=========================preview======================
(mark222)[2009](s)exam1~PPSpider^_10377.pdf
Back to MARK222 Login to download
======================================================
MARK 222 Marketing Research
Spring 2009
Exam I Answer Key

Name (in block letter):
Student ID:

Total 150 points. One-hour exam. Be concise and clear in writing answers. There are 5 pages including this page. Make sure that your copy has all 5 pages.

Part I. 20 points for each
1. (i) Describe Controlled Test Market. (ii) Is it a laboratory experiment or a field experiment? (iii) Is it a disguised method or an undisguised method?
(i) It is a new product test using a forced-distribution channel operated by an outside research firm. The manufacturer of a new product does not have to sell it through a normal distribution channel in the controlled test market. (ii) It is a field experiment as the new product is purchased in a real world setting.(iii) Since consumers do not know that they are tested, it is a disguised method.
2. Explain the difference between Semantic Differential Scale and Likert Scale in term of (i) the purpose and (ii) the way the scales values are averaged (or summated).
(i)
Semantic differential scale measures the market level attitude toward an object along each attribute. Likert scale measures each individual consumers overall attitude toward an object.

(ii)
In semantic differential scale, scale values are averaged across individual consumers for each attribute of an object. In Likert scale, scale values are summated across attributes for each individual respondent.


3. (i) Is reliability a sufficient condition for the validity of a measurement? (ii) Is reliability a necessary condition for the validity of a measurement? (iii) Describe two major ways to assess reliability.
(i, ii) Reliability is a necessary condition for the validity. But not sufficient.
(iii) One way is to assess the stability of a measurement via test-retest method. The other way is to assess equivalence (or internal consistency) via split-half test of measurement questionnaire.
4. Discuss the differences between stratified sampling and cluster sampling in terms of (i) the motivation and (ii) the characteristics of subgroups.
(i)
Motivation: Stratified sampling is used when we want to make sure that the realized sample includes all types of consumers (which is not guaranteed in the simple random sampling) while cluster sampling is purported to reduce costs/efforts associated with data collection.

(ii)
Subgroups: In stratified sampling, consumers are homogeneous within a subgroup and heterogeneous across subgroups. It is opposite in cluster sampling.


5. Describe surrogate information error. Is it a random error or a non-random error? Would increasing sample size reduce this type of error?
Surrogate information error is the difference between the information required for a decision problem and the information being sought by the researcher. It is a non-random error. Large sample would not help in reducing the surrogate information error.

Part II. 50 points
Jane, the manager