I have conducted an experiment to counter COVID-19 vaccine misinformation, with one between-subjects factor (4 levels of corrective message conditions) and one within-subjects factor (3 levels of testing time), and the dependent variable being knowledge scores. Included in my secondary analyses is the exploration of additional variables I have collected (baseline attitudes and posttest perceived quality of corrective messages) with a specific question: Is there a relationship between perceived quality of corrective messages and attitudes towards COVID-19 vaccines; that is, do attitudes predict perceived quality of corrective messages? This is aimed at establishing ‘if’ pre-existing attitudes regarding COVID-19 vaccines significantly predict individuals’ perceived quality of corrective messages towards myths about the vaccines. It is important to note that the perceived quality of corrective message is a between-subjects outcome variable after each four participant group has been exposed to the assigned corrective message.
Now, my question is, should I be doing a simple linear regression to determine how much of the variation in perceived quality of corrective messages can be attributed to their pre-existing attitudes regarding COVID-19 vaccines? (Perceived quality of corrective messages is treated as the outcome variable while attitudes will be used as the predictor variable). Or is this a case for a multiple linear regression with one multi-level nominal variable (4 levels of corrective message conditions) and one continuous variable (attitude)? If the latter is the case, is there any references I can use to run and interpret results through SPSS?
Thank you very much for your help.
Very interesting question. If you consider the two elements in the equation one as a dependent variable and another as an independent variable, then go for the first option and the simple linear regression model will work.
Best of luck.
- Badges
- Science topic
- Similar topics
- Correction