Hello,
Using factor analysis, I recently created a social questionnaire with four factors, each containing four items. My next step is validating the questionnaire.
I want to show that this questionnaire can differentiate between 3 product categories that have different social characteristics.
- What statistical method should I use to prove the differences?
- Should I ask the same respondents to evaluate all 3 products? or it is ok to have separate respondents for each product category.
Thank you
I would say both same or different participants could work. With "same", you may encounter order effects, which may be undesirable. Such oder effects could be avoided by using counterbalancing or by selecting different independent samples. When selecting different independent samples, I would make sure that they each represent a truly random sample from the same population so that differences can be attributed to the product categories with high certainly (rather than pre-existing differences between the samples).
In terms of statistical method, how about using (M)ANOVA for comparing the means across products? In the case of the same participants, you would use methods for dependent samples (e.g., repeated measures ANOVA), whereas for independent samples, you would use non-repeated measures (M)ANOVA.
Alternatively, since you already used factor analysis to develop the questionnaire, you may be able to use multigroup confirmatory factor analysis (CFA) to compare independent groups/samples or longitudinal CFA for dependent samples (repeated measurements of the same people).
As long as you have complete data (i.e., responses on all four scales and evaluations of all three products), then your statistical approach will be identical. So, if you use the same requirements for both samples, then it should not matter which one you use.
Hello
Exploratory factor analysis has an inductive approach, which means that whenever you want to know the items of your questionnaire are able to identify factors beyond the factors planned in the questionnaire, we use exploratory factor analysis. In simpler terms, exploratory factor analysis seeks the hidden structures in the questionnaire items; exploratory factor analysis seeks to identify the underlying structures of your questionnaire.
In contrast, confirmatory factor analysis has a deductive approach, which compares the structured factors in your questionnaire with experimental data and examines the fit between them. To put it simply, this type of analysis does not discover a new factor but will confirm or reject the construct validity of your questionnaire.
In order to identify the ability of these items to distinguish different groups, there are various solutions in multivariate statistical analysis. MANOVA and Discriminant Analysis are the two's that I suggest to you.
good luck
I also recommend that the sample you use to check the validity of your questionnaire be separate from the experimental sample you use to differentiate between products.
- Badges
- Science topic
- Similar topics
- Social Psychology
- Attitude