Question

5 Ie there a relationship berween the race of violent offenders and the race of their victm Data from the U.S. Department of Justice for 2011 are presented belo Characteristics of Offender Characteristics of Victim Black Other White 448 2630 33 White 2,447 193 Black 45 180 99 Other Source U.S. Department of Justice, Expanded Homicide Data, Table 6, 2011 a Letk treat race of offenders as the independent variable and race of victims as the dependent variable. If we first ignore the independent variable and try to predict rnce of victim, how many errors will we make? b. If we now take into account the independent variable, how many errors of t diction will we make for those offenders who are white? Black offenders? Other offenders? . Combine the answers in (a) and (b) to calculate the proportic for this table based on the independent variable. How does this statistic improve our understanding of the relationship between the two variables? 6 Lets continue our analysis of offenders and victims of violent crime. In the following table U.S. Department of Justice 2011 data for the sex of offenders and the sex of victims reported pre- nal reduction in error LS OF SOCIAL STATISTICS FORA DIVERSE SOCIETY

question 5

Answer 1

Charcteristics of Victim	White	Black	Other
White	2,630	448	33	3,111
Black	193	2,447	9	2,649
Other	180	45	99	324
	3,003	2,940	141	6,084

a)
E1 = total sample size - largest row total
= 6084 - 3111
= 2973
b)
E2 = sum of the differences between the largest cell count and its column total
3003 - 2630 = 373
2640 - 2447 = 193
141 - 99 = 42

E2 = 373 + 193 + 42 = 608

c)
PRE = (E1 - E2)/E1
= (2973 - 608)/2973
= 0.795492

A lambda (PRE) of .795 indicates that we would make 79.5% fewer errors in predicting the race of the victim using the race of the offender than not using it.
A lambda of .795 indicates a strong association between these two variables.