I wanted to ensure my understanding is right ...

Paired test: We have two (or more) dependent populations, having therefore dependent samples

Non-paired test Two (or more) independent populations, having therefore independent samples

2) A test of difference in mean can be paired or non-paired, depending on the dependency of the populations?

Can someone please clarify this?


1 Answer 1



I will add some examples corresponding to your definitions:

  1. The same units of analysis (widgets, individual study participants, individual kangaroos, single nations, etc.) are repeatedly measured. For example, we can test for whether units of analysis at one time of measure differ from another time of measure.

  2. Different (as in randomly sampled) units of analysis are independently measured. For example, we can test for whether units of analysis in one category differ in their measure from those in another category.

Finally, I will say that when the number of repeated measures (or matched units of analysis) is greater than 2, "paired" is frequently instead termed "blocked," (and "unblocked" is the corresponding description of more than 2 independent groups), for example in repeated measures ANOVA and vanilla ANOVA.

  • $\begingroup$ @F0l0w feel free to click the up triangle to upvote my answer if you find it worthwhile. :) $\endgroup$
    – Alexis
    Dec 13, 2020 at 3:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.