For those learning (or teaching beginners) to analyse survey data with SPSS, I’ve just added some new draft tutorials to my page 2.3 Data transformations:
http://surveyresearch.weebly.com/23-data-transformations.html
They use data from the British Social Attitudes survey to address the research question, “What is the distribution of respondents' personal gross income from paid work? What shape does the distribution have? What is the distribution for women only? What is the distribution for men only? Are there any differences. What other variables might account for differences in income?”
The data are from the 1989 wave, but the same variables are repeated in all waves, so at a later stage there will be an exercise in merging all the files and investigating changes (if any) from 1983 to 2012.
They’re the last ones dealing with the analysis of single variables via frequencies and conditional frequencies as a transition to analysing two or more variables via contingency tables. For now analyses are limited to charts and to percentages in univariate tabulations. As a transition to bivariate tabulation, SELECT IF is used to produce conditional frequencies and charts. (It’s too early in the course to use SPLIT FILE: students will forget to turn it off!). As yet there are no descriptive statistics or statistical tests: these will come later once the logic of partitioning a zero order statistic into 1st and 2nd order statistics is clear (elaboration). Later tutorials will use the same logic for partitioning means.
Incidentally I experimented with an output table of sex * income group by copying it to Word, deleting the % signs in the body of the table, copying the % rows for men and women into Excel, calculating percentage differences (epsilon) and then importing them back into the (very wide) table and changing the colours to red for negative and blue for positive differences, but the bar-charts are much more effective at demonstrating income differences between men and women.
Once students have got a firm grasp of basic SPSS mechanics and of simple descriptive statistics, they can move on to significance testing, but without ever losing sight of the underlying substantive social research or social policy question.
The tutorials are syntax-based, but a couple of examples use the GUI to produce the same amendments to the data dictionary.
[Brand new homework beta versions added 22 June 2013]
2.3.1.5 Conditional frequencies homework [BSA89] http://www.weebly.com/uploads/2/9/9/8/2998485/2.3.1.5_conditional_frequencies_homework_bsa89.pdf
[Specimen answers for 2.3.1.5]
2.3.1.6.1 Specimen answer for conditional frequencies homework [Tasks 1 and 2]
http://surveyresearch.weebly.com/uploads/2/9/9/8/2998485/2.3.1.6.1__specimen_answer_for_conditional_frequencies_homework_tasks_1_and_2.pdf
2.3.6.1.2 Specimen answer for conditional frequencies homework [Tasks 3 and 4]
http://surveyresearch.weebly.com/uploads/2/9/9/8/2998485/2.3.1.6.1__specimen_answer_for_conditional_frequencies_homework_tasks_1_and_2.pdf
John F Hall (Mr)
[Retired academic survey researcher]
Email: [hidden email]
Website: www.surveyresearch.weebly.com
SPSS start page: www.surveyresearch.weebly.com/spss-without-tears.html
Locked
