 |  Marketing Research: Within a Changing Information Environment, 2/e Joseph Hair,
Louisiana State University
Robert Bush,
University of Memphis
David Ortinau,
University of South Florida
Observation Techniques, Experiments, and Test Markets
Chapter SummaryDiscuss the characteristics, benefits, and weaknesses of observational techniques, and explain how these techniques are used to collect primary data.
Observational techniques are used by researchers in all types of research designs (exploratory, descriptive, causal). In addition to the general advantages of observation, major benefits are the accuracy of collecting data on actual behavior as it unfolds, reduction of confounding factors, and the amount of detailed behavioral data that can be recorded. The unique limitations of using observation methods are lack of generalizability of the data, inability of explaining current behaviors or events, and the complexity of setting and recording the behavior.
Describe and explain the importance of and differences between the variables used in experimental research design.
In order to conduct causal research, the researcher must understand the four key types of variables involving experimental designs (independent, dependent, extraneous, control) as well as randomization of test subjects and the role that theory plays in creating experiments. The most important goal of any experiment if to determine what, if any, relationships exist among different variables (independent, dependent). Functional, or cause-effect, relationships require systematic change in one variable as another variable changes.
Explain the theoretical importance and impact of internal, external, and construct validity measures in experiments and interpreting functional relationships.
Experimental designs are developed to control for contamination, which may serve to confuse the true relationship being studied. While a variety of issues exist regarding the concept of contamination, internal validity, external validity, and construct validity are at the center of discussion. Internal validity refers to the level of exact conclusions the researcher draws about a demonstrated functional relationship. The question is, are the experimental results truly due to the experimental variables? External validity is concerned with the interaction of experimental manipulations with extraneous factors causing a researcher to suspect the generalizability of the results to other settings. Construct validity is important in the process of correctly identifying and understanding both the independent and dependent variables included in an experimental design.
Several techniques unique to the experimental design are used to control for problems of internal and external validity. These techniques center on the use of control groups, pre-experimental measures, exclusion of subjects, matching subjects into groups and randomization of group members. These dimensions, built into the experimental design, provide true power for controlling contamination.
Discuss the three major types of experimental designs used in marketing research. Explain the pros and cons of using causal designs as a means of assessing relationship outcomes.
Pre-experimental designs fail to meet internal validity criteria due to a lack of group comparisons. Despite this weakness, three designs are still used quite frequently in marketing research: the one-shot study; the one-group, pretest-posttest design; and the static group comparison.
True experimental designs ensure equivalence between experimental and control groups by random assignment of subjects into groups. Three forms of true experimental designs exist: pretest-posttest, control group; posttest-only, control group; and the Solomon Four Group.
Quasi-experimental designs are appropriate when the researcher can control some of the variables but cannot establish true randomization of groups. While a multitude of these designs exist, two of the most common forms are the nonequivalent control group and the separate-sample, pretest-posttest.
Explain what test markets are, the importance and difficulties of executing this type of research designs, and how the resulting data structures are used by researchers and marketing practitioners.
Test Markets are a specific type of field experiment and are commonly conducted in natural field settings. Most common in the marketing research field are traditional test markets, controlled test markets, electronic test markets, simulated test markets, and virtual test markets. Data gathered from test markets provide both researchers and practitioners with invaluable information concerning customers' attitudes, preferences, purchasing habits/patterns, and demographic profiles. This information is very useful in predicting new product/service acceptance levels and advertising and image effectiveness, as well as in evaluating current marketing mix strategies. |
|
2003 McGraw-Hill Higher Education