SPSS - Descriptive statistics of questionnaire items under categories

0
votes

I have a question and would like to know whether it is possible or meaningful to do the following.

I conducted a questionnaire about public transportation. Public transportation includes bus, tram and taxi. Under each type of transportation, there are 3 questions using five-level Likert scale (1 means 'strongly disagree' and 5 means 'strongly agree'). In total, I have 9 questions. I received 100 responses and could summarize my data file as below:

  • Bus: or B for short with 3 questions (3 items) and named them with variable names as B1, B2 and B3.
  • Tram: or T for short3 questions (3 items) and named them with variable names as T1, T2 and T3.
  • Taxi: or TA for short 3 questions (3 items) and named them with variable names as TA1, TA2 and TA3.

When I put everything in SPSS, it looks like:

No | B1 | B2 | B3 | T1 | T2 | T3 | TA1 | TA2 | TA3
1  | 1  |  2 | 3  |  3 | 2  | 5  | 2   | 4   |  2    
2  | 2  |  2 | 2  |  5 | 4  | 5  | 4   | 3   |  2         
3  | 3  |  5 | 4  |  2 | 2  | 1  | 5   | 2   |  3           
4  | 2  |  4 | 5  |  1 | 3  | 2  | 1   | 4   |  1   
5  | 2  |  4 | 5  |  1 | 3  | 2  | 1   | 4   |  1
.
.
.

100  | 4  |  3 | 4  |  3 | 4  | 2  | 2   | 4   |  2

Now, I could run descriptive analysis for each item, no problem. There is a separate descriptive analysis table for each item from B1 ... TA3.

However, I would like to know whether I could group B1, B2 and B3; and run a descriptive Analysis for B (representing B1, B2 and B3) to explore the level of agreement or disagreement of respondent to Bus in general? Does it make sense for me to run this kind of analysis? Or I could only run a descriptive analysis for each item?

1
statisticsanalysisspss

1 Answers

1
votes

You could develop a simple sum index 

comp bus=b1+b2+b3.

or calculate averages 

comp bus=(b1+b2+b3)/3.

Then you could run the descriptives for the new variables.

Having said this, I would think there is a reason that you used 3 different questions for each mode of transport, so I would analyse them separately.

Building a "general" assessment with the indices above is a huge leap in theory that assumes that all 3 variables carry the same weight and are able to represent the assessment of the "bus in general". I did similar studies and my experience is that even with many more items you do not arrive there.

A general assessment of something usually is a different beast and should be asked in a dedicated question.