I am working with two dependent variables that are both measuring a country's level of nuclear security. The first variable is a country's nuclear security index rating (range 1-100). The second variable is the number of nuclear security incidents within a country - so this variable measures the number of times nuclear material has been stolen, lost, improperly disposed of, etc within a given country. Is it possible to combine these two variables into a single measure, and if so, how would I go about doing that? I have a feeling I might want to do something with the z scores, but I'm not sure if I'm on the right track. Thanks in advance!


Welcome to CV!

While you are of course free to come up with your own new index, combining the existing index and the number of incidents, I see two issues with this:

  1. Is the security index rating not at all based on number of incidents?
  2. Wouldn't you rather regress both separately to preserve the counts?

As for (1): If the security index in any way incorporates security incidences, then combining these numbers is futile, as the latter is already accounted for in the former.

Concerning (2): Counts of events can be modeled quite nicely with discrete GLMs (e.g., Poisson or negative binomial). This will render your results much more interpretable, as you can directly express the effect that certain variables have on the number of incidences. Any kind of combination of an index and a count will no longer be a count, which means that it will be harder to come up with an appropriate underlying distribution and it will be harder to interpret the magnitude of effects on your outcome variable.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.