Question

To find out whether violent crime rates differ by region, a researcher has regressed the violent crime rate (measured in acts per 100 000 residents) on a constant and regional dummy variables. How would you interpret the regression coefficient for the "East" dummy?

​

Multiple Regression for Violent Crime Rate
	Multiple	R-Square	Adjusted	StErr of
Summary	R		R-Square	Estimate
	0.4660	0.2172	0.1672	240.273231
	Degrees of	Sum of	Mean of	F-Ratio	p-Value
ANOVA Table	Freedom	Squares	Squares
Explained	3	752760.9105	250920.3035	4.3464	0.0088
Unexplained	47	2713367.599	57731.22552
	Coefficient	Standard	t-Value	p-Value	Confidence Interval 95%
Regression Table		Error			Lower	Upper
Constant	423.3076923	66.63980418	6.3522	< 0.0001	289.2456984	557.3696862
West	-97.86324786	104.1894626	-0.9393	0.3524	-307.4654109	111.7389151
East	-61.14102564	96.18627221	-0.6357	0.5281	-254.6428463	132.360795
South	200.7511312	88.52580383	2.2677	0.0280	22.66018515	378.8420773

	a.	The violent crime rate in the East is 61.14 acts per 100 000 residents lower than normal.
	b.	The violent crime rate in the East is 61.14 acts per 100 000 residents lower than the national average.
	c.	The violent crime rate in the East is 61.14 acts per 100 000 residents.
	d.	The violent crime rate in the East is 61.14 acts per 100 000 residents lower than in the North.

Answer 1