1
$\begingroup$

I will openly admit that statistics and statistical programming is not really in my wheelhouse. That said, my job requires it. I am trying to analyze the data gathered from a pre and post test survey where responses were measured with a 1-5 likert scale. 1-strongly agree and 5 is strongly disagree. Essentially we would like to prove that our intervention changes attitudes closer to strongly agree for the majority.

I would like to be able to show the percentage of growth for each of our 14 questions from pre to post test. Essentially I need to know what is the change?

Again I am using SPSS and in way over my head.

$\endgroup$
2
  • $\begingroup$ You could study mean changes from pre to post or you could study %agree or strongly agree changes from pre to post for each question. Is that what you want? $\endgroup$
    – tomka
    Commented Jan 8, 2014 at 16:57
  • $\begingroup$ I guess what I'm trying to say is I want to prove percent change from before and after intervention but I'm concerned because 1, a lower value, represents strongly agree the results will not necessarily be representative of what we want. $\endgroup$
    – Andrea
    Commented Jan 8, 2014 at 17:02

2 Answers 2

2
$\begingroup$

I think you are looking for a test of equality of two proportions in paired data. A proper test may be the McNemar test.

http://de.wikipedia.org/wiki/McNemar-Test

In SPSS, you can check McNemar test under 'statistics' in the crosstabs dialogue. It will provide the proper test for a table using the pre/post measurement.

If you only want to test one proportion, you first need to recode the variable to an indicator variable, e.g. 1 vs. 4 to 5 (i.e., strongly agree vs. all other).

$\endgroup$
2
$\begingroup$

If you want to prove that the majority moved closer to strongly agree then you can simply look, for each question, at the proportion who moved closer to strongly agree. If 50% or more moved closer, then a majority did. If you want to generalize to a population then you can find a confidence interval for that proportion.

But then you say "What is the change?" That is a different question. For that, you'd have to estimate the change on the likert scale (assuming you have a likert scale)

However, if all this is way over your head, you might need to hire a consultant.

$\endgroup$
1
  • $\begingroup$ Unfortunately hiring a consultant is out of the question. I am a graduate assistant in a largely underfunded department of a state school. The budget just isn't there. $\endgroup$
    – Andrea
    Commented Jan 8, 2014 at 18:08

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.