My recollection is that religious activity and well being is very well researched but possibly not with medical front liners. If I am correct your study purpose has been examined before and likely a lot but your argument for contributing to the knowledge base is that it has never been done with your sample. That is not a reason in and of itself for doing the study, so you would have to argue, for example, that stress of front line workers may severely impact their well being and religious activity such as prayer may be particularly helpful. I see nothing wrong with doing a mixed methods design using correlations but a correlational design does not establish causality if that is what you mean by explanatory research.

Lovely Nobleza The goal of correlational research is to uncover variables that have a link to the extent that a change in one causes a change in the other. Unlike experimental research, which is totally based on scientific methods and hypothesis, this form of study is descriptive.

1. Correlations measure the association between variables. Unlike what Fatemeh says, they do not require that "a change in one causes a change in the other." This of course is one way that an association may arise, but there are others like another variable influences both. This will be covered in most intro textbooks. 2. You may have a hypothesis and use a sample correlation for inferential purposes (i..e, making inference about the population). This again is covered in intro books (and is why software gives p-values and CIs for correlations. 3. Given anyone answering these should know these, it unclear motivation underlying why someone appears to be misleading you (maybe they are "borrowing" from https://www.formpl.us/blog/correlational-research), but take some responses cautiously. 4. On your question, the use of different correlation statistics is fine in different research designs. Afterall, you can report a Pearson correlation from a group t-test from a placebo and experimental group, or observational studies. Exploratory versus confirmatory does affect what you might do with the results, a good reference is: author = {Spiegelhalter, David J.},

title = {Trust in numbers},

journal = {Journal of the Royal Statistical Society: {S}eries {A} (Statistics in Society)},

volume = {180},

pages = {948--965},

doi = {10.1111/rssa.12302},

year = {2017},

You say you are doing mixed methods, but I did not see anything about your qualitative data. One popular approach in mixed methods is a sequential explanatory design (QUANT --> qual), where you use a qualitative follow-up study to help understand the results of a prior quantitative result. If you have only the two quantitative variables, you could look for "outliers" from a correlation between them.