hello everyone, i need your help regarding these statistical tests since i'm not really big on stats. i want to know if there are any significant difference between the two groups in my study. i'm confused which statistical treatment i will be using.
here's a little background on my study. i grouped the respondents in my study into two groups. respondents with successful outcome and respondents with unsuccessful outcome. i have several independent variables in my study which include: age, gender, place of residence, marital status, income, employment status, substance use and co-morbidities. i want to know if there are significant difference between respondents with successful outcome and respondents with unsuccessful outcome with regards to the variables listed or if any of the variables can affect the outcome of the respodents.
age and income are the only continuous variables in my study while the rest are categorical variables.

which among these statistical treatment is better to analyze my data?

Hi plethoraldork,

If I understood well you have two different groups of participants (not-paired) tested in different conditions.
If I understood well, you have only one outcome that is a binary variable (either passed or failed?)

Mann-Whitney and independent t-test can be used when your outcome is a continuous variable (so no use for you).

Because you have a binary outcome (yes or no), then you need to do a Pearson' chi-square test (or Likelihood Ratio) to test if the difference between the two groups is significant.

So, say that you determined that the difference between the two groups is significant, then you want to see which predictors are significant. First of all, you need to know if your participants are independent. Assuming they are independent observations, you will need to do logistic regression, so you can include both categorical and continuous variables. So you will be able to determine to what extend substance use affects your outcome after controlling for place of residence for example.

It is very common in social sciences and epidemiology to have dependent participants. You can estimate the intraclass correlation coefficient (ICC) to see if they are dependent. Even a small dependence will affect your results. For example, participants living in the same neighborhood might be more similar than participants living in different neighborhoods. In order to take into account the hierarchy of the data, you will need to perform multilevel logistic regression (binary outcome: yes/no). Multilevel modelling allows to take into account contextual effects (effect of neighborhood), and also individual effects (age), as well as the interaction between these factors. Multilevel logistic regression is very similar to traditional logistic regression. There are special-purpose software to deal with dependent data.

Another way you can analyse your data, is to pair them but I am not going to confuse you with that, unless you organised the experiment like that?

Hope it helps, let me know if you have more questions.

I am a bit confused by your description.... Did any of these people receive an intervention, or is this simply one group where some did well and some did not?

You need to be a very careful with what you are planning to do.... It sounds like a fishing trip to me. What is your hypothesis? Did you have a pre-defined statistical plan?