Degrees of freedom in correlational test

Hello, I am a novice in statistics, so please forgive me if my question is too basic...
At the uni we did a research in our group and used a correlational test to investigate the relationships... however our degrees of freedom equaled to the sample size (df = n) and when the tutor was asked why it isn't n - 2 she said it was because we were not interested in the population, but she couldn't explain it further.
Could someone please explain in what cases df = n and why...
I cannot find anything on the internet :(
thank you


Less is more. Stay pure. Stay poor.
Were you all not interested in the population because you had all of the data already? If you have all data, you don't have to worry about precision values such as standard errors, but there is not doubt because you have the full population. Was it something like this?
thank you, yes we had all the data your reply does make sense to me I think.
but just to confirm, so df in general connected to the sample mean distribution and because we had the sample that was also the population, we didn't have to take into account this stander error of the mean, am I getting it right?


Less is more. Stay pure. Stay poor.
Statistics are based on making inferences from the sample level to the population. This is usually done using test statistics or estimates with confidence intervals. If you did not have a sample but the whole population, you wouldn't need to run a test or generate an estimate with precision intervals - since you don't have to inference to the population.

Look at the U.S. presidential election, if you knew how each person was going to vote via a survey everyone completed truthfully, you wouldn't need a margin of error of estimate you would be able to say 62% would vote for person X. Instead if you had a sample you have to say 62% people would vote for X with a +/- 4% margin of error - to quantify the uncertainty. You have no uncertainty when the sample = population.