Which test can be essential for validation of a questioneir to report in an article?
Scale validation is a rather complicated issue but the first thing one should probably always do is examine the dimensionality/factor structure ("factorial validity") of the items using either item response theory (IRT) or factor analysis methods. This would allow you to examine, for example, whether your items measure the intended dimensions ("factors" or "traits").
Unfortunately, SPSS has very limited options for both IRT and item-level (ordinal) factor analysis (but see https://www.ibm.com/support/pages/item-response-theoryrasch-models-spss-statistics). Specialized programs for confirmatory factor analysis and structural equation modeling such as lavaan (in R) and Mplus offer more flexibility in the proper analysis of ordinal items (Mplus does both IRT and ordinal factor models).
Aisha Iftikhar You may use SPSS to evaluate a Likert scale questionnaire by running many statistical tests to ensure the questionnaire's reliability and validity. You can do the following steps:
1. Cronbach's alpha is used to calculate the questionnaire's reliability. This will assist you in determining the questionnaire's internal consistency.
2. Using histograms and the Kolmogorov-Smirnov test, check for univariate normality. This ensures that the data has a normal distribution, which is required by many statistical tests.
3. Using Pearson's correlation coefficient, look for relationships between items. This can assist you in identifying any duplicate things that may be causing a high Cronbach's alpha rating.
4. Compare the correlations between pairs of things that are anticipated to be connected to each other with the correlations between pairs of items that are not expected to be related.
5. Convergent validity may be determined by analyzing the correlations between the questions on your questionnaire and a measure of the idea that the questionnaire is designed to evaluate.
6. Check for known-groups validity by comparing the mean questionnaire scores for distinct groups that are known to differ on the subject under consideration.
7. Face validity may be determined by having a panel of professionals assess the questionnaire and provide input on its substance and language.
I hope this was helpful! let me know if you have any more queries.
Christian Geiser Fatemeh Khozaei thankyou so much for giving your precious suggestions.
You can check: https://www.npmj.org/article.asp?issn=1117-1936;year=2015;volume=22;issue=4;spage=195;epage=201;aulast=Bolarinwa
- Badges
- Science topic
- Similar topics
- Statistics
- Survey