Dear all,
The data for my dependent variable is a grade given to a company. The grading system differ between one country to another and in my sample I have companies from different countries. In order to create the dataset for the dependent variable: can I standardize each company grade, i.e. use the equivalent Z-score (according to its country sub-group) in order to be able to judge them on the same scale?
Thanks in advance
any help will be highly appreciated :-)
Because the grading systems are different from country to country, I would say that the answer is no. If you decide to do it anyway, you would need a section in your paper that fully discusses the issue and how you made your decision. Best wishes, David
Thank you David Eugene Booth! in different grading systems I mean that in one country a company is evaluated based on two categories and in another country a (different) company is evaluated based on 3 categories. No company is evaluated in two different countries, just in its original country. in addition, the institute that is evaluating and giving them grades is the same institute.
still not a good method to standardize? this way I can "correct" each grade according to its sub group and then put them all (all companies from all countries) on the same scale, no?
@Rany
Normative standardization separately by group is the only hope to compare companies between country.
In another RG thread I provided some links to papers addressing the use of normative (for between-groups studies) and ipsative (for longitudinal studies) standardization as a means of circumventing paradoxical confounding (when that is possible). Here is the link to that thread:
- https://www.researchgate.net/post/Standardized_vs_Unstandarized_scores_for_regression
- Badges
- Science topic
- Similar topics
- Econ
- Business Economics