Question

I am trying to figure out which test analysis to use for my research questions. I was thinking about think about multivariate because of the number of variable being addressed in the study but there is also the possibility to use univariate to address each question.

What are the current levels of police satisfaction in CMPD jurisdictions?

What is the public’s perception of crime in CMPD jurisdictions?

Does “hot spot” policing reduce crime in CMPD jurisdictions?

How does broken windows policing impact racial and ethnic groups in CMPD jurisdictions?

Answer 1

For this research, multivariate as well as univariate analysis can be conducted to explore the research question. The most important thing is that which test analysis to use for this research questions. First of all we see, in first question we need to explore "current levels of police satisfaction in CMPD jurisdictions". For this question, we need to setup likert scale questions adressing police satisfaction in CMPD jurisdictions. After that, likert scale questions, can be analysed using non-parametric test like Mann Whitney U test for comparing two groups like gender (Male versus Female), status (One rank versus other rank) and so on. Chi square test for association can be also employed to assess the association between different parameters of police satisfaction in CMPD jurisdictions. Similar approach can be employed in case of "What is the public’s perception of crime in CMPD jurisdictions?".

This question "Does “hot spot” policing reduce crime in CMPD jurisdictions?" can be well analysed using logistic regression technique (Multivariate Analysis). Logistic regression will help us identifying whether really “hot spot” policing reduce crime in CMPD jurisdictions.

The question "How does broken windows policing impact racial and ethnic groups in CMPD jurisdictions?" can be well analysed by association tests like Chi square and fishers exact tests to explore how racial and ethnic groups in CMPD jurisdictions are getting impact from broken windows policing.