BU442 Market Research Final Exam
May 16, 2013
6. There are four specific methods for implementing survey research. They include on-site surveys, telephone surveys, computer surveys, and mail surveys. Each has pros and cons, which are listed as follows: On-site surveys:
Pros—great for probing; best for non-vested markets; should be no more than 6 pages long; you can use physical props for demonstrations; higher general response rate at around 75%. Cons—people may not answer sensitive information questions correctly or at all (age, income, marital status); most costly of the 4 types; labor intensive; can’t go back for a follow-up because you usually don’t ask for address or phone number. Telephone surveys:
Pros—central location; less costly now than in the past due to lower phone rates; data is collected quickly; cost is only about 40% of on-site or mail surveys; people are anonymous so they are more likely to share sensitive information. Cons—some investment required; not able to physically test products. Computer-based surveys:
Pros—inexpensive way to reach a lot of people; sensitive questions are more likely to be answered; respondents can take their time; faster data collection; ability to track respondents Cons—people may ignore requests for surveys thinking they are spam; technical problems may arise. Mail surveys:
Pros—good for sensitive questions because most don’t require a name or address; bulk mail can be used to keep costs down. Cons—internet surveys decreasing mail survey importance; can be costly due to postage, printing and paper costs; not recommended if the survey has a time constraint.
5. There are two major sources of errors in survey research: random sampling errors and systematic errors. A random sampling error is a statistical fluctuation that occurs because of chance variation in the elements selected for a sample. A systematic error is error resulting from some imperfect aspect of the research design that causes respondent error or from a mistake in the execution of the research. Other errors include respondent error, which is a category of sample bias resulting from some respondent action or inaction such as nonresponse or response bias. A nonresponse error is the statistical differences between a survey that includes only those who responded and a perfect survey that would also include those who failed to respond. Nonrespondents are people who are not contacted or who refuse to cooperate in the research. No contacts are people who are not at home or who are otherwise inaccessible on the first and second contact. Refusals are people who are unwilling to participate in a research project. Self-selection bias is a bias that occurs because people who feel strongly about a subject are more likely to respond to survey questions than people who feel indifferent about it. Response bias occurs when respondents either consciously or unconsciously tend to answer questions with a certain slant that misrepresents the truth.
4. When using descriptive data, everyone involved is known. With inferential data everyone is unknown, with a finite number of people to draw a conclusion about, and the purpose is to make a judgment about a population. Population parameters are what is measured, for example price, and sampling statistics describe who is measured. Frequency distributions are frequency percentages (what is) and probability (what may be), which is the long-run relative frequency with which an event will occur. It is one of the most common ways to summarize a set of data. A significance level is a critical probability associated with a statistical hypothesis test that indicates how likely it is that an inference supporting a difference between an observed value and some statistical expectation is true. It is the acceptable level of Type I error.
3. The sampling frame is a list of elements from which a sample may be drawn. It is also called the working population because these units...
Please join StudyMode to read the full document