, 2007 to solicit advice from the

social science research community. The major focus of the workshop was to discuss

methodological and substantive challenges of the GSS in 2010 and beyond and to ensure the best

use of NSF funds for supporting social science research infrastructure.

The workshop brought together twenty-one experts in survey research methodology and

scholars with intimate knowledge of the GSS. These participants prepared background papers

and discussed major areas of the GSS, including operations and governance, survey topics and

content, conceptual and methodological innovations, collaboration and integration with other

major national and international survey programs, outreach to the social science and other

scholarly communities, dissemination of data and results dissemination, and other broader

impacts. These background papers are included in the second section of this report. The

workshop discussions and papers provided advice and generated recommendations to NSF on

how to improve and strengthen the GSS as it moves into the next decade.

An Overview of the General Social Survey

The General Social Survey (GSS) has provided a wealth of data on contemporary American

society for approximately 35 years by measuring social change and trends and constants in

attitudes, behaviors and attributes of the adult population. The GSS is a regular, ongoing

interview survey of U.S households conducted by the National Opinion Research Center. The

mission of the GSS is to make timely, high-quality, scientifically relevant data available to social

science researchers. The GSS is a personal interview survey and collects information on a wide

range of demographic characteristics of respondents and their parents; behavioral items such as

group membership and voting; personal psychological evaluations, including measures of

happiness, misanthropy, and life satisfaction; and attitudinal questions on such public issues as

abortion, crime and punishment, race relations, gender roles, and spending priorities. Since 1972

the GSS has conducted 26 in-person, cross-sectional surveys of the adult household population

of the U.S. Interviews have been conducted with a total of 51,020 respondents. The 1972-74

surveys used modified probability designs and the remaining surveys were completed using a

full-probability sample design, producing a high-quality, representative sample of the adult

population of the U.S. The GSS has a response rate of over 70 percent above that of other major

social science surveys and 40-45 percentage points higher than the industry average.

Current GSS Design

The basic GSS design is a repeated cross-sectional survey of a nationally representative sample

of non-institutionalized adults who speak either English or Spanish. Subsampling of non-

respondents is done to limit survey costs while maintaining a nationally representative sample.

Each GSS formally includes an A sample and a B sample. The preferred interview mode is in-

person interviews; however, a few interviews will be done by telephone in the event that an in-

person contact cannot be scheduled. Each respondent is asked the replicating core of socio-

demographic background items, along with replicated measurements of sociopolitical attitudes

and behaviors. Many of the latter are measured by way of a “ballot” design such that each item

is answered by a random 2/3 of each sample. Each GSS sample (A and B) includes an

International Social Survey Program module (ISSP). Each sample is also asked to respond to

several topical modules that may be supported by NSF or others, but are no longer supported by

the basic grant from NSF. Some of these topical modules, however, extend across both samples

in a given GSS survey.

This section summarizes materials provided by the GSS PIs in funded proposals, reports, and commentary

prepared for the workshop.

A Selective Chronology of the General Social Survey

1972 First GSS; subsequently conducted almost annually until 1993

Initial sampling design was block quota/modified probability

Many replicating core items measured on rotation design (2 years on and 1 year off)

Board of Advisors established; remained in existence until 1983

1975 Shift to full-probability sampling design

1977 Board of Methodological Advisors added; operated until 1983

1982 First African-American oversample

Bilateral cross-national collaboration with ALLBUS initiated

1983 Board of Overseers established

1985 Expansion of topical modules initiated

First International Social Survey Program module; topic was the role of government

1987 Second African-American oversample

1988 Split-ballot rotation system for replicating core items adopted

1991 First auxiliary study (National Organizations Study)

1994 Shift to two-sample design with target N of 3000

Major reduction in size of replicating core to accommodate more topical modules

Family mobility module and affiliated sibling study

1996 Segments of replicating core on race, gender, religion refreshed to alter series tracked

(until 2003)

1998 First National Congregations Study

2000 Clergy and Congregational Attendees studies

2002 Shift to computer-assisted personal interviewing (CAPI) methods

Second GSS-linked National Organizations Study

2004 Subsampling of non-respondents design adopted

2006 Target population expanded to include Spanish-speaking adults

Third sample added to accommodate overage of topical modules

Baseline wave of data collected for first GSS panel, spanning 2006-2010

National Voluntary Associations Study underway, linked to 2004 data

Second National Congregations Study now in progress

Survey Development

The GSS has six components, which include a replicating core, topical modules, cross-national

modules, experiments, re-interviews, and follow-up studies. The replicating core makes up one-

third of the GSS and the topical and cross-national modules the other two-thirds. Experiments

are conducted in both the core and supplemental modules. Re-interview and follow-up studies

are completed through additional data collections. The replicating core is the part of the GSS

that the Sociology Program has continually supported over the past 35 years. The “core”

consists of questions that regularly appear on the GSS. The contents of the core are periodically

reviewed by the grant principal investigators and Board of Overseers (consisting of a multi-

Prepared by Peter Marsden for the workshop.

disciplinary group of scholars with expertise in survey and other social science research

methodologies) to insure that the content remains relevant. Core questions are updated when

deemed necessary by the Board of Overseers. The core is about one-third demographic

information and two-thirds items that capture attitudes and behaviors. GSS content is wide

ranging with 5084 variables overall in the 1972-2006 file. The topical modules are used to

introduce new topics not previously investigated by the GSS and to cover existing topics in

greater detail with more fully specified models. The concept for a module may originate with the

PIs, the Board, or other scholars. Many prominent scholars help develop topical modules. For

example, 74 researchers representing many fields from 48 universities and research institutes

have served on the GSS Board and 253 social scientists in a dozen disciplines from 147

institutions have participated in the design of the first 28 topical modules.

The GSS Core

The GSS is a 90-minute in-person interview. Forty-five minutes of the GSS are devoted to the

core items, 15 minutes are devoted to the questions selected as part of the International Social

Survey Program (ISSP), and 30 minutes are allocated to topical modules that are funded by

sources other than the main NSF grant for the GSS. The socio-demographics in the core are

administered to all respondents and most attitudinal and behavioral measures are administered on

split-ballots with background items and a range of replicated measures of sociopolitical attitudes

and behaviors. The core questions are administered as a split-ballot, with each respondent

answering two-thirds of the core questions. Topical modules are focused on a wide range of

substantive issues and are funded by a variety of government agencies, private foundations, and

other organizations. These funds supplement the main NSF grant funding and allow the GSS to

fully fund the 90 minute survey and collect a larger sample than would be possible with only

NSF funds.

The GSS has a “replicating core” that emphasizes collection of data on social trends through

exact replication of question wording over time. Core items fall into two major categories—

socio-demographic/background measures, and replicated measurements on social and political

attitudes and behaviors. Many of the latter items appear on three GSS “ballots”, each of which is

administered to a random two-thirds of most GSS samples. Over the course of the project, there

have been many changes in replicated items. In addition, there are “quasi-core” items that are

repeatedly funded by other sources. In addition to the planned trend items included in the GSS

core, there are other data series that arise through unplanned repetition of topical modules and

the replication of ISSP items.

Recent Innovations

The 2006 GSS has two major ongoing innovations. First, it serves as the baseline sample

for the new GSS panel component, with a sub-sample of cases to be re-interviewed in 2008 and

2010. Second, the GSS core was translated into Spanish and administered in either English or

Spanish as needed in 2006. This practice will continue in 2008 and 2010.

GSS Panel Component

The GSS now includes a panel study component for the first time in order to allow the direct

observation of change over time in the same individuals. The GSS switched from a repeating,

cross-section design to a combined repeating cross-section and panel-component design. The

2006 GSS was the base year for the first panel. A sub-sample of 2006 GSS cases (most likely

about 2000) will be selected for reinterview in 2008 and again in 2010 as part of the GSS in

those years (see Table 1). The 2008 GSS will consist of a new cross-section of about 2000 plus

the 2006 reinterviews. The 2010 GSS will consist of another new cross-section of about 2000,

the second reinterview wave of the 2006 panel cases and the first reinterview wave of the 2008

panel cases. The 2010 GSS will be the first one to fully implement the new, combined design. In

2012 and later General Social Surveys, there will likewise be a fresh cross-section, wave two

panel cases from the immediately preceding GSS, and wave three panel cases from the next

earlier GSS.

Rotating, Three-Wave Panel/Cross Section Design

Samples

2006

2008

2010

2012

New Cross Section

3000 2000 2000 2000

Initial Reinterview Target 2000 2000 2000 2000

Expected Wave 2 Completed Interviews N/A 1500 1500 1500

Expected Wave 3 Completed interviews N/A N/A 1200 1200

Total Sample 3000 3500 4700 4700

Spanish Language Translation

In 2006 the GSS added Spanish to its standard, English-language version. This translation

allowed the GSS to expand its target population to adults living in U.S. households able to be

interviewed in either English or Spanish. The addition of the Spanish language interviews

“notably increased the number and proportion of Hispanics in the GSS.” In addition, the

composition of the Hispanic population changed in several notable ways. The adding of Spanish-

language interviews shows that Hispanics are notably less assimilated than indicated in the

previous English-only samples and also differ on several other demographics. This variation

across demographics is often, but not always, linked to the differences in level of assimilation

across the language-use/ability groups. The analysis of non-demographics further indicates that

Hispanics often significantly differ across language-use/ability groups.

Further analysis of the differences across language-use/ability groups focusing on the English

and Spanish bilinguals identifies a few items on which language effects may be occurring. These

will be explored further by building Spanish-wording experiments into the 2008 GSS.

Cross-National Research

The GSS first spurred cross-national research by inspiring other nations to develop data

collection programs modeled on the GSS, including Allgemeinen Bevolkrungsumfragen der

Socialwissenschaften (ALLBUS) in Germany, the British Social Attitudes Survey, the National

Social Science Survey in Australia, the Polish GSS, and the Japanese GSS. Second, it joined

with these and other programs to form the International Social Survey Program (ISSP), a

collaborative program of comparative survey research. The fundamental goal of ISSP is to study

important social processes in a comparative perspective by both examining differences across

societies and changes within countries over time. Since 1984, ISSP has grown to 43 nations,

which includes the founding four-- US, Germany, Britain, and Australia-- plus Austria, Belgium,

Brazil, Bulgaria, Canada, Chile, China, Croatia, Cyprus, the Czech Republic, Denmark, the

Dominican Republic, Finland, France, Hungary, Ireland, Israel, Italy, Japan, Latvia, Mexico, the

Netherlands, New Zealand, Norway, the Philippines, Poland, Portugal, Russia, Slovakia,

Slovenia, South Africa, South Korea, Spain, Sweden, Switzerland, Taiwan, Turkey, Uruguay,

and Venezuela. Data from ISSP modules on the role of government, social networks and support

systems, social equality, the family, work orientation, religion, the environment, national

identity, and citizenship are available from various national archives and the Inter-university

Consortium for Political and Social Research (ICPSR) in the United States.

Experimentation and Re-Interview, Follow-up and Methodological Studies

Since its inception, experiments have been included as part of the GSS and the GSS has

completed numerous re-interviews and follow-up studies. Experiments are an integral part of the

GSS program of methodological research, and dozens of studies have been completed as part of

the replicating core and topical modules to examine most aspects of survey methods. Re-

interview studies have included both methodological and substantive studies and the GSS has

also served as the source for follow-up studies of employers, voluntary associations, religious

organizations/leaders, and family mobility. In a continuing effort to improve data, the GSS

conducts methodological research on topics such as survey error, sensitive topics, sample-frame

comparability, third-person effects, contextual effects, the measurement of race and ethnicity,

item non-response, cross-national comparisons, and network measurements.

Data Dissemination and Usage

Since 1972 the Roper Center for Public Research has been the original point of deposit for GSS

Data. Upon release, data users may immediately secure data on a CD from the Roper Center.

Data, however, are also simultaneously released to ICPSR and the University of California at

Berkeley Survey Documentation and Analysis (SDA) Archive where, after additional

preparation, data are made publicly available for download and analysis. GSS data are also

distributed by over half a dozen archives around the world. In 2007 the GSS received support

from NSF to update their website. The funds are being used to create a GSS data dissemination

portal which will allow researchers and the general public to browse the GSS 1972-2006

cumulative data file, run basic statistical tabulations on-line within their web browser, and select

variables of interest for download into a selected statistical package format for additional

analysis.

GSS data are made available to a broad-based user community. Currently, there are over 150

different versions of GSS/ISSP. Documentation comes in five major forms-- the GSS

Cumulative codebook, a SPSS system file, the GSS Report Series, the GSS Data and Information

Retrieval System (GSSDIRS), and, for the ISSP, electronic and hard-copy codebooks in English,

data files, and a CD-ROM with codebooks and files, plus copies of original language

questionnaires. The GSSDIRS website seems popular with users, having approximately

4,000,000 visits annually. The ISSP website is visited approximately 200,000 times a year. The

user community includes researchers, college teachers, university students, business planners,

media and public officials. Academic scholars who use the GSS as a data source include

sociologists, political scientists, economists, statisticians and survey methodologists,

anthropologists, geographers, biologists, engineers, psychologists, criminologists and legal

scholars, medical/health researchers, and business administration and management scholars.

The GSS is a well used research tool and there are now 14,000 documented uses, but the detailed

categorization has not been undated since 2003. In 2003, the PIs were able to document 8,662

uses of the GSS: 4,862 journal articles, 1,664 books, 1,364 scholarly papers, 568 reports, and 188

dissertations and theses. Most users (82%) were academics with college affiliations. Usage has

grown over the years, increasing from 200 per annum in the late 1980s to over 600 per annum in

2003. With the exception of the Census and its Current Population Survey, the GSS is the most

frequently used data set in the three leading sociology journals. The GSS has been used about as

often as the total of the next six most frequently used data sets combined.

Recommendations

The workshop participants strongly emphasized that the GSS had made many invaluable

contributions to social science, to U.S. policy debates, and to public understanding of the

characteristics of U.S. society. GSS time series and cross-sectional data provide the foundation

for social science understanding of a wide range of issues, allowing for the tracking of changes

in attitudes regarding race and ethnic relations, religious beliefs and practices, and family life.

The GSS occupies a unique niche as an effective social science omnibus survey, which it has

occupied for about 35 years. Its future strengths lie in being able to continue to operate as a very

high quality survey of the behavior and attitudes of a representative sample of the adult U.S.

population. The GSS is the “gold standard” for survey research in the U.S. and globally. Thus,

the current basic structure of the GSS provides a strong foundation for moving into the future.

Workshop participants recommended that the basic structure of the GSS should remain the same.

However, they also offered recommendations that focus on how to move forward and enhance

the GSS for the future.