- 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
3.3 Multiple response
[Page last updated 16 June 2020]
Explanation of multiple response questions, examples from real surveys, and exercises in analysis using SPSS command MULT RESPONSE
3.3.1 Analysing multiple response with SPSS - an introduction
Many questions permit more than one answer.(eg tick as many as apply). This document explains different practices in coding such questions and how the data can be analysed using the SPSS command MULT RESPONSE
3.3.2a First exercise in multiple response] [pdf, SPSS 15]
3.3.2b First exercise in multiple response [SPSS 24] [in preparation]
Worked example from the pre-course self-completion questionnaire analysing the multiple response question on previous training in, or experience of ,typing, word-processing, social statistics or other computing. Follows on from previous exercises in Block 1 sections 1.3 and 1.4 (checking and coding data to a transfer sheet, typing data into a raw data file, reading the data into SPSS and completing the data dictionary. The cumulative data set is contained in the SPSS saved file class.sav ).
3.3.3 Multiple response questions in the 1986 British Social Attitudes survey
Note giving examples of multiple response questions in the 1986 British Social Attitudes survey and the different ways in which they have been coded and entered in the raw data. See also the sample data to get an idea of the content and layout (5 cases extracted).
Beta versions [pdf] of the following fully worked through tutorials are now available.
3.3.3.1 Analysing multiple response exercise 1 - One field per code
Deals with a question on qualifications obtained by education and/or training. There is a list of 15 qualifications (coded from 01 to 15) each of which has been allocated its own fixed 2-column field: the tutorial demonstrates how to read in the raw data, specify a group variable for qualifications obtained and produce a grouped frequency table.
3.3.3.2 Analysing multiple response exercise 2 - More values than fields
Deals with a question on whether the respondent is prejudiced against people of other races and, if so, which races. There are 13 possible code values, but only 3 fields allocated.
3.3.3.3 Analysing multiple response exercise 3 - More replies than values [New tutorial added 5 March 2016]
Using a question about welfare benefits received, this exercise demonstrates how to handle a question in which the same code values 1 - 6 are repeated within a set of 12 responses to the same question (but in different fields). This involves recoding the second set of values, but needs great care whn dealing with missing values.
3.3.3.4 Analysing multiple response 4 - Dichotomous mode [New tutorial added 20 Feb 2018]
Sets of questions with only two response categories (e.g. Yes/No) can be analysed as a block by treating them as multiple response questions. Indeed, any variable can be dichotomised and thus treated. The example asks whether the respondent (as a member of a trade union or staff association) has ever done any of a list of six actions (coded 1="Yes" or 2="No") and the exercise demonstrates how to define a dichotomous group variable comprising all the "Yes" answers and then tabulate it.
Back to Block 3 menu
Explanation of multiple response questions, examples from real surveys, and exercises in analysis using SPSS command MULT RESPONSE
3.3.1 Analysing multiple response with SPSS - an introduction
Many questions permit more than one answer.(eg tick as many as apply). This document explains different practices in coding such questions and how the data can be analysed using the SPSS command MULT RESPONSE
3.3.2a First exercise in multiple response] [pdf, SPSS 15]
3.3.2b First exercise in multiple response [SPSS 24] [in preparation]
Worked example from the pre-course self-completion questionnaire analysing the multiple response question on previous training in, or experience of ,typing, word-processing, social statistics or other computing. Follows on from previous exercises in Block 1 sections 1.3 and 1.4 (checking and coding data to a transfer sheet, typing data into a raw data file, reading the data into SPSS and completing the data dictionary. The cumulative data set is contained in the SPSS saved file class.sav ).
3.3.3 Multiple response questions in the 1986 British Social Attitudes survey
Note giving examples of multiple response questions in the 1986 British Social Attitudes survey and the different ways in which they have been coded and entered in the raw data. See also the sample data to get an idea of the content and layout (5 cases extracted).
Beta versions [pdf] of the following fully worked through tutorials are now available.
3.3.3.1 Analysing multiple response exercise 1 - One field per code
Deals with a question on qualifications obtained by education and/or training. There is a list of 15 qualifications (coded from 01 to 15) each of which has been allocated its own fixed 2-column field: the tutorial demonstrates how to read in the raw data, specify a group variable for qualifications obtained and produce a grouped frequency table.
3.3.3.2 Analysing multiple response exercise 2 - More values than fields
Deals with a question on whether the respondent is prejudiced against people of other races and, if so, which races. There are 13 possible code values, but only 3 fields allocated.
3.3.3.3 Analysing multiple response exercise 3 - More replies than values [New tutorial added 5 March 2016]
Using a question about welfare benefits received, this exercise demonstrates how to handle a question in which the same code values 1 - 6 are repeated within a set of 12 responses to the same question (but in different fields). This involves recoding the second set of values, but needs great care whn dealing with missing values.
3.3.3.4 Analysing multiple response 4 - Dichotomous mode [New tutorial added 20 Feb 2018]
Sets of questions with only two response categories (e.g. Yes/No) can be analysed as a block by treating them as multiple response questions. Indeed, any variable can be dichotomised and thus treated. The example asks whether the respondent (as a member of a trade union or staff association) has ever done any of a list of six actions (coded 1="Yes" or 2="No") and the exercise demonstrates how to define a dichotomous group variable comprising all the "Yes" answers and then tabulate it.
Back to Block 3 menu