Hi all,
I am currently in the process of writing up my report for a BMedSci on the quality of registries assessing pain.
Each participating registry has completed a questionnaire which included questions such as method of data collection and participant number.
I would like to calculate statistically whether there is an association between the number of participants (continuous) and method of data collection (paper, online or paper and online).
Would anyone be able to advise me on the correct statistical test to use for this comparison?
Thank you in advance
Hello Eleanor,
Either the eta-squared from a one-way anova (collection method = IV/factor, with 3 levels; participants = DV, though this is a count variable) or the R-squared from a multiple regression with two dummy variables (Dummy1 = 1/0 for paper, yes/no; Dummy2 = 1/0 for online, yes or no) and participants as the DV will quantify how much variance in number of participants may be explained by the choice of collection method.
Good luck with your work.
i think the best option is using one-way anova by three levels . which you take dummy variables for online , paper, and both. on other hand you can easily use correlation if you don"t care details and regression by taking particioants as Dv and taking others as dummy variables online(yes/no), paper( yes/no) , bith(yes/no)
I agree with David Morse and Abdullahi Farhan Abdi
In your scenario, the one-way ANOVA with 3 levels would be the best fitting test.
David Morse, Abdullahi Farhan Abdi, Shakil Ahmed
Thank you all for sharing your expertise within this field, I am so grateful!
Use one-way ANOVA. "Number of Participants" is your DV. Your IV can be "Method," where 1 = paper, 2 = online, and 3 = Both.
- Badges
- Science topic
- Similar topics
- Psychology