- Welcome
- Important notice
- About the author
- About this site
- Site guide + Search box
- Dedications
- Acknowledgments
- My personal pantheon (of the great and the good in survey research)
- Recent and planned activities
- Textbooks for Research Methods and Data Analysis
- 1: Survey Analysis Workshop (SPSS)
- 1a: Statistical concepts and methods
- 1b: Teaching with Survey Data
- 1c: Developing research projects using survey data
- 1d: Workshop and presentations for ASSESS (SPSS users in Europe)
- 2: Survey Research Practice
- 2a: Survey Research Methodology, Practice and Training
- 2b: Major survey series
- 3: Subjective Social Indicators (Quality of Life)
- 4: Survey Unit, Social Science Research Council (UK)
- 5a: Polytechnic of North London (1976-1992)
- 5b: Survey Research Unit (1978-1992)
- Village life in Normandy
- Contact
- Origins of the British Crime Survey
- British Crime Survey
SPSS files and documentation used for tutorials and exercises
[Page still under construction: last updated 22 August 2014]
Accessing SPSS files
Getting SPSS to access files directly from websites can be complicated, so you will need to download some files to your own computer first, but it won't be necessary to keep everything once you have extracted what you need. On my machine SPSS saved files, once downloaded, can be opened immediately with SPSS. All unrestricted data sets used in my tutorials can be accessed from data sets and documents. Some facsimile questionnaires are included, but if not, links to extracts are provided in tutorials as and when necessary.
I hope eventually to use more recent data from the British Social Attitudes survey, the European Social Survey and other major data sets in the public domain. The original data remain copyright of the depositors and are not available from this site, although I may be able to negotiate making small subsets available for tutorial purposes. However, the Roper Center has given me permission to use small extracts of original data from the General Social Survey (GSS) but I'm having trouble locating a decent text version of the questionnaire to work from: the only one I can find seems to be a BLAISE program for CAPI.
Meanwhile I have downloaded the SPSS saved files for the following surveys from:
Roper Center:
General Social Survey (NORC, 2008 wave) .
UK Data Service (UKDS, based at Essex University):
British Social Attitudes 2011 and
Understanding Society (Wave 2, 2010: previously the British Household Panel Survey)
I have a number of reservations about the immediate usability of these files for secondary analysis or for teaching, particularly for new users.
For the General Social Survey file, my main dislike is aesthetic: the use of UPPER CASE text throughout for most variable and value labels.
For British Social Attitudes and Understanding Society my concerns are more serious. These include confusing, inefficient and sometimes ungrammatical labelling of variables and values, incorrect (or non-) specification of measurement levels and inconsistent (or non-) declaration of missing values, of which there are sometimes eight when you can only specify three in SPSS.
If you wish to access the full original data for any or all of these, you will have to register with the appropriate archive. You can then use my notes to see what the files look like and, if appropriate, run some SPSS syntax files to modify the data yourself.
Whilst data sets and documentation for British Social Attitudes and Understanding Society are available from UK Data Services, and may be used in future tutorials, checking and modifying them, then writing tutorials and obtaining necessary permissions to upload corresponding extracts from the data, is daunting and exhausting. UKDS is currently investigating ways of making small subsets available with unrestricted access and without the need for registration.
To make the distributed SPSS saved files easier to understand and use I am preparing extended introductions to, and commentaries on, the distributed SPSS files for each survey as follows:
British Social Attitudes
Commentary on SPSS file for British Social Attitudes 2011
Notes on British Social Attitudes 2004 teaching data set (as used by Marsh and Elliott, 2008)
Understanding Society
Commentary on Understanding Society 2010
ONS Opinions Survey: Well-being module (April 2011)
Unrestricted Access Teaching Dataset (ONS Opinions Survey, Well‐Being Module)
(Data set and user guide from Cathie Marsh Centre for Census and Survey Research, Manchester)
NORC General Social Survey (GSS)
Commentary on full NORC General Social Survey 2008
Commentary on subset of General Social Survey 2008 (as used by Sweet & Grace-Martin)
Commentary on GSS 2008 SPSS files for Babbie et al (as used by Babbie, Halley, Wagner & Zaino)
I am extremely grateful to Jon Peck (Senior Software Engineer, IBM/SPSS) who has patiently provided version after version of Python code to modify variable and value labels in the distributed SPSS files for some or all of the above surveys.
The modified SPSS files will not be available on this site unless permission is given by the original depositors, but the Python code and SPSS setup files to modify the original SPSS files will be posted later. These will convert labels which are all UPPER CASE or all lower case to Mixed Case, ensure that each label starts with an upper case letter and, where appropriate, move question numbers from the end to the beginning of the variable labels.. Other SPSS syntax files will correctly specify measurement levels, recode positive missing values to negative in a consistent manner thus enabling them to be specified using a range of values (value1 THRU value2).and fully specify values to be treated as missing.
Also of interest is a document containing some comments I addressed to the European Social Survey team in 2004 about the SPSS file for the 2002 wave: this includes a reply from Caroline Roberts.
I hope eventually to use more recent data from the British Social Attitudes survey, the European Social Survey and other major data sets in the public domain. The original data remain copyright of the depositors and are not available from this site, although I may be able to negotiate making small subsets available for tutorial purposes. However, the Roper Center has given me permission to use small extracts of original data from the General Social Survey (GSS) but I'm having trouble locating a decent text version of the questionnaire to work from: the only one I can find seems to be a BLAISE program for CAPI.
Meanwhile I have downloaded the SPSS saved files for the following surveys from:
Roper Center:
General Social Survey (NORC, 2008 wave) .
UK Data Service (UKDS, based at Essex University):
British Social Attitudes 2011 and
Understanding Society (Wave 2, 2010: previously the British Household Panel Survey)
I have a number of reservations about the immediate usability of these files for secondary analysis or for teaching, particularly for new users.
For the General Social Survey file, my main dislike is aesthetic: the use of UPPER CASE text throughout for most variable and value labels.
For British Social Attitudes and Understanding Society my concerns are more serious. These include confusing, inefficient and sometimes ungrammatical labelling of variables and values, incorrect (or non-) specification of measurement levels and inconsistent (or non-) declaration of missing values, of which there are sometimes eight when you can only specify three in SPSS.
If you wish to access the full original data for any or all of these, you will have to register with the appropriate archive. You can then use my notes to see what the files look like and, if appropriate, run some SPSS syntax files to modify the data yourself.
Whilst data sets and documentation for British Social Attitudes and Understanding Society are available from UK Data Services, and may be used in future tutorials, checking and modifying them, then writing tutorials and obtaining necessary permissions to upload corresponding extracts from the data, is daunting and exhausting. UKDS is currently investigating ways of making small subsets available with unrestricted access and without the need for registration.
To make the distributed SPSS saved files easier to understand and use I am preparing extended introductions to, and commentaries on, the distributed SPSS files for each survey as follows:
British Social Attitudes
Commentary on SPSS file for British Social Attitudes 2011
Notes on British Social Attitudes 2004 teaching data set (as used by Marsh and Elliott, 2008)
Understanding Society
Commentary on Understanding Society 2010
ONS Opinions Survey: Well-being module (April 2011)
Unrestricted Access Teaching Dataset (ONS Opinions Survey, Well‐Being Module)
(Data set and user guide from Cathie Marsh Centre for Census and Survey Research, Manchester)
NORC General Social Survey (GSS)
Commentary on full NORC General Social Survey 2008
Commentary on subset of General Social Survey 2008 (as used by Sweet & Grace-Martin)
Commentary on GSS 2008 SPSS files for Babbie et al (as used by Babbie, Halley, Wagner & Zaino)
I am extremely grateful to Jon Peck (Senior Software Engineer, IBM/SPSS) who has patiently provided version after version of Python code to modify variable and value labels in the distributed SPSS files for some or all of the above surveys.
The modified SPSS files will not be available on this site unless permission is given by the original depositors, but the Python code and SPSS setup files to modify the original SPSS files will be posted later. These will convert labels which are all UPPER CASE or all lower case to Mixed Case, ensure that each label starts with an upper case letter and, where appropriate, move question numbers from the end to the beginning of the variable labels.. Other SPSS syntax files will correctly specify measurement levels, recode positive missing values to negative in a consistent manner thus enabling them to be specified using a range of values (value1 THRU value2).and fully specify values to be treated as missing.
Also of interest is a document containing some comments I addressed to the European Social Survey team in 2004 about the SPSS file for the 2002 wave: this includes a reply from Caroline Roberts.
SPSS files with unrestricted access on this site
Pre-course survey of skills and interests (Block 1)
Students on the original Survey Analysis Workshop completed a short questionnaire at the beginning of the course. The questionnaire included questions on previous experience of computing and statistics and a selection of questions replicated from the British Social Attitudes survey: it also asked for principal mode of travel to the campus, gender, age and height. It was part of an exercise in which they had to copy data to a transfer sheet and then type the data into a computer file, prior to using SPSS.for processing and analysis. Students with little or no previous experience of computing or statistics found this procedure very helpful in understanding where data come from and how they get into the computer. There is a cumulative data set for all students over several waves and a related SPSS saved file. These files form the basis for the first part of this course. If you like you can complete the questionnaire yourself and add your own data.
British Social Attitudes
The tutorials use the same data from the 1986 and 1989 waves of the British Social Attitudes survey series that were used on my original course. The raw data are used to help you to learn how build up your own files: you will not be using the entire data set. There are also SPSS saved files for both 1986 and 1989: these have the full original variable set, but no derived variables.
I am working on more recent data from 2011 (UKDS SN 7237), but there's a lot of preparation required before the published SPSS file can be made easily usable. In any case, the original data remain copyright of the depositors and will not be available from this site, although I may be able to negotiate making small subsets available for tutorial purposes.
I am working on more recent data from 2011 (UKDS SN 7237), but there's a lot of preparation required before the published SPSS file can be made easily usable. In any case, the original data remain copyright of the depositors and will not be available from this site, although I may be able to negotiate making small subsets available for tutorial purposes.
Quality of Life in Britain
Some of the new tutorials use examples from the series of surveys conducted by Mark Abrams and myself at the SSRC Survey Unit between 1971 and 1975 (in collaboration with Angus Campbell, ISR, Michigan) to develop subjective social indicators to measure the quality of life.. They may be 40 years old, but they contain interesting and innovative questions designed as benchmarks to test the feasibilty of asking a general population sample to describe and assess their living conditions and satisfactions across a range of life domains, together with measures of psychological well-being. For now the only file we shall be using is the SPSS saved file for the 1975 nation-wide survey.
Playground to Politics
A 1981 study of values and attitudes from a Fifth Form survey (15 and 16-year olds) in a North London comprehensive school.. The data are used in tutorials demonstrating the generation of derived variables using RECODE, COUNT and COMPUTE