Free Version

Upgrade subject to access all content


Interpreting Parameter Estimates - Crimes


Suppose you estimate a model of the number of crimes in a city and the number of police officers in the city, and you estimate this model for cities in the Unites States and in Great Britain. In the United States, you find a negative effect of police officers on crimes and the effect is statistically significant at a $1\%$ level.

You find a similar negative effect in Great Britain, but it is only significant at a $10\%$ level.

What can you conclude?


Both the U.S. and Great Britain should increase their police force to reduce crime, because both models showed negative and statistically significant impacts.


It is not possible to conclude whether police force should be increased.


Only the U.S. should increase their police force, because the effect has a higher statistical significance than for Great Britain.


It is not possible to conclude whether police force should be increased because the estimated marginal effects are not statistically significant in both countries.


There are fewer crimes in Great Britain, so the effect of additional police officers in that country is less pronounced.