The question attempts to get counsel on the ethics of reporting when during exploratory data analysis reclassification of a variable from "High, Moderate, Low" categories to "High and Low" only yield different results.
Will it be to report the one that favour's the research questions raised or the one that amend to the conference theme if the paper is been prepared for such?
As expected there will be different result because you the moderate class is taken off or merge.
how did you classify variables as high, moderate and low? It attached to numbers e.g 1-3 low, 4-7 moderate and 8-10 high,
Then you can reclassify as 1-4 low, 5-10 high. Nothing should be left out.
I am unsure why you need to yake off the moderate class, because 5 effect range is commonly used - very low, low, moderate, high, very high.
SORRY NOT SURE IF THIS HELP
@ Omololu_I am getting an insight from your 5 effect range categorisation now. I think I will experiment with this too as it amends to the Likert type scaling system. I am sure I will yet get another output using this. I think one can then see the one that maximises the variables I.e. two, three and five categorisations. Your response has sure opened new frontiers to help narrow my conjecture. Thank you.
- Badges
- Science topic
- Similar topics
- Philosophy
- Ethics