This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Stat 651 (Winter 2011) Kaizar Homework 2 Solutions 40 points total Exercises 1. (6 points) This question has many problems. I first assume that the researcher has already determined that the respondent owns a vehicle. Even so, the question wording still contains a presupposition that the respondent does have regular service work performed on the vehicle. There is also a vague concept in the question where it asks about service work that the respondent has done on their vehicle. Many owners do some service work (such as changing oil) themselves. In this instance, should they include an oil change as part of their response? Another vague concept lies in the term usually. This term may mean different time periods to different people (e.g., annually vs. when it needs it). Retrieval failure may be an issue if the owner of the vehicle is not the person who cares for it. This question would be improved by a preceding question about knowledge of usual care. The closed categories are poor. They are neither complete nor clearly distinct. (e.g., Is oil different from fluid?) They also do not include an other option for write-in responses. Asking the respondent to choose one or two options will lead to difficult analysis. It may introduce recency problems in that the respondent may choose the first two that they read, rather than two that they perform most often, or make a choice on some other more important criteria. 2. (5 points) One of the options that this researcher may choose would be to conduct a focus group. She could gather six to twelve vehicle owners in a group. I do not think she will need several groups based on heterogeneity, since this survey does not seem to be about sensitive questions. It would be best if the researcher used a trained moderator to run her group by asking a series of open ended questions about her questionnaire. The discussion should be recorded. Using a focus group may be advantageous to her because she can get a lot of information in a short amount of time with relatively little expenditure. It is perhaps better than a field test in that she can receive active feedback about the question, rather than trying to infer this feedback from the answers her test respondents give. The field test will also allow her to get information from actual potential respondents that an expert may miss, and the group discussion of focus groups may bring out more issues than would be produced by individual-focused cognitive interviews. 3. (5 points) If the survey was conducted online, all the respondents would have access to the internet. This, of course, would make the internet-based sampling population on the whole more likely to look at news online, as opposed to a sampling population in which some do not 1 have internet access. This, in turn, would likely cause an overestimate of the proportion of people who get most of their news from the internet....
View Full Document
This note was uploaded on 07/26/2011 for the course STA 651 taught by Professor Kaizar during the Winter '11 term at Ohio State.
- Winter '11
- Sampling Methods