Chapter 8 – Causal Research Design: Experimentation

Chapter 8 – Causal Research Design: Experimentation

True/False Questions

1. Validity takes place when the occurrence of X increases the probability of the occurrence of Y.

False (moderate, page 212)

2. Concomitant variation occurs when the presumed cause and presumed effect are both present and both vary in a manner predicted by the researcher's hypothesis.

True (moderate, page 212)

3. The relevant cause/effect relationship is identified when both time order and concurrent variation conditions are satisfied.

False (difficult, page 213)

4. The process of manipulating one or more independent variables and measuring their effect on one or more dependent variables while controlling for the extraneous variables is called an experiment.

True (moderate, page 214)

5. According to the text, experiments can be described in terms of interdependent variables.

False (moderate, page 214)

6. Dependent variables are variables that are manipulated by the researcher and whose effects are measured and compared.

False (moderate, page 214)

7. Individuals, organizations, or other entities whose response to independent variables or treatments is being studied are called test units.

True (moderate, page 214)

8. Variables that measure the effect of the independent variables on the test units are called dependent variables.

True (moderate, page 214)

9. Consumers are good examples of dependent variables.

False (moderate, page 214)

10. Variables, other than the independent variables, that influence the response of the test units are called extraneous variables.

True (moderate, page 214)

11. According to the text, an experimental group is the group exposed to the manipulated independent variable.

True (moderate, page 215)

12. The control group is not exposed to the independent variable manipulation, but provides a point of comparison when examining the effects of these manipulations on the dependent variable.

True (moderate, page 215)

13. External validity refers to whether the manipulation of the independent variables or treatments actually caused the observed effects on the dependent variables.

False (moderate, page 216)

14. Internal validity refers to whether the cause-and-effect relationships found in the experiment remain the same when replicated in a larger population.

False (moderate, page 216)

15. "History" refers to the occurrence of events before the experiment.

False (moderate, page 217)

16. Extraneous variables represent alternative explanations of experimental results.

True (easy, page 218)

17. Randomization is the preferred method for controlling extraneous variables.

True (moderate, page 218)

18. Designs that do not control for extraneous factors by randomization are called true experimental design.

False (moderate, page 219)

19. Preexperimental designs are distinguished by the fact that the researcher can randomly assign test units to experimental groups and also randomly assign treatments to experimental groups.

False (moderate, page 219)

20. Designs that apply part of the procedures of true experimentation, while lacking full experimental control are called quasi-experimental designs.

True (moderate, page 219)

21. Statistical designs allow for the statistical control and analysis of external variables.

True (easy, page 219)

22. A preexperimental design in which a single group of test units is exposed to a treatment X, and then a single measurement on the dependent variable is taken, is called the one-shot case study.

True (difficult, page 219)

23. The random group is a preexperimental design in which a group of test units is measured before and after exposure to the treatment.

False (moderate, page 219)

24. The static group is a two-group experimental design in which one of the groups acts as a control group, the subjects are not assigned randomly, and measurement are made on both groups following the treatment.

True (moderate, page 221)

25. True experimental designs include the pretest-posttest control group design and the posttest-only control group design.

True (difficult, page 222)

26. An effect in which a prior measurement affects the test unit's response to the independent variable is called nonresponse bias.

False (moderate, page 223)

27. An experimental design in which the experimental group is exposed to the treatment but the control group is not and no pretest measure is taken is called the posttest-only control group design.

True (moderate, page 223)

28. The one-group pretest-posttest design is a quasi-experimental design that involves periodic measurements on the dependent variable for the group of test units.

False (moderate, page 224)

29. A time series design that includes another group of test units to serve as a control group is called a multiple time series design.

True (moderate, page 224)

30. A factorial design is a statistical experimental design used to measure the effects of two or more independent variables at various levels and to allow for interactions between variables.

True (moderate, page 225)

31. According to the text, factorial designs are the most common statistical designs.

True (moderate, page 225)

32. A field environment is an artificial setting for experimentation in which the researcher constructs the desired conditions.

False (moderate, page 226)

33. Responses given because the respondents attempt to guess the purpose of the experiment and respond accordingly are called demand artifacts.

True (moderate, page 226)

34. A laboratory environment involves measurement of behavior, attitudes, or perceptions in the environment in which they occur.

False (moderate, page 227)

35. The internal and external validity of field experiments conducted overseas is generally higher than in the US.

False (moderate, page 229)

Multiple Choice

36. ______________ takes place when the occurrence of X increases the probability of the occurrence of Y.

a. Probability

b. Causality (moderate, page 212)

c. Reliability

d. Validity

e. True scores

37. Concomitant ______________ occurs when the presumed cause and presumed effect are both present and both vary in a manner predicted by the researcher's hypothesis.

a. variation (moderate, page 212)

b. probability

c. reliability

d. validity

e. measurement

38. Which of the conditions must be satisfied in order to justify the inference of a causal relationship between two variables?

· Concomitant variation

· Time order of occurrence of variables

· Absence of other possible causal factors

iv. Sterile measurement environment

a. (i)

· (i), (iii) and (iv)

· (ii) and (iii)

· (i) and (iv)

· (i) (ii) and (iii) (moderate, pages 212-213)

39. The process of manipulating one or more independent variables and measuring their effect on one or more dependent variables while controlling for the extraneous variables is called a(n) ______________.

a. experiment (moderate, page 214)

b. test unit

c. hypothesis

d. theory

e. manipulation measurement

40. According to the text, experiments can be described in terms of all of the following EXCEPT:

a. independent variables.

b. interdependent variables. (moderate, page 214)

c. dependent variables.

d. extraneous variables.

e. b and d

41. ______________variables are variables that are manipulated by the researcher and whose effects are measured and compared.

a. Dependent

b. Interdependent

c. Extraneous

d. Independent (moderate, page 214)

e. Test unit

42. All of the following are good examples of independent variables EXCEPT:

a. price levels.

b. package designs.

c. advertising themes.

d. sales. (moderate, page 214)

e. c and d both

43. Individuals, organizations, or other entities whose responses to independent variables or treatments are being studied are called ______________.

a. hypotheses

b. independent variables

c. test units (moderate, page 214)

d. dependent variables

e. corollaries

44. All of the following are good examples of test units EXCEPT:

a. consumers.

b. price levels. (moderate, page 214)

c. stores.

d. geographic areas.

e. b and d both

45. Variables that measure the effect of the independent variables on the test units are called ______________.

a. interdependent variables

b. concurrent variables

c. extraneous variables

d. concomitant variables

e. dependent variables (moderate, page 214)

46. All of the following are good examples of dependent variables EXCEPT:

a. sales.

b. consumers. (moderate, page 214)

c. profits.

d. market shares.

e. c and d both

47. Variables, other than the independent variables that influence the response of the test units are called ______________.

a. interdependent variables

b. dependent variables

c. extraneous variables (moderate, page 214)

d. concurrent variables

e. influence variables

48. According to the text, a(n) ______________ is the group exposed to the manipulated independent variable.

a. control group

b. hypothetical group

c. experimental group (moderate, page 215)

d. extraneous group

e. corollary group

49. The ______________ is not exposed to the independent variable manipulation but provides a point of comparison when examining the effects of these manipulations on the dependent variable.

a. control group (moderate, page 215)

b. hypothetical group

c. experimental group

d. extraneous group

e. caveat group

50. ______________ refers to whether the manipulation of the independent variables or treatments actually caused the observed effects on the dependent variables.

a. Internal validity (moderate, page 216)

b. Extraneous validity

c. External validity

d. Control validity

e. Corollary group

51. ______________ refers to whether the cause-and-effect relationships found in the experiment remain the same when replicated in a larger population.

a. Internal validity

b. Extraneous validity

c. External validity (moderate, page 216)

d. Control validity

e. Population validity

52. An experiment was conducted to test the effects of coupon value on redemption. Personal interviews were conducted in New York with 280 shoppers who were entering a supermarket. These shoppers were given a coupon for one of four brands prior to entering the store. Two coupon values were used, one offering 15-cents off and the other 50-cents off. Shoppers were randomly assigned to these two coupon-value levels. Four brands--Tide detergent, Kellogg's corn flakes, Aim toothpaste, and Joy liquid dishwashing detergent were used. These same shoppers were re-interviewed upon leaving the store and asked to report on any coupons used in the store. What extraneous variable most likely will influence the results of this experiment?

· bivariate regression

· interdependence

· statistical regression

· history

· testing effects (moderate, page 217)

53. Which of the following is NOT an extraneous variable?

a. maturation

b. statistical regression

c. testing effects

d. selection bias

e. All of the above are extraneous variables. (moderate, pages 217-218)

54. Which of the following is NOT a way of controlling extraneous variables?

· randomization

· matching

· statistical control

· design control

· interactive testing (moderate, page 218)

55. Which of the following refers to the improper assignment of test units to treatment conditions?

· randomization

· matching

· overcontrolling

· statistical regression

· selection bias (moderate, page 218)

56. Which of the following is NOT one of the four broad categories of experimental designs?

a. concurrent design (moderate, page 219)

b. preexperimental design

c. quasi-experimental design

d. statistical design

e. true experimental

57. Designs that do not control for extraneous factors by randomization are called ______________.

a. true experimental designs

b. quasi-experimental designs

c. statistical designs

d. preexperimental designs (moderate, page 219)

e. extraneous designs

58. ______________ are distinguished by the fact that the researcher can randomly assign test units to experimental groups and also randomly assign treatments to experimental groups.

a. Preexperimental designs

b. True experimental designs (moderate, page 219)

c. Quasi-experimental designs

d. Statistical designs

e. Concurrent designs

59. A random sample of respondents would be distributed randomly, half to the experimental group and half to the control group. A pretest questionnaire would be administered to the respondents in both groups to obtain a measurement on attitudes toward Pepsi advertising, brand, and celebrity; only the respondents in the experimental group would be exposed to the TV program containing the Pepsi commercial. Then, a questionnaire would be administered to respondents in both groups to obtain posttest measures on attitudes toward Pepsi advertising, brand, and celebrity. Which of the following experimental design best represents the above scenario?

· one-shot case study

· one-group pretest-posttest design

· posttest-only control group design

· static group design

· pretest-posttest control group design (moderate, pages 222-223)

· Which of the following can be symbolized as:

EG: R X O1

CG: R O2

Where,

X = the exposure of a group to an independent variable, treatment, or event, the effects of which are to be determined

O = the process of observation or measurement of the dependent variable on the test units or group of units

R = the random assignment of test units or groups to separate treatments

CG = control group

EG = experimental group

· one-shot case study

· one-group pretest-posttest design

· posttest-only control group design (moderate, page 223)

· static group design

· Pretest-posttest control group design

61. In a Pepsi case the researcher is interested in examining the effect of humor and the effect of various levels of brand information on advertising effectiveness. Three levels of humor (no humor, some humor, and high humor) are to be examined. Likewise, brand information is to be manipulated at three levels (low, medium, and high). The resulting table would be three rows (levels of information) by three columns (levels of humor), producing nine possible combinations or cells. The respondents would be randomly assigned to one of the nine cells. Respondents in each cell would receive a specific treatment combination. After exposure to a treatment combination, measures would be obtained on attitudes toward Pepsi advertising, brand, and the celebrity from respondents in each cell.

To which category of design does this fall into?

· statistical regression design

· recombinant design

· static group design

· one-shot case study

· factorial design (moderate, page 225)

62. Designs that apply part of the procedures of true experimentation, while lacking full experimental control are called ______________.

a. concurrent designs

b. preexperimental designs

c. quasi-experimental designs (moderate, page 229)

d. statistical designs

e. partial designs

63. ______________ allow for the statistical control and analysis of external variables.

a. Preexperimental designs

b. True experimental designs

c. Quasi-experimental designs

d. Statistical designs (moderate, page 219)

e. External variable designs

64. Which of the following is NOT one of the three types of preexperimental designs?

a. the one-shot case study

b. the one-group pretest-posttest design

c. the random group (moderate, page 219)

d. the static group

e. c and d both

65. A preexperimental design in which a single group of test units is exposed to a treatment X and then a single measurement on the dependent variable is taken is called ______________.

a. the one-shot case study (moderate, page 220)

b. the one-group pretest-posttest design

c. the random group

d. the static group

e. the X-file

66. The ______________ is a preexperimental design in which a group of test units is measured before and after exposure to the treatment.

a. one-shot case study

b. one-group pretest-posttest design (moderate, page 221)

c. random group

d. static group

e. before-and-after study

67. The ______________ is a two-group experimental design in which one of the groups acts as a control group, the subjects are not assigned randomly, and measurements are made on both groups following the treatment.

a. one-shot case study

b. one-group pretest-posttest design

c. random group

d. partial experimental design

e. static group (moderate, page 221)

68. True experimental designs include ______________; ______________.

a. the one-shot case study and the pretest-posttest control group design

b. the pretest-posttest control group design and the posttest-only control group design (difficult, page 222)

c. the posttest-only control group design and the static group

d. the static group and the one-shot case study

e. the random study and the control study

69. An effect in which a prior measurement affects the test unit's response to the independent variable is called ______________.

a. nonresponse bias

b. interactive testing effect (moderate, page 223)

c. randomization error

d. interviewer bias

e. preliminary measurement effect

70. An experimental design in which the experimental group is exposed to the treatment but the control group is not and no pretest measure is taken is called the ______________.

a. one-shot case study

b. random group

c. static group

d. manipulated treatment study

e. posttest-only control group design (moderate, page 223)

71. The ______________ is a quasi-experimental design that involves periodic measurements on the dependent variable for the group of test units.

a. time series design (moderate, page 224)

b. one-group pretest-posttest design

c. factorial design

d. posttest-only control group design

e. quasi-dependent variable design

72. A time series design that includes another group of test units to serve as a control group is called a ______________.

a. factorial design

b. multiple time series design (moderate, page 224)

c. posttest-only control group design

d. one-shot case study

e. alternative group design

73. A ______________ is a statistical experimental design used to measure the effects of two or more independent variables at various levels and to allow for interactions between variables.

a. multiple time series design

b. posttest-only control group design

c. factorial design (moderate, page 225)

d. one-shot case study

e. interaction design

74. According to the text, which of the following are the most common statistical designs?

a. factorial designs (moderate, page 225)

b. multiple time series designs

c. posttest-only control group designs

d. one-shot case studies

e. regular matrix design

75. A ______________ is an artificial setting for experimentation in which the researcher constructs the desired conditions.

a. statistical environment

b. laboratory environment (moderate, page 226)

c. field environment

d. virtual environment

e. construct condition design

76. Responses given because the respondents attempt to guess the purpose of the experiment and respond accordingly are called ______________.

a. nonresponse bias

b. environmental constraints

c. demand artifacts (moderate, page 226)

d. sample controls

e. guessing bias

77. A ______________ involves measurement of behavior, attitudes, or perceptions in the environment in which they occur.

a. statistical environment

b. laboratory environment

c. field environment (moderate, page 227)

d. virtual environment

e. perceptual environment

78. Field experiments feature which of the following?

· Realistic environment

· Low control

· High external validity

· a and c only

· a, b, and c (difficult, page 227)

79. Limitations of experiments include which of the following?

· time

· cost

· administration of an experiment

· a and b

· a, b, and c (moderate, pages 227-228)

80. ______________ is an application of a controlled experiment done in limited, but carefully selected, test markets.

· Random sampling

· Parallel sampling

· Experimentation

· Selective control sampling

· Test marketing (moderate, page 228)

81. Which of the following conditions must be met before causal inferences can be made that X is a cause of Y?

· concomitant variation

· time-order occurrence of variables

· elimination of other causal factors

· a and b only

· a, b, and c (moderate, page 229)

Essay Questions

82. In a short essay, define internal validity and external validity and discuss the difference between each type of validity.

Answer

· Internal validity refers to whether the manipulation of the independent variables or treatments actually caused the observed effects on the dependent variables. Internal validity is threatened when the influences of extraneous variables are mixed with the independent variables. Without proper control of the extraneous variables, the researcher is unable to isolate the effect of the independent variable and thus cannot establish internal validity.

· External validity refers to whether the cause-effect relationships found in the experiment remain the same when replicated in a larger population. Threats to external validity arise when the experiment is conducted in an unrealistic manner, limiting the ability to generalize. This occurs when the experimental conditions do not account for factors likely to be encountered in the real world.

(moderate, pages 216-217)

83. In a short essay, describe the four broad categories of experimental designs and discuss the various applications of each design category.

Answer

· Preexperimental designs are designs that do not use randomization to control for extraneous factors. Thus, they suffer from many threats to internal and external validity. However, with a proper note of their limitations, they can add value when used in an exploratory fashion.

· True experimental designs are designs in which the researcher can randomly assign subjects and experimental groups. Therefore, these designs provide a larger degree of control over extraneous variables. Included in this category are the pretest-posttest control group design and the posttest-only control group design.

· Quasi-experimental designs are designs that result when the researcher is unable to fully manipulate the independent variables or treatments but can still apply part of the apparatus of true experimentation. These designs are typically employed in natural environments, enabling some degree of experimental control in a natural setting.

· Statistical designs are series of basic experiments that allow for statistical control and analysis of external variables. Statistical designs are classified on the basis of their characteristics and use. The important statistical designs include factorial designs.

(difficult, page 219)

84. In a short essay, list and discuss the three specific preexperimental designs.

Answer

· One-shot case study is a preexperimental design in which a single group of test units is exposed to a treatment X, and then a single measurement on the dependent variable is taken. This type of design is constructed using a nonrandom sampling process in which the subjects are self-selected or selected arbitrarily by the researcher. Without randomization, the observed dependent variables are subject to the influences of several extraneous variables.

· One-group pretest-posttest design is a preexperimental design in which a group of test units is measured before and after exposure to the treatment. Although this design is considered better than a case study, the validity of conclusions is questionable since extraneous variables are largely uncontrolled due to lack of randomization and a control group.

· Static group is a two-group experimental design in which one of the groups acts as a control group. Only one group, the experimental group, receives the experimental treatment. The subjects are not assigned randomly, and measurements are made on both groups following the treatment. The treatment effect would be measured as the difference between the control and experimental group.

(moderate, pages 220-222)

85. In a short essay, define randomization and list and discuss the two types of true experimental design.

Answer

Randomization is one method of controlling extraneous variables that involves randomly assigning test units to experimental groups by using random numbers. Treatment conditions are also randomly assigned to experimental groups. Randomization is the preferred procedure for ensuring the prior equality of experimental groups.

· Pretest-posttest control group design is an experimental design in which the experimental group is exposed to the treatment but the control group is not. Pretest and posttest measures are taken on both groups. Test units are randomly assigned. The randomization of this design controls for most extraneous variables. The extraneous effects are presumed to be equally represented in both the control and experimental groups.

· Posttest-only control group design is an experimental design in which the experimental group is exposed to the treatment but the control group is not and no pretest measure is taken. Test units are randomly assigned. One significant advantage of this design over the pretest-posttest control is the elimination of the interactive testing effect that comes with pretesting. Additionally, the simplicity of this design offers time, cost, and sample size advantages.

(moderate, pages 222-223)

86. In a short essay, discuss the difference between laboratory and field experimentation.

Answer

· A laboratory environment is an artificial one, which affords the greatest amount of control over the crucial factors involved in the study. Compared to field experiments, the artificial nature of laboratory environments may cause reactive error in that the respondents react to the situation itself rather than to the independent variable. On the positive side, laboratory experiments do allow for more complex designs than field experiments.

· A field experiment involves measurement of behavior, attitudes, or perceptions in the environment in which they occur. The researcher has much less control over extraneous variables that may affect internal validity. However, if internal validity can be maintained, the results may generalize more easily than those obtained in a laboratory setting.

(easy, pages 226-227)

87. In a short essay discuss the different types of extraneous variables and their effects on experiments in marketing research.

Answer

The discussion on extraneous variables should focus on the following categories of extraneous variables: history, maturation, testing, instrumentation, statistical regression, selection bias, and mortality.

History

Contrary to what the name implies, history (H) does not refer to the occurrence of events before the experiment. Rather, history refers to specific events that are external to the experiment but occur at the same time as the experiment. For example, general economic conditions may decline during the experiment, thereby contaminating the post-treatment observation. The longer the time interval between observations, the greater the possibility that history will confound an experiment.

Maturation

Maturation (MA) is similar to history except that it refers to changes in the test units themselves. These changes are not caused by the impact of independent variables or treatments but occur with the passage of time. In an experiment involving people, maturation takes place as people become older, more experienced, tired, bored, or uninterested. Tracking and market studies that span several months are vulnerable to maturation, since it is difficult to know how respondents are changing over time.

Testing Effects

Testing effects are caused by the process of experimentation. Typically, these are the effects on the experiment of taking a measure on the dependent variable before and after the presentation of the treatment. For example, the respondents try to maintain consistency between their pre- and post-treatment attitudes. As a result of the testing effect, post-treatment attitudes may be influenced more by pretreatment attitudes than by the treatment itself.

Instrumentation

Instrumentation (I) refers to changes in the measuring instrument, in the observers or in the scores themselves. Sometimes, measuring instruments are modified during the course of an experiment. Consider an experiment in which dollar sales are being measured before and after exposure to an in-store display (treatment). If there is a non-experimental price change between measurements, this results in a change in instrumentation because dollar sales will be measured using different unit prices. Instrumentation effects are likely when interviewers make pre- and post-treatment measurements. The effectiveness of interviewers can be different at different times.

Statistical Regression

Statistical regression effects (SR) occur when test units with extreme scores move closer to the average score during the course of the experiment. In the advertising experiment, suppose that some respondents had either very favorable or very unfavorable attitudes. On post-treatment measurement, their attitudes might have moved toward the average. People's attitudes change continuously. People with extreme attitudes have more room for change, so variation is more likely. This has a confounding effect on the experimental results because the observed effect (change in attitude) may be attributable to statistical regression rather than to the treatment (test commercial).

Selection Bias

Selection bias (SB) refers to the improper assignment of test units to treatment conditions. This bias occurs when selection or assignment of test units results in treatment groups that differ on the dependent variable before the exposure to the treatment condition. If test units self-select their own groups or are assigned to groups on the basis of the researchers' judgment, selection bias is possible. For example, consider a merchandising experiment in which two different merchandising displays (old and new) are assigned to different department stores. The stores in the two groups may not be equivalent to begin with. They may vary with respect to a key characteristic, such as store size. Store size is likely to affect sales regardless of which merchandising display was assigned to a store.

Mortality

Mortality (MO) refers to the loss of test units while the experiment is in progress. This happens for many reasons, such as test units refusing to continue in the experiment. Mortality confounds results because it is difficult to determine if the lost test units would respond in the same manner to the treatments as those that remain. Consider again the merchandising display experiment. Suppose that during the course of the experiment three stores in the new display treatment condition drop out. The researcher could not determine whether the average sales for the new display stores would have been higher or lower if these three stores had continued in the experiment.

(difficult, pages 217-218)

Дата добавления: 2015-11-04