If all the scores of the evaluated variables are in numerical, or in case they are a percentage and can be converted to decimals, from this you can perform spearman correlation because it is likely a lack of normal distribution of the data. Although I always recommend the use of relationship models based on structural equations that allow you to generate more robust analysis and more accurate results.

Apologies But If you believe there is a relationship beween variables why not actually find it with regression.. Now I am not sure but it looks like your DV is the Likert variable , level of usage. However, I do not understand why and/or how you got the total maturity score. But let'say you can provide a good JMR reference on it. Now I assume the toal maturity scores are integers. Now I am completly confused by what you say in Number 2.

BUT I had a very good friend in Marketing Research that did research at least somewahat like this, Prof. David Fritsche. I suggest what you really want to do is,to Quote Wikipedia:"

'Conjoint analysis' is a survey-based statistical technique used in market research that helps determine how people value different attributes that make up an individual product or service. Wikipedia" You may find a lot of useful information here: https://www.google.com/search?biw=1584&bih=779&ei=t2S4XuOlDruE9PwPtdG-qAI&q=conjoint+analysis+marketing&oq=Conjoint&gs_lcp=CgZwc3ktYWIQARgAMgcIABBDEIsDMgcIABBDEIsDMgIIADICCAAyBAgAEEMyAggAMgQIABBDMgIIADIECAAQQzICCAA6DggAEOoCELQCEJoBEOUCOgUIABCLAzoICAAQgwEQiwM6BwgAEIMBEENQycYBWPnmAWCanQJoAXAAeACAAXCIAbgFkgEDNi4ymAEAoAEBqgEHZ3dzLXdperABBrgBAw&sclient=psy-ab

To be blunt about it, I think your design is unclear, poorly developed, and lacked any clue for what to do after you had data. This a classic description of research failure. ALWAYS PLAN WHAT TO DO AFTER YOU HAVE DATA. You didn't do that so here we are with a salvage operation as statisticians usually call it. So what to do?

you appear to be a beginning graduate student in Marketing. This may be a thesis project. Your best chance of success is to redeisign your study SEE: https://www.google.com/search?biw=1600&bih=795&ei=3GS4XsiGMdW6tAaVkIboCA&q=Designing+the+Marketing+Research+Study&oq=Designing+the+Marketing+Research+Study&gs_lcp=CgZwc3ktYWIQAzIICCEQFhAdEB4yCAghEBYQHRAeMggIIRAWEB0QHjIICCEQFhAdEB4yCAghEBYQHRAeMggIIRAWEB0QHjIICCEQFhAdEB4yCAghEBYQHRAeMggIIRAWEB0QHjIICCEQFhAdEB46BAgAEEc6BAgAEEM6AggAOgYIABAWEB46DggAEOoCELQCEJoBEOUCOgUIABCDAToHCAAQQxCLAzoICAAQgwEQiwM6BQgAEIsDUNrrPVjctT9g1rw_aAFwAXgEgAGrAogBxyySAQcyOS4xNS41mAEAoAEBqgEHZ3dzLXdperABBrgBAg&sclient=psy-ab&ved=0ahUKEwjI5KiIkqrpAhVVHc0KHRWIAY0Q4dUDCAw&uact=5

If you are interested in the topic of this post see if it fits in the framework of conjoint analysis. If it does design it from that perspective. If not, and you probably should do this anyway, get advice from a senior researcher. Actually that is what professors are for when you are in grad school. Be sure to be clear about 4 things:

1 your research question

2 sampling and how your data will be collected(including things like (1),(2), etc in your question above-CLEARLY STATED

3 What will you do with the data after collection?. Perhaps Conjoint analysis can be of help here. But whatever it is write it out and try it on set of made up data of the sort you plan to collect. If you can analyze this you should be OK.

4 Finally make sure this will answer your research question. If so you are ready to go collect the real data. Remember this: Proper Prior Planning Prevents Poor Performane. Feel free to return here for assisance if needed.

Best wishes on a fine career.

David Booth

The type of data may determined the type of regression. If the dependent Variable is 0 and 1 and the independent variables are discrete variables you can use logistic regression which will be valuable to determine which of the independent variables is risky or protective.

You can separate the data and analyze each group alone.

David Eugene Booth Thank you for your answer.

Yes it is a Master Thesis, and I designed my research from a maturity model that a media agency provided to me. So the aim of my research is to find a relationship between the results of this model applied to 8 selected brands, that measures the level of maturity of brands on social media, and social media users' perceptions of the selected brands. This is why I have two different variable sources and my description looks confused.

At this point, I cannot redesign my survey to include conjoint analysis. Therefore, I can only research on how to best interpret my data. In particular, what kind of correlation and regression since the hypothesis is that there is a relationship between the levels of maturity (measured by me using the agency's model) and the consumer mindset metrics (measured on a Likert scale in the survey).

Hoping that I responded correctly to your critiques.

Well as far I can see it is confused. To do regression or correlation you must measure the 2 variables together at each point. You havent done that as far as I can see so how are you going to do your data analysis? Best, D. Booth

Para corroborar la relación entre dos variables es sencillo, se aplica Rho de spearman, Chi cuadrado en general, las variables específicas de igual manera sugiero que trabajes con dimensiones.(No puedes usar análisis de regresión porque tu nivel de investigación no es causal o explicativo. ( tienes que ordenar bien la investigación tiene que haber lógica).