When mentioning the null/alternative hypothesis is not required in research proposal? Any good reference? In which study design this can be ignored? Someone told me about cross sectional designs which don't require, is that true?
Practice varies in different disciplines. These terms are part of a formal approach to hypothesis testing in frequentist statistics.
1) If you aren't using this formal approach then the terms are not required (e.g., in qualitative research).
2) If you have clearly specified your formal test it may not be necessary to state the null and alternative because they are implicit in any clearly formulated hypothesis test. This is I think an issue of style and to some extent who your readership is.
Always even if you use an alternate inference meth such as a confidence interval. Knowing.what.you want to do in.your research question requires thinking about these ideas see the attached and a modern book on Research Methods. Best wishes, David Booth
The null hypothesis is usually that the results were to chance, and I can't see how cross-sectional versus longitudinal data would make any difference in that.
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Materials Science
- Materials
- Inorganic Materials
- Metals
- Alloys
- Steel