SPSS for beginners, Part 3
2 likes2,110 views
This document discusses useful SPSS tools for analyzing survey data, including entering different types of survey questions as variables, computing new variables, recoding variables, selecting cases, and analyzing groups separately. It demonstrates how to enter single-response, multiple-response, ranked-response, and scaled-response survey questions as variables in SPSS. It also shows how to use the "Split File" function to generate separate frequency outputs for groups to analyze the relationship between variables split by a grouping variable like gender.
SPSS for beginners, Part 3
- 1. Part 3: Dr. Christine Pereira Survey data; Useful SPSS tools
- 2. » Entering data from a survey » Compute a total score and mean score » Recode variables » Select cases to analyse Dr. Christine Pereira » Analyse separate groups of data ASK at Brunel (2013) 2
- 3. Dr. Christine Pereira What are the variables? ASK at Brunel (2013) 3
- 4. Select ‘ONE’ Response Questions What is your main mode of transport to Uni during the week? What is your age? (please tick one) 18-25 26-35 36-45 46-55 56+ (please tick one) Car Bus Tube Rail Walk Cycle Other: ____________ Dr. Christine Pereira Yes No Do you try to minimise your carbon footprint? (please tick one) What is your gender? (please circle one) ASK at Brunel (2013) Male Female Single response survey questions. Each question is a new variable. 4
- 5. Select ONE response Gender Age Group Mode CO2 1 Male 36-45 Car No 2 Male 26-35 Tube Yes 3 Female 46-55 Car No 4 Dr. Christine Pereira Participant Male 18-25 Tube Yes 5 Female 26-35 Rail Yes ASK at Brunel (2013) Single response survey questions. Each question is a new variable. 5
- 6. Multiple Response Questions Which mode(s) of transport do you usually take to get to Uni during the week? Dr. Christine Pereira (please tick all that apply) Car Bus Tube Rail Walk Cycle Other: ____________ ASK at Brunel (2013) ‘tick all…’ instead of ‘tick one’ 6
- 7. Multiple response Transport 1 Car 2 Bus, Tube 3 Car 4 Dr. Christine Pereira Participant Tube, Rail, Walk 5 Bus, Rail ASK at Brunel (2013) Can we enter responses like this? 7
- 8. Multiple response Transport 1 Car 2 Bus, Tube 3 Car 4 Dr. Christine Pereira Participant Tube, Rail, Walk 5 Bus, Rail ASK at Brunel (2013) Can we enter responses like this? Why not? 8
- 9. Multiple response Transport 1 Car 2 Bus, Tube 3 Car 4 Dr. Christine Pereira Participant Tube, Rail, Walk 5 Bus, Rail ASK at Brunel (2013) Can we enter responses like this? Why not? Only ONE response per cell for statistical analysis 9
- 10. Multiple response Car Bus Tube Rail Walk Cycle 1 Yes No No No No No 2 No Yes Yes No No No 3 Yes No No No No No 4 Dr. Christine Pereira Participant No No Yes Yes Yes No 5 No Yes No Yes No No ASK at Brunel (2013) Multiple response survey question Each response is a new variable 10
- 11. Multiple response Car Bus Tube Rail Walk Cycle 1 Yes No No No No No 2 No Yes Yes No No No 3 Yes 4 Dr. Christine Pereira Participant No No Yes Yes Yes No 5 No Yes No Yes No No ASK at Brunel (2013) Yes = They travel by car No No No No No = They do not travel by car Multiple response survey question Each response is a new variable No 11
- 12. Multiple response Car Bus Tube Rail Walk Cycle 1 Yes No No No No No 2 No Yes Yes No No No 3 Yes No 4 Dr. Christine Pereira Participant No No Yes Yes Yes No 5 No Yes No Yes No No ASK at Brunel (2013) Yes = They travel by bus No No No No No = They do not travel by bus Multiple response survey question Each response is a new variable 12
- 13. Ranked Response Questions » Obtain preferential judgements or opinions. Rank the following subjects from 1-5 based on how much you like them. (1 = Favourite, 5 = Least favourite) Rank in order of importance, from 1-5, the following features when considering a destination for your holiday. Dr. Christine Pereira (1 = Most important) History Science Math Writing IT I’ve never been before Weather Surroundings Cost Accommodation These also are multiple response questions ASK at Brunel (2013) 13
- 14. Ranked multiple response Never been Weather Surroundings Cost Accommodation 1 Most Impt 4 2 3 5 2 5 2 3 Most Impt 4 3 5 Most Impt 2 4 3 4 Dr. Christine Pereira Participant 3 4 5 Most Impt 2 5 4 2 Most Impt 5 3 Multiple response survey question Each response is a new variable ASK at Brunel (2013) 14
- 15. Scaled Response Questions » Obtain preferential judgements or opinions Likert scale: Level of agreement Other scales: Level of importance or satisfaction Dr. Christine Pereira How important are the following when considering where to live? (circle one) Very Important Important I don’t know Unimportant Very Unimportant Cost 1 2 3 4 5 Distance to Uni 1 2 3 4 5 Distance to work 1 2 3 4 5 I feel safe 1 2 3 4 5 ASK at Brunel (2013) 15
- 16. Scaled multiple response Cost Dist Uni Dist Work Safety 1 Very Impt Important I don’t know Important 2 Important Very Impt Very Impt Important 3 Unimportant Very Impt Important Very Impt 4 Dr. Christine Pereira Participant Unimportant I don’t know Important Important 5 Very Impt Important Important Very Impt Single response survey questions. Each question is a new variable. ASK at Brunel (2013) 16
- 17. Dr. Christine Pereira Using the ‘Compute variable’ tool ASK at Brunel (2013) 17
- 18. Dr. Christine Pereira Reverse coding or regroup responses ASK at Brunel (2013) 18
- 19. Dr. Christine Pereira Select specific cases of the data to analyse ASK at Brunel (2013) 19
- 20. Dr. Christine Pereira Analyse groups separately ASK at Brunel (2013) 20
- 21. Dr. Christine Pereira ASK at Brunel (2013) 21
- 22. Dr. Christine Pereira Two methods: 1. ‘Explore’ option 2. Split file, then ‘Frequencies’ option We will use the 2nd method. ASK at Brunel (2013) 22
- 23. Use descriptives to summarise the relationship between Annual Income split by Gender 1 Dr. Christine Pereira 2 What is your gross annual income? Gender (please tick one) ASK at Brunel (2013) Male Female 23
- 24. Dr. Christine Pereira ASK at Brunel (2013) 24
- 25. Output for males and females will be in one table Dr. Christine Pereira Separate output tables for males and females Categorical Variable File must be sorted! ASK at Brunel (2013) 25
- 26. Dr. Christine Pereira The file is sorted All females are listed first, then all males ASK at Brunel (2013) 26
- 27. Dr. Christine Pereira Now get stats for Annual Income • Use ‘Frequencies’ • Output will be split by Gender ASK at Brunel (2013) 27
- 28. Dr. Christine Pereira Use descriptives to summarise the relationship between Annual Income split by Gender ASK at Brunel (2013) 28
- 29. Use descriptives to summarise the relationship between Annual Income split by Gender Split File Option to Compare Groups Dr. Christine Pereira Output for males and females is in one table ASK at Brunel (2013) 29
- 30. Dr. Christine Pereira Separate output tables for males and females ASK at Brunel (2013) 30
- 31. Dr. Christine Pereira Use descriptives to summarise the relationship between Annual Income split by Gender Split File Option to Organise Output by Groups Separate output table for males and females ASK at Brunel (2013) 31
- 32. Dr. Christine Pereira When finished, return the file to normal. ASK at Brunel (2013) 32
- 33. Dr. Christine Pereira ASK at Brunel (2013) 33