I have rt and accuracy measurements for 4 types of linguistic stimuli differing only on semantic terms. I ran the experiment twice on the same sample, one of them included an experimental manipulation. Effects of the experimental manipulation is found to be significant on both rt and accuracy measurements of the all 4 types of stimuli.
What I need to learn in the next step is which type of stimuli was more affected by the experimental manipulation. I ran a pairwise comparison test in repeated measures ANOVA but I felt that it is not OK.
Which type of analysis should I use?
Difficult to answer with the information given. What are the variables? Categorical? Ordinal? Binary? Are you comparing each sample against itself? Are sample sizes balanced?
One can find statistical significance, which could be noise in a small sample or nouse in a large sample. What does statistical significance mean practically? Are you trying to classify data?
https://opentextbc.ca/researchmethods/chapter/multiple-dependent-variables/
My sample consists of 72 participants. They answered 160 questions consisting of 4 semantic categories and all items are standardized on syntactic terms (number of words in the sentence, lenght of the words etc.) Measurements of accuracy and reaction times were taken. 10 days later they were given the same task but with an experimental manipulation. The total effect of experimental condition on rt and accuracy was significant but I also want to learn if there is a difference on the effects of experimental manipulation on the semantic categories.
The dependent variables are continous since they are rt and accuracy scores. However they can be divided into 4 different semantic categories.
Özge Yılmaz Sounds like you had 2 conditions one of which you designated as experimental. In both condition, your participants answered questions with 4 semantic categories. It’s yet unclear what the nature of the experimental manipulation was and so it’s difficult to understand what “total effect of experimental condition on rt and accuracy was significant” really means. Did you determine the rt and accuracy for each semantic category in each condition or just the aggregate rt and accuracy were used? What statistical analysis was conducted for your significance conclusion?
- Badges
- Science topic
- Similar topics
- Psychology
- Mental Processes
- Learning