Interpretation of a variable within an interval

[an nhu do]

Hi everyone,

I am doing a research on the impact of institutions income level. The proxy used for institutions is Rule of law estimates from the world bank. Income level is measured by gdp per capita (constant 2010 US$)

I was wondering how should I best interpret this coefficient because Rule of law is in the interval of -2.5 and 2.5? If I say " A unit increase in Control of corruption by 1 unit leads to an increase in income by 91 percent", wouldn't a 1 unit increase extreme in the variable is confined within an interval.

I am looking forward to your advice. Thank you in advance

Edited: This is an example that I practiced on which is from Rodrik et al. (2004)

[Rich Goldstein]

you don't tell us anything about what the data actually look like (please see the FAQ esp re -dataex-); I guess that you have a categorical variable that you are using factor variable notation to estimate a set of indicator (dummy) variables; if there are several then we are talking about a comparison to whatever the reference range is; a completely different guess would be that you have coded the variable somehow and enter it into your estimation procedure as a quantitative variable treated linearly - then the interpretation depends on the coding system used; so, please see the FAQ on how to has questions more clearly

added: include, as the FAQ requests, your exact command and the output (within CODE delimiters)

[Marcos Almeida]

Just an additional note: income tends to perform better under glm. Logically, it is necessary to do the modeling, but a gamma family with a log link can be an interesting starting point. That being said, by seeing the name of the variables, it seems there might be some hierarchical structure of the data...

[Carlo Lazzaro]

An:
welcome to this forum.
As an aside to Rich's helpful advice, please note that FAQ reminds welcomed newcomers to consider the multidisciplinary flavour of this forum (which is one of its greatest advantage). Obviously, this comes at a cost of keeping in mind that contributions that sound familiar to you might be perfectly obscure to other listers who are engaged in different research fields.
Hence, please provide full reference (aka, "may the source be with you!) Thanks.

[an nhu do]

thank you for all for your reminders.
The data I practiced with is from

Rodrik, D., Subramanian, A. and Trebbi, F., 2004. Institutions rule: the primacy of institutions over geography and integration in economic development.

Journal of economic growth, 9 (2), pp.131-165.