A question on academic writing in social science research. How do you frame hypothesis where independent variable is categorical and dependent variable is continuous?

In my specific case, i am interested to research if robo-analyst (Some sort of AI/automatic analyst) investment recommendations is profitable. The independent construct is 'analyst types' and the dependent variable is 'recommendation performance'.

Currently what i have is the following: Do Robo-Analyst analyst type relate positively to recommendation performance ?

But i am not sure if it is correct. Thanks !

  • 2
    I’m voting to close this question because I think it's a question about statistics not academia, which belongs here: stats.stackexchange.com
    – Kimball
    Oct 8, 2022 at 11:08

1 Answer 1


There is a difference between a research question and a hypothesis.You need both. What you proposed is a question, as there is a question mark at the end. What I miss here is the reference: Compared to who are the robo-analyst better? Compared to all humans, compared professional humans?

A hypothesis is a possible answer to (a part of) the research question. You base it on your theory. If the theory you presented is true, then the answer should be ... It is a statement, not a question. At the end of your project you can say that that statement is supported or not supported by your research. If it is, then that is counted as evidence supporting the underlying theory, and if it is not, then that is counted as evidence against the underlying theory.

  • Thank you Maarten. Yes, the reference is versus human analyst. Actually i wanted to test 2 hypothesis, (1) Whether the robo-analyst performance is positive stand-alone (2) Whether the robo-analyst is better than human. I think i know how to frame hypothesis part 2. i am stuck how to frame part 1 as a hypothesis statement. If you could enlighten me, i will be highly appreciative. thanks ! Sep 8, 2022 at 8:23
  • I think the problem you have is that 1) is too general. If you refine the term "positive" with something more specific like "make a profit", or something more appropriate for your exact research problem, then it would become easier. Sep 8, 2022 at 11:52

Not the answer you're looking for? Browse other questions tagged .