You do not necessarily need to use hypotheses, but could use research questions in the research proposal! Refer to some of my earlier journal articles on ResearchGate, which all formed part of my doctoral thesis by publication, and you will see that I only used research questions (the same was applicable to my research proposal).

Fortune Abla Nyamadaxe Djirackor No, hypotheses are not required in every quantitative research. Hypotheses are not required in descriptive investigations. However, hypotheses are required for RCT and experimental investigations, as well as when using inferential statistics.

Hypotheses are testable claims that are related to your research issue. Hypotheses bridge the gap between the broad subject you want to study (i.e., the research question) and short assertions of what you think the relationship between your variables is.

It is a precise, explicit, and testable hypothesis concerning the research's result. It is not required for all types of study. It is required for quantitative research but is not required for qualitative research.

I highly agree with Rodney Duffett , as I also prefer the research question route. However there is nothing wrong in choosing hypothesis route either but it is not a necessity in quantitative research.

Hypotheses are used to predict how variables interact or relate to each other. If you are doing a quantitative study that is correlational, quasi-experimental or experimental, you should have hypotheses. They are what guide your outcomes and eventually get reported in the conclusions. Research questions are typically used in descriptive quantitative studies since those types of studies do not usually predict relationships among variables.

Good luck with your study.

Fortune Abla Nyamadaxe Djirackor ,

Hypotheses are testable assertions about the research question and objective. Hypotheses bridge the gap between the broad question you intend to investigate (i.e., the research question) and concise statements of the relationship between variables that you hypothesise in the conceptual model. However, hypotheses are not needed for descriptive analysis, but Pearson, MRA, SEM, T-test, ANOVA, etc., are better to include research hypotheses.

Kind Regards,

Thank you all very much for your contributions.

Without clearly stated hypotheses a priori it can be tempting for researchers to collect and analyze data to 'prove' their pet theory. The null hypothesis requires quantitative researchers to begin with the assumption that their theory is wrong - i.e. there is no difference between groups or correlation between variables. If they correctly follow clearly described statistical methods, and reject the null hypothesis, they may have produced findings that suggest that their theory may not be wrong. (Mind you, this is not the way press releases are worded.)

I am not sure how a "research question" differs from a hypothesis. In particular, if the question has a yes or no answer, then that strikes as just the kind of testable difference that I would associate with a hypothesis.

That said, I don't think you need hypotheses in purely descriptive research, such as "What does X look like?"

RQ1: Are basketball players taller on average than jockeys? (test H0 with t-test).

RQ2: How much taller on average? (calculate difference in mean height with confidence interval, no hypothesis)

RQ3: Why are they taller? (describe requirements of the two sports associated with height/weight - no statistics, hypotheses or confidence intervals)?

Mark R Speechley My concern with your RQ 3 is that is that it doesn't require collecting any quantitative date.

Knowing variables suffice sometimes. No need for additional paradigms.

David - I agree, no data required for narrative explanation/interpretation. I guess what I'm getting at is they're all 'research questions'. I can imagine reading a quantitative kinesiology thesis concerning RQ1 and 2, and a qualitative thesis that tackled RQ3 using interviews of coaches and athletes, etc. (And I definitely could have come up with better examples of RQs!)

Thanks, Mark

Good to follow the question and answers. Thank you for sharing, professors.

Usually best practice to have a clear null hypothesis to start with as it forms the basis of your significance testing and other inferential testing. Clarity helps with resolving problems which may arise during data analysis which means your results and conclusions are valid. Your work is likely to be relevant and valuable, not heavily critiqued and therefore less valuable to use. Hope this helps.

Danni

The hypothesis in the proposal is very important because it can influence a study by directing a researcher to have a nice objective, research question, and a design for your study.

Hypothesis brings clarity and unhinder progress in the research process as well. That is the reason hypothesis fromulation is important. I agree with all the comments made here. Best.