For my thesis I have done a longitudinal study of sending emails to a group of participants
Before the experiment began I asked people a question which some answered.
Then afterwards I asked the participants the same question to seek for differences.
Because it was completely annonymous I do not have a perfect 1 on 1 answer ratio where I could have preformed a paired sample t -test.
Now I just have the mean and loose answers of the group.
Which test should I use to say something about the answer to the question?
The question is an 5-point likert scale question.
All help is appreciated.
Watch this video on “How to choose the correct statistical test when we are comparing the groups for statistical difference” https://youtu.be/aUp8sZXDzc8
It seems that you have a repeated measures design which refers to measuring each participant under several different conditions. Since you expect missing data over time, your design could be unbalanced, having an inconsistent number of observations on participants. Therefore, you can go for generalized linear models (GLMs) for correlated outcomes due to their capability in dealing with such variations in a unified manner. You might go through chapter 21 of the book by Fitzmaurice et al. (2011) for insightful inputs on approaches for longitudinal data of repeated measures and related designs. You could also check out the article by Arnau et al. (2016) for an account of the analysis of longitudinal data and the use of GLMs. Here are the full citations.
Arnau, J., Bono, R., Bendayan, R., & Blanca, M. J. (2016). Analyzing longitudinal data and use of the generalized linear model in health and social sciences. Quality & Quantity, 50(2), 693–707. https://doi.org/10.1007/s11135-015-0171-7
Fitzmaurice, G. M., Laird, N. M., & Ware, J. H. (2011). Applied longitudinal analysis (2nd ed). Wiley. https://www.wiley.com/en-us/Applied+Longitudinal+Analysis%2C+2nd+Edition-p-9780470380277
Good luck,
A video on “Choosing the Correct Statistical Test (For Relationship between variables) with examples” https://youtu.be/1M5DQphChOg might be helpful.
The classification of variables represents one of the parameters that make it possible to choose statistical tests, which is why it is important in the research process.
Among the criteria for selecting the statistical test to be used, with the support of heuristics, are the following:
- The research level.
- The statistical objective.
- The type of variable.
- The type of research.
- The research design.
- The behaviour of the data.
- Badges
- Science topic
- Similar topics
- Thesis Research