I'm doing my dissertation. I did my data analysis. I have a total of 130 responses. My survey had three sections. Demographics, and two instruments. For the first instrument, I had 130 responses. However, for the second instrument, I had 111 participants. my main analysis is multiple regression analysis. so the SPSS only included the shared responses, which are a total of 111. My methodologist advised me to repeat the analysis of everything and only work on 111 participants because the reader would wonder why I included participants who are not included in the test of my hypotheses, meaning exclude all participants who did not answer the second instrument. For me, I don't think that having an uneven number of participants is a problem. However, I want to know what experts would think when they read my paper.
also, when or what would determine the decision of including or excluding participants?
If you do any analyses that include both your instruments, that will automatically be done on an N of 111, because the coefficients can only be calculated on cases where complete data is present.
Alternatively, if you do all your analyses separately, you will not be able to compare any that involve the two different instruments because you used different samples.
So, either way, you are better to use the subset with complete data.
You could, however, ask your statistician about a more complex alternative, which would involve the "imputation" of your missing data.
Alaa Alhejaili - It is not only important to define the appropriate inclusion and exclusion criteria when designing a study but also evaluate how those decisions will impact the external validity of the results of the study. Inclusion criteria are the key features of the target population that the investigators will use to answer their research question. In contrast, exclusion criteria are defined as features of the potential study participants who meet the inclusion criteria but present with additional characteristics that could interfere with the success of the study or increase their risk for an unfavorable outcome.
Thank you so much David L Morgan and Dr Manzoor Hussain I really appreciate your explanations. So I did exclude the extra responses and re-run the analyses again. Results of regression, correlations, and reliability, and so forth, remained the same. I think what also played a role is that the number of responses to the multiple regression analysis in my study was decent (130) total, and even when excluding the extra (19) participants, I still got 111.
Thank you so much
Alaa
- Badges
- Science topic
- Similar topics
- Cognitive Science
- Thinking