SDA: Survey Documentation and AnalysisNews
June 2024
SDA 4.1.5 has been released. This update includes a major upgrade to the usage report capability of SDAMANAGER. Usage reports can now be limited to a range of dates using a popup date picker. Also, asterisks can now be used as wildcards when filtering on dataset ID's or client addresses. These new options, and other improvements, make the useage report feature much more useful. This update also includes bug fixes, version upgrades to libraries, etc.News From Previous Dates
November 2023
The 1972-2022 General Social Survey (GSS) Cumulative Dataset (Release 2) has been added to the SDA Archive. See the title page of the online SDA HTML codebok for links to more information on this version.October 2023
The 2022 Survey of Consumer Finances (SCF) has been added to the SDA Archive. See the Abstract for more information on this version.August 2023
SDA 4.1.4 has been released. This update includes several enhancements: the C++ programs are now 64-bit executables (instead of 32-bit), MySQL 8 is now supported, plus several other enhancements, version upgrades to libraries, bug fixes, etc.March 2023
Release 3b of the 1972-2021 General Social Survey (GSS) Cumulative Dataset has been added to the SDA Archive. See the title page of the online SDA HTML codebook for links to more information on this version.December 2021
The 1972-2021 General Social Survey (GSS) Cumulative Dataset has been added to the SDA Archive. See the title page of the online SDA HTML codebook for links to more information on this version.July 2021
The 2020 American National Election Study (ANES) Full Release has been added to the SDA Archive. This is the first full release of the dataset and supersedes the two previous preliminary releases (including the preliminary release added to the SDA Archive in June). See the Abstract for more information on this version.June 2021
The 2020 American National Election Study (ANES) Preliminary Release has been added to the SDA Archive. SDA 4.1.3 has been released. This new version of SDA includes enhancements to XCODEBK's HTML output. SDA codebooks were improved stylistically, made easier to navigate and made fully compliant with the HTML 5 standard. The HTML output of the various analysis programs was also modified for HTML 5 compliance. However, the changes do not affect the appearance of the analysis output in any significant way.September 2020
The 2019 Survey of Consumer Finances (SCF) has been added to the SDA Archive. See the Abstract for more information on this version.August 2020
SDA 4.1.2 has been released. This new version of SDA includes an option in TABLES and MEANS to output statistical results in CSV format for importing into other applications (such as Excel) for creating custom charts. CSV files can also be useful for preparing tables for inclusion in manuscripts. All datasets on the SDA Archive page that use SDA 4 have now been updated to use 4.1.2.March 2020
SDA 4.1.1 has been released. This new version of SDA can handle non-ASCII character variables in the raw input data file. The data file must be encoded in UTF-8, but character variables are no longer limited to the US-ASCII subset of characters. See the Internationalization document in the online SDA 4.1 Manual for more informaiton on using non-English languages in SDA.February 2020
The 2018 and 2016 Cooperative Congressional Election Studies (CCES) have been added to the SDA Archive. See the introductions in the online HTML codebooks for more information on these datasets.October 2019
SDA 4.1 has been released. This update enhanced the SDA Manager web app in a number of ways:- SPSS .sav files, Stata .dta files, CSV files and TSV files can be imported and automatically converted into SDA datasets.
- The user interface was reorganized to enhance usability.
- The online help was extended and improved.
- A CASEID variable is no longer required in an SDA dataset or in SDA metadata (DDL).
April 2019
The 1948-2016 American National Election Study Cumulative Dataset has been added to the SDA Archive. See the introduction in the online HTML codebook for more information on this version.March 2019
The 1972-2018 General Social Survey (GSS) Cumulative Dataset has been added to the SDA Archive. See the introduction in the online HTML codebook for more information on this version.June 2018
All datasets on the SDA Archive page that use SDA 4.0 have now been updated to use 4.0.3. This new version of SDA will be released to ACS members in the near future. Enhancements include:- Percentiles Added to MEANS
An option was added to calculate and display a specified percentile (from the 1st to the 99th percentile) of the dependent variable in each cell. - Bar Charts Added to XCODEBK
When a frequency distribution is displayed for a variable, each category is now displayed with a horizontal bar to show the relative size of the percent of cases falling into each category. - Charts Added to REGRESS
A chart showing the values of the regression coefficients and their confidence intervals can be displayed. You can choose either the 'B' (unstandardized coefficient) or the 'Beta' (standardized coefficient) to display. - Charts Added to LOGIT
A chart showing the values of the regression coefficients and their confidence intervals can be displayed. You can choose either 'B', or 'Exp(B)' (if Logit regression is specified), or (if 'Probability differencess' are specified) 'P-Diff 1 unit' or 'P-Diff 1 SD'.
September 2017
The 2016 Survey of Consumer Finances (SCF) has been added to the SDA Archive.June 2017
All datasets on the SDA Archive page that use SDA 4.0 have now been updated to use 4.0.2. This new version of SDA will be released to ACS members in the near future. Enhancements include:- Drag and drop from variable tree to input text boxes
You can drag and drop a variable from the variable tree to an input field on a form. (Note that although the variables in the tree display both the variable's name and the label, only the variable name will be "dropped" into the input field.) - Right-click on variable tree to view variable description
You can right-click on a variable in the tree to bring up a context menu where you can click on "View variable" to view a description of the variable's contents. - Bar charts added to "View variable" output
When a frequency distribution is displayed (for variables with 256 or fewer categories) each category is now displayed with a horizontal bar to show the relative size of the percent of cases falling into each category.
May 2017
All datasets on the SDA Archive page that use SDA 4.0 have now been updated to use 4.0.1. This new version of SDA will be released to ACS members in the near future. Enhancements include:- Bookmarkable URLs for Analysis Output
The URL for a page of analysis results can now be saved and re-run. This allows a user to bookmark the output for later reference, email the URL to someone else to view the results, or copy and store the URL in a document. (In technical terms, the analysis request is now a "get" instead of a "post".) - SDAWEB Accessibility Enhancements
SDAWEB has various accessibility enhancements including a new accessibility mode for Section 508 compliance. There is a checkbox at the top of the option screen to enter that mode.
April 2017
The 1972-2016 General Social Survey (GSS) Cumulative Dataset has been added to the SDA Archive. See the introduction in the online HTML codebook for more information on this version. The 2016 American National Election Study Dataset has been added to the SDA Archive. See the introduction in the online HTML codebook for more information on this version.October 2015
SDA 4.0 has been released. SDA 4.0 is a major update. The user interface has been substantially modified, providing a more modern look with rich, dynamic components. Also, a new SDA Manager web application is part of the SDA distribution package. The SDA Manager consolidates all of the management functions for an SDA archive into one web interface. The SDA Manager webapp also supports the configuration of personal user workspaces. These user workspaces enable analysts to create and store recoded and computed variables in their own private storage areas. The SDA Manager also allows an archive manager to easily configure dataset-level access control -- specifying which users can access which datasets.April 2015
The 1972-2014 General Social Survey (GSS) Cumulative Dataset has been added to the SDA Archive. See the introduction in the online HTML codebook for more information on this version.May 2014
Video tutorials about SDA are now available on the SDA YouTube Channel. Want to learn about using the SDA 4.0 user interface? Try the new SDA YouTube Channel.April 2014
SDA 4.0 interface now available for datasets in the SDA Archive. Most datasets on our SDA Archive page are now available with the new SDA 4.0 -- in addition to the older SDA 3.5.August 2013
The 2012 American National Election Study Dataset has been added to the SDA Archive. See the introduction in the online HTML codebook for more information on this version.March 2013
The 1972-2012 General Social Survey (GSS) Cumulative Dataset has been added to the SDA Archive. See the introduction in the online HTML codebook for more information on this version.October 2012
The SDA Web site was down for several days due to multiple hardware failures. For more information on the problems during October with the SDA Web site, please see the Server Update Page.July 2012
The Survey of Consumer Finances (Extract File 1989-2010) has been added to the SDA Archive. Data from the SCF are widely used, from analysis at the Federal Reserve and other branches of government to scholarly work at the major economic research centers.March 2011
The 1972-2010 General Social Survey (GSS) Cumulative Dataset has been added to the SDA Archive. See the introduction in the online HTML codebook for more information on this version.February 2011
SDA 3.5 has been released. Note that version 3.5 now calculates appropriate standard errors for regression coefficients for complex sample designs (like GSS). This requires extra computation time, especially for logit and probit regression. You can still use the previous simple random sample computations as an option.January 2011
All the studies on the Archive page are now accessed using SDA 3.5. This new version of SDA will be released to ACS members in February. A new book has been published that uses SDA for data analysis exercises. Conducting Empirical Analysis: Public Opinion in Action, by Rosalee A. Clawson and Zoe M. Oxley, is a public opinion workbook that includes a variety of data analysis exercises that use SDA.May 2010
The Public Policy Institute of California (PPIC) is using SDA's Quick Tables software to power its new Statewide Survey Interactive Tools.January 2010
SDA 3.4 has been released. Versions for Linux (on Intel x86 and compatible platforms), Windows and Solaris (on the Sparc platform) are now available for ACS members. All the studies on the Archive page are now accessed using SDA 3.4. Question text has been added to the 2008 American National Election Study Dataset in the SDA Archive.November 2009
The SDA Web Site has had a stylistic update. But the organization of the site and its content remains largely the same.September 2009
A preliminary version of the 2008 American National Election Study Dataset has been added to the SDA Archive. See the introduction in the online HTML codebook for more information on this version.June 2009
SDA 3.3 has been released. This new version of SDA includes enhancements to customized subsetting, including the output of CSV format files (for import into a spreadsheet) and the inclusion of user-created variables in subset datasets. Disclosure rules that protect confidentiality can also be specified in this version. For a complete list of the new features see the "Updates" section in the online SDA 3.3 Manual. Versions for Linux (on Intel x86 and compatible platforms), Windows and Solaris (on the Sparc platform) are now available for ACS members.May 2009
All the studies on the Archive page are now accessed using SDA 3.3. It will be released in mid June.March 2009
A preliminary version of the 1972-2008 General Social Surveys Cumulative Dataset has been added to the SDA Archive. See the introduction in the online HTML codebook for more information on this version.October 2008
The SDA Web site has moved to a new server. The new server is a Dell PowerEdge 2950 III (2 Quad Core Xeon E5440 processors, 8 GB RAM) running Red Hat Enterprise Linux 5. (SDA's previous server was an old Solaris/Sparc machine that is being retired by the Berkeley campus.)May 2008
SDA 3.2 has been released. Searching is the major new feature of this release. (You can try it out on our Archive page.) Versions for Windows, Linux (on Intel x86 and compatible platforms) and Solaris (on the Sparc platform) are now available for ACS members.March 2008
All the studies on the Archive page are now accessed using SDA 3.2. This new version of SDA provides a search function and will be released soon. Quick Tables for the GSS 1972-2006 Cumulative dataset have been added to the SDA Archive page.December 2007
New user contributions added to the Community page: ICPSR contributed an SPSS-Python script that converts an SPSS system file or portable file into the files required to set up the dataset in SDA. The University of Victoria and University of Toronto contributed an SDA Tutorial.November 2007
The 1998 Dutch Prejudice Survey has been added to the SDA Archive.August 2007
The 1948-2004 American National Election Study Cumulative Dataset has been added to the SDA Archive. The study includes a search utility that is currently under development.July 2007
The 1972-2006 General Social Surveys Cumulative Dataset has been added to the SDA Archive. The 2006 GSS includes a search utility that is currently under development.April 2007
SDA 3.1 has been released. Versions for Solaris (on the Sparc platform), Windows and Linux (on Intel x86 and compatible platforms) are now available for ACS members. Also, we have switched to using SDA 3.1 for all the studies listed on our Archive page.December 2006
"Investigating Community and Social Capital", an ICPSR instructional module, has been added to the Community page. This site "teaches data analysis of social capital, as discussed in Robert D. Putnam's Bowling Alone." The exercises use SDA.September 2006
SDA 3.0 has been released. Versions for Solaris (on the Sparc platform), Windows and Linux (on Intel x86 and compatible platforms) are now available for ACS members.August 2006
The SDA Archive page is now using SDA 3.0 for all studies. We plan to release this version to ACS members in September.May 2006
A test version of the new SDA interface is now available in our archive. The next version of SDA will provide users with a new interface. It includes program selection, variable selection and program options all on one screen. To try it out, go to the Archive page and select the first GSS 2004 link. Comments are welcome. We expect this version to be released to ACS members in the fall.January 2006
SDA 2.1 has been released and is now being used for our SDA Archive. Versions for Solaris (on the Sparc platform), Windows and Linux (on Intel x86 and compatible platforms) are now available for ACS members.October 2005
The "Other Archives" page has been changed to the "Community" page. The "Community" page now includes various SDA resources: books that use SDA for exercises, contributions from SDA users, and the previous listing of other archives that use SDA.September 2005
The 1972-2004 General Social Surveys Cumulative Dataset has been added to the SDA Archive. The dataset for the 2004 American National Election Study (ANES) has been added to the SDA archive. This dataset has been set up using a new test version of SDA -- version 2.0. This new version allows users to run tables using character variables. Since the ANES 2004 has a number of character variables (like the name of the state where the interview was carried out), this dataset is a good one to use to test out those new features. As always, let us know (at sda@csm.berkeley.edu) if you encounter any problems with the new version. SETUPS (Supplementary Empirical Teaching Units in Political Science) has been added to the "Other Archives" Page. The SETUPS site features Voting Behavior: The 2004 Election, an instructional module based on the 2004 American National Election Study (ANES) data. Additional data files have been added to the SDA Archive. Census Data: The American Community Surveys for 2000-2003 have been added to the SDA Archive. The American Community Surveys are designed to replace the long form of the Census. The 1990 U.S. 1% PUMS dataset has also been added to the archive. Election Studies: The Cumulative ANES 1948-2000 has been added.March 2005
The 1986 Bay Area Race and Politics Survey has been added to the SDA Archive Although this dataset is relatively old, it is of interest as the main dataset used for the book The Scar of Race by Paul M. Sniderman and Thomas Piazza. Funding recently became available to mark up the dataset and expand the documentation so that it could be put up in SDA.January 2005
SDA 1.4 has been released. Versions for Solaris (on the Sparc platform), Windows and Linux (on Intel x86 and compatible platforms) are available. Although a number of enhancements are included in the 1.4 release, the major addition is the ability to produce charts with the crosstabulation output.November 2004
SDA 1.3 for Linux on Intel x86 and compatible platforms has been released. (Although we had previously developed a test port of SDA 1.2 to Linux, this is the first official Linux release of SDA.)September 2004
The SDA crosstabulation program in our SDA Archive now generates charts automatically. Users have a choice of charts and can also specify the size of the chart and the orientation (horizontal or vertical) for bar charts. This chart-generation capability is still under development and not yet ready for release. We hope to include charts within other SDA programs in the future.August 2004
The 2000 California 5% PUMS Dataset has been added to the SDA Archive. This dataset is a 5% sample of all households and persons in California, taken from the 2000 Census. The dataset has one record for each person in the sample and the household variables have been appended to each person's record. The dataset contains a sample of 1.7 million California residents. The smallest identifiable geographic area contains 100,000 persons. SDA's Quick Tables application now automatically generates charts. You can try it out with the GSS Cumulative Datafile on the SDA Archive page.April 2004
The 2000 U.S. 1% PUMS Dataset has been added to the SDA Archive. This dataset is a 1% sample of all households and persons in the United States, taken from the 2000 Census. The dataset has one record for each person in the sample and the household variables have been appended to each person's record. SDA 1.3 Released SDA 1.3 is now available for ACS members. This version includes enhancements to several programs. For a summary of all the updates since 1.2b, see the "updates" section of the online SDA 1.3 Manual (in the "Documentation" section).February 2004
The 'Listcase' program is now available for the datasets on the SDA Archive page. Select 'List values of individual cases' to access the program.January 2004
The SDA Archive page is now using the forthcoming version 1.3. SDA 1.3 will be made available to ACS members in the near future.December 2003
The 2000 California 1% PUMS Dataset has been added to the SDA Archive. This dataset is a 1% sample of all households and persons in California, taken from the 2000 Census. The dataset has one record for each person in the sample and the household variables have been appended to each person's record.September 2003
The 1972-2002 General Social Surveys Cumulative Dataset has been added to the SDA Archive.June 2003
Two new features were added to the GSS Cumulative Datafile in the SDA Archive. A link to a set of "Quick Tables" -- a simplified way of obtaining analysis results -- has been added to the archive page. As time permits, more Quick Tables will be added. (For more on the "Quick Tables" project see the "Projects" page.) Also, an alternate HTML codebook version -- with variables stratified by year -- has been made available to help users determine if specific questions were asked in a certain year or range of years.'Crosstabulation' Program UpdatedThe statistic in the crosstabulation program that determines the color code for each cell in a table has been revised. The statistic now used for that purpose is called a "standardized residual." Documentation for the revised residuals (now referred to as a Z-statistic) is available online by clicking on the Z-statistic option on the crosstabs option page.
April 2003
The new Cultural Policy and the Arts National Data Archive (CPANDA) has been added to the listing of 'Some Other Archives that Use SDA'.August 2002
The California Workforce Survey 2001-2002 and the Chicago African American Survey 1997 have been added to the SDA Archive. The comparison of means program now includes an option for producing a multiple classification analysis (MCA).December 2001
Recode and Compute are now available for some studies. The SDA site now provides an option to run the RECODE and COMPUTE programs on some of the datasets (GSS 2000, ANES 2000, and National Race and Politics). These programs allow you to create new variables derived from existing variables in each dataset. The new variables are placed into a public work area, which is available to all users of the SDA Website.August 2001
The 2000 American National Election Study (ANES) has been added to the SDA archive.July 2001
The 1972-2000 General Social Survey Cumulative Dataset has been added to the SDA archive.March 2001
The 1972-1998 General Social Survey Cumulative Dataset has been added to the SDA archive.September 2000
The American Political Science Association (APSA) presented SDA the 2000 "Best Instructional Software Award" at their annual meeting in Washington D.C. The 1998-1999 Multi-Investigator Study has been added to the archive. The study was a national random-digit telephone survey funded by the National Science Foundation. It involved the active participation of thirteen teams of investigators and was focused on public opinion related to a wide variety of social and political issues.July 2000
The 2000 Innovators Award was recently presented by the American Association for Public Opinion Research (AAPOR) for SDA and the ICPSR Web site for the General Social Survey (GSS).March 2000
SDA 1.2 has been released. This version includes enhancements to several programs. For example, the XCODEBK program now uses HTML tables and frames to both improve the appearance and ease the navigation of HTML codebooks. Also, the output of various analysis programs has been enhanced with color-coding of the cells in the tables. (These improvements can be viewed in the SDA Archive at this site.) In addition, SDA 1.2 includes new programs: correl (correlation matrix), regress (multiple regression), and listcase (list contents of selected variables for selected cases). For a summary of all the updates since 1.1, see the "updates" section of the online SDA 1.2 Manual (in the "Documentation" section).October 1999
The on-line GSS and ANES codebooks in the SDA Archive have been updated using a beta version of the forthcoming 1.2 XCODEBK program. The improvements include a persistent table-of-contents frame for easier navigation and enhanced HTML formatting of various aspects of the codebook.September 1999
Some analysis programs have been updated to include the color-coding of table cells. The cells are color coded in order to aid in detecting patterns. See the help for each program for details on how the color-coding works for that program.April 1999
Variable name specifications are now case-insensitive for most studies. Most of the problems users have reported with using the SDA Archive turned out to be problems with the case of variable names. For example, many GSS analysts were accustomed to seeing the GSS variable names in upper case and therefore would specify the variable "life", for example, as "LIFE" -- which would produce a "variable not found" message. In response, we've added an option to the archive configuration ("HARC") file so that user-specified variable names will automatically be converted to upper or lower case as appropriate for the study. Now, for most studies, case is not an issue when specifying variable names. The SDA 1.1 documentation -- available from this site's "Documentation" section -- has been updated to reflect the features in the final, release version. The "Projects" section has been updated with information about current work on enhanced HTML codebooks. There's also new information about our collaboration with UC Berkeley's Survey Research Center on a project funded through ICPSR's Data Documentation Initiative (DDI). We've added additional links to some other sites using SDA software. See the "Other Archives" section.January 1999
Version 1.1 of the SDA software has been installed on our Web site. The major changes are as follows:- Button for extra codebook window: the main option screen for a study allows you to open a separate window for the codebook. Then you can switch easily between the codebook window and the window to run analysis programs. You can even use the regular "copy-and-paste" procedures to copy variable names from the codebook window to the option screen where you specify the names of variables. This is particularly useful for studies that have long variable names.
- Weight variables are pre-specified for most studies. You can select the desired weight (or no weight) from a drop-down list.
- In the crosstabulation program the horizontal variable is now optional. If you specify only the vertical variable, and you request statistics, the program will generate univariate statistics. The calculation of the chi-square statistics has been modified to avoid excessive inflation of the statistics due to the use of population weights. The weighted chi-square is now scaled down by the factor: (Unweighted N) / (Weighted N).
- Comparison of Means program: standard errors can be requested that take into account complex sample designs (such as for stratified cluster samples). The availability of this option depends on our having sufficient information to specify the complete sample design. Eventually we hope to have this information for all studies in the data archive.
- Instrument Documentation: Procedures to document computer-assisted interviewing instruments have been released and are now available.
November 1998
The first release version of the IDOC programs and documentation has been completed. IDOC is a joint project of the Census Bureau and CSM to develop network-browsable documentation for CAI instruments written in the CASES Q language.February 1998
The SDA Web Site has had an aesthetic make-over. The content remains largely the same but we hope the new look will improve navigation between pages and make it easier for our users to locate specific topics.January 1998
All of the online datasets available for analysis have now been converted to the new SDA (CSA version 3) format. The last two datasets converted were the ANES cumulative file for 1952-92 and the California 1% Microdata Sample from the 1990 Census. Standard errors project: A project is underway to add the capability of calculating standard errors for complex samples to the SDA comparison of means program. Other data archives are beginning to use SDA. Our home page now provides links to some of those archives that may be of general interest.October 1997
Most of the online datasets available for analysis have been converted to the new SDA (CSA version 3) format. New features available with SDA programs:- New "Comparison of Correlations" program
- Temporary recoding of variables
- Output from the analysis programs is in HTML instead of basic ASCII.
Note that this new HTML output uses the tables mechanism. Old character-based browsers (like LYNX) will not format that output into the appropriate columns. However, even version 1 of Netscape will handle the output correctly.
- 1996 American National Election Study
- Health studies from Brazil (in Portuguese)
July 1997
Downloading and subsetting are now standard. The procedures to allow downloading and subsetting of data files and documentation are now included in the main archive as options for many of the datasets. Previously, these options were in the section for "Projects Under Development." Additional Version 3 (SDA) programs: work is continuing on the new version of the CSA programs for the Web. Since these programs are designed specifically for the Web, and have a different interface and output format than the older CSA programs, we intend to give this new set of programs a new name: SDA -- Survey Documentation and Analysis. Previously, the only program available to test was 'tables'. Additional programs and test datasets will now begin to appear.March 1997
New location: you will have noticed that the CSA Demonstration Archive has moved to a new computer. New format for demonstration archive: the one-stage screen for selecting both a study and an action has been replaced by a two-stage procedure. First you select a study (perhaps after viewing study abstracts), and then you select the action (browse codebook, run a table, etc.). This two-stage procedure is closer to what an actual data archive will need to do. The one-stage procedure is still available as one of the "Other Versions of Data Archive." Projects under development: A new version of the procedures to create customized subsets of data files and documentation has been installed. You can try it out on the CSA test dataset. In addition, if you want to try it out on larger datasets, including the large GSS file, you can access the procedure by selecting "Projects Under Development.".September 1996
New and revised datasets:- 1991 National Health Interview Survey (part)
- 1994 National Multi-Investigator Survey
- Tables with HTML output
- Subset data files for downloading
March 1996
New and revised datasets:- General Social Survey - Cumulative File 1972-94
This dataset is now online, ready for analysis. An HTML codebook is also online. - California Census Public Use Sample - 1 percent
Several introductory and appendex files have been added. Individual items have more complete descriptive text. A revised HTML codebook is online.
January 1996
The following documents were added for access online:- New documents about how to set up datasets on the Web for browsing and analysis
- The CSA User's Guide, Chapters 1-5. (The documents on individual programs and topics, included in the Manual as Chapter 6, were already available online.)
- A 1994 survey carried out in Italy on regional and ethnic prejudice in that country; the variable labels and other textual information are in English.
- The same Italian survey, but with labels and other textual information in Italian. It is instructive to see what does and does not remain in English in the codebook with the current CSA software.