One major option is to create scales from each of your four sets of related variables. Given that you already have assignments of variables to scales, this could be done via a Confirmatory Factor Analysis (CFA). However, its sounds like you are a relative beginner in this area, so it would be simpler to use a combination of coefficient alpha and Exploratory Factor Analysis (EFA) for this purpose.

Once you create your scales, you could treat them as interval variables and use them for correlations, ANOVA, etc. In contrast, the Wilcoxon signed rank test is for ordinal variables, and thus does not lend itself to the creation of scales.

Daniel Stone, if you can be confident that each of your four scales is unidimensional, you could certainly add up the responses to obtain a composite score for each scale (after reverse coding if necessary), and you could regard those composite scores as interval for subsequent analyses (despite them being comprised of ordinal items; I support David Morgan's point about that).

If you are unsure at the start about whether your four scales are unidimensional, that introduces a complication. It could well be advisable to conduct exploratory factor analysis on each scale to see what bubbles up.

It might be worth considering ANCOVA for your subsequent analyses - but you've not provided enough details to obtain authoritative advice concerning that.

Every good with with your research.

Before starting the analysis process, first, the paragraphs must be read carefully and after applying the apparent validity, which is one of the criteria of psychometric characteristics. Second, we must determine the negative and positive paragraphs because this makes a difference in the analysis.