Negative interaction term but main variables are positively correlated with each other and the dependent variable, how do I interpret?

[Eve Enright]

I am looking at the determinants of childcare use. I have local authority places available and household income as two of my independent variables for my regression on childcare use, they both have a positive impact on childcare use and when regressed on each other also show a positive relationship. However, when I include the interaction term between places available and household income, the coefficient of this is negative. How do I interpret this? Is it correct to interpret this as when household income is higher, the places available has less of an impact on these households using childcare, then for households at a lower income level. Many thanks in advance!

[Carlo Lazzaro]

Eve:
welcome to this forum.
As per FAQ, please share what you coded and what Stata gave you back. Thanks.

[Eve Enright]

This was my code: regress ccuse hhinc avail_pl c.hhincl##c.avail_pl

Thanks,
Eve

[Andrew Musau]

If you want to consider the marginal effects from the estimated model with interactions, use margins. But the following explains how to interpret the coefficient on the interaction: https://stats.oarc.ucla.edu/stata/fa...tion-stata-12/

[Carlo Lazzaro]

Eve:
...and what Stata gave you back. Thanks.