Data sets used in Survey Analysis Workshop

[Page last updated 10 October 2011]

Raw data sets are exactly as they would arrive from the fieldwork agency.

SPSS saved files for British Social Attiudes and Quality of Life in Britain are initial files only.  Variables are named using the positional naming convention  There are no mnemonic names and no derived variables, but these will be generated during the exercises.

Pre-course survey of interests and experience

                Questionnaire  on pre-course interests and experience.
                Transfer sheet for data from pre-course questionnaire
                myclass.txt            Cumulative raw data set -  (6 kb, fixed-width Courier 12)

                                                                          SPSS Syntax files   >>             SPSS saved files

                1st step: reading raw data -              myclass1.sps      >>           myclass1.sav (5 kb)
                2nd step: adding variable labels-       myclass2.sps      >>           myclass2.sav (6 kb)
                3rd step: adding value labels -           myclass3.sps      >>           myclass3.sav (6 kb)

British Social Attitudes

                              
     1986          3100 cases, 23 records per case, 798 variables
                      Raw data file       bsa86.txt         (3.4 mb)  
                        SPSS saved file     bsa86.sav       (2.0 mb)

The saved file for the 1986 wave has variables renamed from the original mnemonic names to my preferred positional variable names (see: 1.3.1 Conventions for naming variables in SPSS) but it still has the original variable and value labels (written by John Curtice et al in 1986).   These are all in UPPER CASE and quite cumbersome (limit respectively of 40 and 16 characters in those days).  It's best to build up your own file with mixed case labels, as in the tutorials and exercises (limits on characters almost totally relaxed in later versions of SPSS).  You should try the exercises yourself before looking at these.  They are here as a crib (or as backup) not for you to copy and run, otherwise there's no point in you taking the course.  See entries 2.2.1.2 to 2.1.2.5
                
                                                     SPSS Syntax files   >>   SPSS saved files
Nominal and ordinal variables
 1: Reading raw data -                                 mybsa86_1.sps      >>      mybsa86_1.sav  
 2:  Adding missing values and labels-        mybsa86_2.sps      >>      mybsa86_2.sav  

Interval scale variables
 1: Reading raw data -                                 mybsa86_3.sps      >>      mybsa86_3.sav  
 2:  Adding missing values and labels-        mybsa86_4.sps      >>      mybsa86_4.sav  
_______________________________________________________________________

       1989               3024 cases, 23 records per case, 980 variables
                               Raw data file         bsa89.txt            (3.8 mb)           
                               SPSS saved file    BSA89 UK       (3.2 mb)

The saved file for the 1989 wave has variables renamed from the original mnemonic names to my preferred positional names (see: 1.3.1 Conventions for naming variables in SPSS).  Variable labels are in mixed case, but the question numbers remain at the end of the labels (where you can't see them!) as per original file written by John Curtice et al in 1989 (limit of 40 characters in those days).  The value labels are also mixed case, but are otherwise untouched (limit of 16 characters in those days).  The variable measurement levels have not been accurately specified (SPSS defaults keep changing when reading numeric variables) and some missing values have not been declared, but it's easier to check and/or amend these when you use the file. 

Again. it's best to build up your own file(s) with mixed case labels, as in the tutorials and exercises (limits on characters almost totally relaxed in later versions of SPSS).  The following files repeat the 1986 exercises on the 1989 data.  You should try the exercises yourself before looking at these.  They are here as a crib (or as backup) not for you to copy and run, otherwise there's no point in you taking the course.  See entries 2.2.1.6 to 2.2.1.8.

Nominal, ordinal and interval variables. 
                                                     
                                                               SPSS Syntax files   >>   SPSS saved files 
                                                                                                                                                   
 1: Reading raw data into SPSS                      mybsa89_1.sps     >>      mybsa89_1.sav  
 2:  Adding missing values and labels     mybsa89_2.sps     >>      mybsa89_2.sav  
 3:  Merging two files                               mybsa89_3.sps      >>       mybsa89_3.sav
 4:  Renaming variables                                mybsa89_4.sps        >>       mybsa89_4.sav

Quality of Life in Britain

   1975    Questionnaire [pdf, 27pp, 3.4 mb]    
    SPSS saved file  - QOL1975.sav (0.6 mb)

Other documents and data sets from the QoL surveys are not used in tutorials, but can be seen from the main menu on Subjective Social Indicators >> SSRC Survey Unit Quality of Life in Britain surveys

Playground to Politics

Questionnaire [pdf, 13pp, 4.63 mb]
Raw data file fifthdat.txt   [0.05 mb]
SPSS saved file fifthx.sav [0.06 mb]

The user manual  containing frequency counts for all variables, basic coding
information and a facsimile of the original questionnaire is on another site because it is more than 5mb

Other data sets

Once all the main tutorials are up and running, I hope to include later waves of British Social Attitudes, European Social Survey and other major surveys in the public domain.