How to know if I should use correlation or hypothesis test?

I just took a statistics course and since we did correlation, my grades went downhill in this course.
For example, in lab assignments where we had different data sets and had to find/use the appropriate test, I went from 98% to 74%.

I don't know what went wrong because we only got back the final grade but not the assignment itself but I think the reason why I got 74% could be due to the fact that I maybe(?) couldn't really distinguish between when to use correlation and when to used t-tests.

I know that correlation is used to find out if/how two variables are related and t-test are used to find out if there is a significant difference between them.

But how do I know which one to use when I get a data set with a question that doesn't ask directly "Is there a significant relationship" etc
Is there a trick to find out which test to use?

Thank you
I think the 'trick' is to really understand what each method/test is for (e.g., one sample t test tests whether an unknown population mean mu is different from a certain fixed value, two sample t test tests whether an uninown population mean is different from another unknown population mean, etc.). Be sure you know all the differences and similarities between different methods. Once you have a good overview of when to use which method, it becomes easier to select an appropriate method given the information in a question.


Omega Contributor
You can test hypotheses using correlations.

Correlations examine a relationship between two variables, if one increase what does the other paired variable do.
Thank you for your answers, but I actually followed these rules in my assignment...
We also had to test the assumptions for (for example) Pearson correlation. Does it matter in which order (linearity/scatterplot and then normality or vice versa) you test the assumptions? I'm wondering if I lost points with that.


Omega Contributor
You should just request a meeting with the professor to discuss the exam. Speculation and assumptions may not help you truly resolve the issues going forward.