My analysis uses students' test scores. A similar question was answered by Stefano Giuseppe Lazzarini in a forum but I am posting this message separately that the old discussion might not be visited by others as it is considered the issue was solved. However, my confusion is not cleared. My meta-analysis using Revman 5.4 showing the diamond box in the opposite direction. I am analyzing students' scores. Even though Pretest and Posttest score is not recommended in the meta-analysis, but this conflict exists exactly in a similar data structure, therefore I intentionally use pretest and posttest scores because I know posttest is definitely better. But the output result showed pretest is more effective, as the diamond box is on the pretest column. The diamond box on the forest plot is showing in the opposing direction, in the wrong direction. In one forum suggested by Stefano Giuseppe Lazzarini, I multiply the data by -1, then the diamond box moved to the posttest score. It solved in this particular case. I requested the Revman software team for clarification. They suggested consulting with the expert statistician. My question here is: When should I multiply by -1 or not? How should I know in a more complex situation that the diamond box is in the wrong direction? In the above example, as I use data from pretest and posttest, I could detect this conflict. I also used Mean Difference and Standard Deviation Difference setting, but nothing has changed in the direction of the diamond box if I do not multiply by -1. I need help for more clarification on this issue. I attached forest plots of my examples hereby. Thank you. Mr. Dina
These results are based on secondary data .Thus they should be reported as it is.It would not be a good idea to to take the post experiment as effective .Further refer to https://www.youtube.com/channel/UCj38Topqj8yC3SdrV6HypIg
@ Jaspreet Kaur, thank you very much. Your youtube link is a very much valuable resource.
If your analysis is correct, then leave it as is as Jaspreet Kaur mentioned :)
- Badges
- Science topic
- Similar topics
- Counseling Psychology
- Confusion