Trouble interpreting interaction term

[Jacob Green]

Hi, I was wondering if I could get some help as I am having trouble interpreting a interaction term.

As you can see, I am running an ordinal regression to try and explain the differences in digital literacy between individuals. The continuous variable "Q1" looks at the age of the individual , while the dummy variable "Q101" is whether a individual is male or female (0 as male and 1 as female).Is it true that the factor "Q1" shows that a unit increase in the age of a respondent would result in their ordered log-odds of being in higher Digital Literacy category decrease by .018 when "Q101" is equal to 0? So the interaction term of .009 is the impact that age has on digital literacy when the individual is female, so when "Q101" is set to 1. I am probably incorrect, so it would be great if someone can explain it to me- thanks.

[Clyde Schechter]

Is it true that the factor "Q1" shows that a unit increase in the age of a respondent would result in their ordered log-odds of being in higher Digital Literacy category decrease by .018 when "Q101" is equal to 0?

More or less. Since this is evidently not a randomized study, I would avoid words that imply causality, like "result," and use phrasing like "is associated with." But you have it right except for that.

So the interaction term of .009 is the impact that age has on digital literacy when the individual is female, so when "Q101" is set to 1.

This is not correct. The interaction term of .009 is the difference between the "impact" (in the log odds metric) of a unit increase of age in females and an increase of 1 unit in age in males. So the "impact" of a unit increase of age in females would actually be -0.018 + 0.009 = -.009.