hw10 - Soc 63993, Advanced Social Statistics II Homework...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
Homework #10—Year in Review Special Edition—Page 1 Soc 63993, Advanced Social Statistics II Homework No. 10 Year in Review Special Edition This homework is a mini-review of much of the material you have covered throughout the year (including last semester). In a few instances, the “best” answer requires advanced techniques we have only briefly or recently mentioned. Skimming through the syllabi from both semesters and the advanced section of the readings packet (or at least my summary of those readings) may help you. I. For each of the following circumstances describe the statistical technique you would use for revealing the relationship between the dependent and independent variables. Write a few sentences explaining and justifying your answer. Part of your answer may include an explanation of why other apparent alternatives would not be as good. In some instances more than one technique may be reasonable. EXAMPLE: The President of Notre Dame wants to see whether anything can be done about student drinking. He intends to have a random sample of 25 students fill out a questionnaire reporting on how much alcohol they drank in the last month. The students will then see a film on alcohol abuse. A month later they will again report on how much alcohol they drank in the last month. SOLUTION: A matched pairs T-Test is called for. Individuals are interviewed before and after seeing the film, with the goal of seeing whether the amount they drink has changed. An independent samples T-Test would not be appropriate since the same people are measured twice. Such a T-test would be appropriate if, say, half the subjects were shown the film and half were not. EXAMPLE: For a sample of 10,000 births you know whether the baby died in the first year of life, its birth weight, mother's age and education, and its race. How would you assess the influence of these variables on infant mortality? SOLUTION: Since the dependent variable is a dichotomy, Logistic regression would probably
Background image of page 1

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

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 4

hw10 - Soc 63993, Advanced Social Statistics II Homework...

This preview shows document pages 1 - 2. Sign up to view the full document.

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