1
$\begingroup$

I have the following output from Stata (see below), which is based on a linear regression model with an interaction term (female*region) and an independently added continuous predictor of age. The outcome (or dependent variable) is the bmi.

I can interpret the coefficients of female, region, and female*region. However, I am not quite sure how to correctly interpret the coefficient of age, which is not involved in the interaction term.

Can someone explain how to interpret the coefficient of age?

enter image description here

Improve this question
$\endgroup$
2
  • 1
    $\begingroup$ A very off-topic sidenote. Older people have a larger BMI than younger people. In many countries with an aging population this is a main driver for the increase in the BMI of the average population (big exceptions are the USA and the UK where the weight is increasing at every age). $\endgroup$
    – Sextus Empiricus
    Commented Jun 2 at 13:02
  • $\begingroup$ @SextusEmpiricus Thanks for your comment. Indeed, I believe this dataset is just 'fake' data for a Stata tutorial class $\endgroup$
    – KLee
    Commented Jun 3 at 14:04

1 Answer 1

Reset to default
4
$\begingroup$

As age is continuous, you would interpret the coefficient as all other variables kept constant, a one year increase in age leads to a 0.0488 unit increase in bmi.

Improve this answer
$\endgroup$
2
  • $\begingroup$ Hi @cha116 Thanks for your reply. Just one follow up question: With "all other variables held constant", does this refer to the case where the region is NE (baseline) and female is 0 (baseline)? $\endgroup$
    – KLee
    Commented May 31 at 14:56
  • 2
    $\begingroup$ Because there is no interaction, in this case you do not need to specify what the values of the other variables are. As long as age is the only value that is changes, the model predicts an increase of 0.05 per 1 year increase in age. $\endgroup$
    – cha116
    Commented May 31 at 15:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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