ex3_2011ak - Sociology 63993 Exam 3 Answ er Key[Draft May...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
Sociology 63993—Exam 3—Page 1 S o c i o l o g y 6399 3 E x a m 3 A n s w e r K e y [ D r a f t ] M a y 1 0 & 1 2, 2011 I. True-False. (20 points) Indicate whether the following statements are true or false. If false, briefly explain why. 1. If a model fails the Brant test, Manova should be used instead. False. The Brant test is used after ordered logit models. If the model fails the test, you may want to use multinomial logit or generalized ordered logit models instead. 2. The probability of an event occurring is .5. This means that the odds of the event occurring are 1. True. 3. Unlike OLS regression, a Wald test in logistic regression requires that you estimate both the constrained and unconstrained models. False. As with OLS, you just estimate the unconstrained model. 4. A model with reciprocal causation is under-identified. One way to solve the problem is to add other variables from the data set that are totally uncorrelated with the variables that are already in the model. False. The added variables must have direct effects on some variables and indirect effects on others (which means they must be correlated with the variables already in the model.) 5. The dependent variable is coded 1 = Catholic, 2 = Protestant, 3 = Jewish, 4 = Muslim, 5 = Other. Because the DV has more than two categories but is not continuous, an ordered logit model is called for. False. Categories are not ordered so use mlogit. II. Short answer . (25 pts each, 50 pts total). Answer both of the following. II-1. (25 points): The New York Times (4/25/2011) recently reported that “Ever since Congress passed the federal gender- equity law known as Title IX, universities have opened their gyms and athletic fields to millions of women who previously did not have chances to play. But as women have surged into a majority on campus in recent years, many institutions have resorted to subterfuge to make it look as if they are offering more spots to women.” One of the most questionable practices is that some schools “are counting male practice players as women.” For example, Texas A & M, which recently defeated Notre Dame for the national championship in women’s basketball, reported that it had 32 players on its team, 14 of whom were men. Besides raising concerns about whether schools are trying to dodge their legal obligations under Title IX, some coaches are also wondering whether having male practice players on a team actually affects the team’s success. Therefore, data have been gathered on every Division I women’s sports team in the nation. The variables are
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Sociology 63993—Exam 3—Page 2 Variable Description tournament 1 = made the postseason tournament in the team’s sport, 0 = did not make tournament maleplayers 1 = school listed males as part of the women’s team, 0 = school did not list males as part of the team experience 1 = Head Coach has at least 5 years experience, 0 = coach has less than 5 years of experience salary Head Coach’s salary in thousands of dollars (centered to have a mean of 0)
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

Page1 / 8

ex3_2011ak - Sociology 63993 Exam 3 Answ er Key[Draft May...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online