1
$\begingroup$

The UK Government publishes deprivation scores for geographic areas. Often these scores are converted into ranks and then into quantiles to see if groups of people are more or less deprived than one would expect.

The IMD score is ordinal, Area A may be more deprived than Area B, but we cannot say by how much.

Is it valid to use IMD Rank as an independent variable in a simple or multiple linear regression?

More information on IMD here: English Indices of Deprivation 2010

$\endgroup$
1
$\begingroup$

Yes: you're not invalidating any assumption of the linear model. Of course there's no guarantee your model will be much use - there never is. I'd be inclined to try the raw deprivation scores first if they were available.

| cite | improve this answer | |
$\endgroup$
  • $\begingroup$ Thanks @Scortchi, raw scores are available to me, in fact I even have the domain scores that are combined to make up the overall IMD score so I'm tempted to split them out and add as separate variables. $\endgroup$ – Tumbledown Jan 31 '13 at 10:17

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.