Statistics in Criminal Justice:

1. Which of the following is not a criticism of logit regression? A. The coefficients are based on the values of the independent variables B. Coefficients can be biased for small samples C. Multicollinearity D. Influential cases 2. Which of the following procedures is not suitable for limited dependent variables? A. OLS regression B. Logit regression C. Poisson regression D. Tobit regression E. None of these is correct. 3. Overdispersion occurs when: A. the variance is equal to the mean. B. the standard deviation is equal to the mean. C. the variance is greater than the mean. D. the variance is less than the mean.

4. Tobit regression can deal with censored or truncated data. a. True b. False

5. Poisson and negative binomial regression analyses both deal with count data. a. True b. False 6. Nagelkerke R and the Cox and Snell R are equivalent to the R in OLS regression. a. True b. False

7. Multinomial logit regression employs a series of dichotomous logit regression analyses. a. True b. False

8. The test of the significance of a logit model is done by a likelihood ratio test. a. True b. False