In short, my MSc thesis involves providing recommendations to industry and these recommendations have been put to a series of experts in the form of a survey in order to validate the recommendations.
For each recommendation, four questions are asked relating to:
1. Whether the expert would advise clients to implement the recommendation
2. Whether businesses can generally afford to implement the recommendation
3. Whether businesses would understand the recommendation
4. Whether businesses could implement the recommendation without external support
Each of the four questions utilises a 5 point likert scale ranging from very unlikely to very likely.
I have never undertaken statistical analysis before and have historically been more of a words person than a numbers person. I'm intending to use SPSS for data analysis but if anyone were able to point me in the right direction of appropriate tests to run I would be exceptionally grateful.
https://students.shu.ac.uk/lits/it/documents/pdf/questionnaire_analysis_using_spss.pdf
Your Likert-scored variables are ordinal-level measures, so you should look up statistics for the analysis of ordinal data.
Dear Thomas,
Could you tell me what is your hypothesis?
I can say what you should use. But if I know it, My recommendations will be clear.
for Likert Scale analysis the Ordinal Scale measures are applied.
Like- Chi Square ( with Nominal Data), Correlational, t Test, F test, ANOVA etc.
It sounds like you’re analyzing four individual Likert-type items. I wouldn’t call these scales as that can cause confusion with scales that are composed of several items that are summed up or averaged, such as are used in psychology. The terms “Likert” and “scale” aren’t used consistently, but calling your items “items” or “questions” will avoid confusion.
If I’m correct about your analyzing individual Likert items, you want to treat your data as ordinal in nature, as David Morgan suggests. Appropriate analyzes may include traditional non-parametric tests like the Wilcoxon-Mann-Whitney test and ordinal regression. You can see what I wrote about this here:
https://www.researchgate.net/publication/337785034_How_Should_We_Analyze_Likert_Item_Data
https://www.nacaa.com/journal/index.php?jid=1001
(Both links the same article.)
That being said, it’s impossible to suggest a specific analysis without knowing what it is you are trying to determine. For example, you may want to know A) if the responses for each question were generally greater than 3. Or B) you might want to compare the responses from Region 1to those from Region 2. Or C) you might want to see if the responses are correlated to the size of the business.
Each of these scenarios would use a different analysis or test.
Treating your data as ordinal, some of the recommendations in some other responses are probably not the best. You probably don’t want to use a chi-square test of association (independence), nor t-test, nor anova.
Salvatore S. Mangiafico, your answer is complete and helpful.
Of course, SPSS allows you to use chi-square, but is it reasonable? It is not probably the best test for ordinal data, neither is it any parametric statistical test; but, again, as Magiafico states, "it’s impossible to suggest a specific analysis without knowing what it is you are trying to determine."
Bear in mind that the Chi-square statistics SPSS CROSSTABS reports include a so-called test of linear-by-linear association. It is the ordinal Chi-square that the late David Howell discussed in these notes:
- https://www.uvm.edu/~statdhtx/StatPages/More_Stuff/OrdinalChisq/OrdinalChiSq.html
I see that Salvatore discusses it here too:
- https://rcompanion.org/handbook/H_09.html
;-)
- Badges
- Science topic