Chapter 12 Experimental Research 110

Chapter 12

Experimental Research

True-False Questions

T 1. The purpose of experimental research is to allow the researcher to control the situation
257 so that causal relationships among variables may be studied.

F 2. In a well-designed experiment, the control group and the experimental group consist of
261 the same people.

T 3. Test units may be people, stores, departments, or sales territories.

262

F 4. A laboratory experiment is an experiment conducted in a natural setting, often for a

270 long period of time.

F 5. A field experiment is an experiment conducted in an artificial setting that gives the

270 researcher almost complete control over the research setting.

T 6. A quasi-experimental design is an experimental design that fails to control adequately

275 for loss of external or internal validity.

T 7. We would diagram a one-shot design as follows: X O1.

275

F 8. A typical quasi-experimental design might be diagrammed like this:

278
Experimental group R X O1
Control group R O2

F 9. The most common way to prevent test units from differing from one another on key
264 variables is to use a matching technique.
Chapter 12 Experimental Research 111

F 10. An experiment diagram as follows would be a static group design:

276
Experimental group 1 R O1 X O2
Control group 1 R O3 O4
Experimental group 2 R X O5
Control group 2 R O6

T 11. The before-after with control group is the classic experimental design and looks like
277 this:

Experimental group R O1 X O2
Control group R O3 O4

T 12. Debriefing refers to the process of providing subjects with all pertinent facts about the
268 nature and purpose of the experiment after its completion.

T 13. Demand characteristics are a source of constant error in experiments.

265

F 14. It is not possible to assess the effects of more than one independent variable in a single
281 experiment.

T 15. The most prominent demand characteristic in most experiments is the person who
265 administers the experimental conditions..

F 16. In factorial experimental designs, a single independent variable is changed to observe

283 its effect on a single dependent variable.

T 17. Interaction is said to occur when the effect of one treatment is different at different
284 levels of the other treatment.

F 18. In basic experimental designs, the interaction of two independent variables can be
275 studied.

F 19. Field experiments are conducted in laboratory settings.

270

F 20. A factorial design has two or more dependent variables.

283

F 21. External validity has to do with the determination of whether the independent variable
Chapter 12 Experimental Research 112

273 was the sole cause of the changes in the dependent variable.

F 22. A major advantage of the static group design is that we have some assurance that the
276 groups were equal on the variables of interest before the experimental group received
the treatment.

T 23. The effect of testing can only occur in "before-after" experiments.

272

T 24. One of the experimental researcher's main goals is to identify the possible extraneous
265 variables other than the treatment that may affect the results of the experiment and to
eliminate or control them.

T 25. Blinding refers to the control of subjects' knowledge of whether or not they have been
267 given a particular experimental treatment, such as not informing the subjects whether
they have the diet or regular formulation of a cola drink.

T 26. Internal validity has to do with the interpretation of the cause-and-effect relationship in
271 an experiment.

T 27. In internal validity, testing is also called the pretesting effect.

272

T 28. Factorial designs allow for the testing of the effects of two or more treatments at
283 various levels.

F 29. In a business experiment, the researcher manipulates all the relevant, extraneous
260 variables and records the impact on the independent variable.

T 30. The major difference between experimental research and other research is the
257 experimenter's ability to "hold constant" conditions and to manipulate the independent
variable.

F 31. Internal validity has to do with the ability of the researcher to generalize the results of
271 the study to the external environment.

F 32. Naturalistic field experiments tend to have less external validity than laboratory
274 experiments.

T 33. Often the dependent variable selection process, like problem definition, is not as
262 carefully considered by the researchers as it should be.

T 34. The experimenter's choice of a dependent variable determines what type of answer is
Chapter 12 Experimental Research 113

262 given to the research question.

F 35. Quasi-experimental designs are a type of true experimental design.

275

T 36. Random assignment of subjects to various experimental groups is the most common
267 technique used to avoid the problem of subjects differing from each other on key
variables.

T 37. In one-shot designs, there is no way to control extraneous variables.

275

T 38. The term "demand characteristics" refers to experimental design procedures that
265 unintentionally give hints to subjects about the experimenter's hypothesis.

T 39. The one-group pretest-posttest design allows comparisons of the same subjects both
275 before and after training in a human relations study.

T 40. If two different taste formulations of an orange juice are marked in cans with either a
267 triangle or a square and neither the subjects nor the experimenters know which taste
formulation is in which can, the experiment is said to be double-blinded.

T 41. The first step in a true experimental design is to assign subjects randomly to
277 experimental (and control) conditions.

T 42. In a basic experimental design, a single independent variable is manipulated to observe

269 its effect on another single dependent variable.

F 43. External validity refers to the question of whether the experimental treatment was the
273 sole cause in observing changes in the dependent variable.

F 44. Internal validity is the quality of being able to generalize beyond the data of the
271 experiment.

T 45. In many studies, the use of student subjects as surrogate businesspeople causes a
274 problem of external validity.

T 46. Naturalistic field experiments tend to have greater external validity than artificial
274 laboratory experiments.

F 47. The "Hawthorne effect" demonstrates that subjects do not normally perform differently
266 when they know they are experimental subjects.
Chapter 12 Experimental Research 114

F 48. An experimenter should avoid disguising the purpose of the experiment because
266 participants will then not know how to be good subjects to help confirm the hypothesis.

T 49. The Solomon Four-Group design combines the before-after with control group design
279 with the after-only with control group design.

T 50. Field experiments are generally utilized to "fine tune" business strategies and to
274 determine sales volume.

T 51. An experiment is a research technique allowing researchers to control the research

257 situation so that causal relationships among variables may be evaluated.

T 52. An independent variable may be changed in value or altered independently of any

257 other variable.

F 53. In the pretest-posttest with control group design, all groups are exposed to the
277 experimental treatment.

T 54. Experimentation may require that an artificial situation be created that is not exactly
270 like the purchasing situation.

T 55. If the independent variable is defined as levels of advertising dollar expenditures, the
260 independent variable is a continuous variable.

F 56. Alternative manipulations of the dependent variable are called experimental treatments.
260

F 57. The Solomon Four-Group design is frequently used in business research.

279

T 58. The independent variable is hypothesized to be the causal influence on the dependent
257 variable..

T 59. Supermarkets, sales territories, or people may be test units.

262

T 60. The random assignment of experimental treatments to the different experimental

268 groups might be called "control by chance."

T 61. Suppose a before-after experiment was being conducted to test a new and very unusual
272 packaging strategy. If the unusual package receives much publicity from the broadcast
media, a history effect influences internal validity.
Chapter 12 Experimental Research 115

F 62. A history effect jeopardizing internal validity might be a group of subjects in the
272 experimental group getting hungry or bored because of the long duration of the
experiment.

T 63. The mortality effect jeopardizing internal validity may occur if many subjects
273 withdraw from one experimental treatment group and not from the control group.

T 64. The problem of mortality in internal validity may also be referred to as sample
273 attrition.

T 65. A loss of internal validity due to maturation might likely occur in an experiment of a
272 new compensation program for sales personnel that lasts over a one-year period of
time.

T 66. In the one-shot design, there is no means for controlling extraneous influences.
275

T 67. Interaction effects are assumed to be either minimal or nonexistent in Latin square
287 designs.

F 68. The one-shot design controls for history effects, but not any other sources of
275 extraneous variation.

T 69. Assignment errors can be controlled by the use of a matching technique on the basis of
264 important background information..

T 70. A static-group design is an after-only design in which subjects in the experimental

276 group are measured after exposure to the experimental treatment, and the control group
is measured without exposure to the experimental treatment; no premeasure is taken.

Multiple Choice Questions

C 1. “Does the size of the package affect the amount of the product which is consumed?” is
256 a question that should best be studied through:
A. exploratory research.
B. descriptive research.
C. experimental research.
D. focus group research.

A 2. An experimenter has some degree of control over the:

Chapter 12 Experimental Research 116

257 A. independent variable.

B. correlative variable.
C. history effect.
D. All of the above, if the experiment is conducted properly.

C 3. An experimental test of a new outboard motor lubricant gave misleading results

263 because one of the test sites was Florida, where the lubricant was never exposed to cold
winter weather. This is an example of:
A. test unit failure.
B. the instrumentation effect.
C. sample selection error.
D. O1 X O2 error.

B 4. If subjects change their normal behavior in order to cooperate with the experimenter,
265 affecting the results of the experiment, this is called:
A. the double-blind effect.
B. the guinea pig effect.
C. experimenter bias.
D. the interaction effect.

B 5. If a researcher wants to answer the question: “Will adult women (ages 25-49) use
256 more of a cooking oil when it is in a full container or a half-empty one?”, the best type
of study to use would be a(n):
A. descriptive study.
B. experimental study.
C. exploratory study.
D. focus group study.

D 6. After an experiment is completed, a researcher explains the nature and purpose of the
268 experiment to all the subjects. This is called:
A. blinding.
B. instrumentation.
C. compromise design.
D. debriefing.

A 7. In a basic experimental design, variable(s) is/are manipulated.

275 A. one
B. two
C. three to five
D. ten or more

D 8. The one-group pretest-posttest design:

Chapter 12 Experimental Research 117

276 A. is an example of a “good” experimental design.

B. uses a control group.
C. the influence of the experimental treatment is: O2.
D. does not control for history.

A 9. The unit price example in the text showed that a change in the presentation of unit
257 price information (the variable) had caused a change in the average unit price paid
(the variable).
A. independent, dependent
B. experimental, independent
C. dependent, independent
D. research, control

B 10. Decisions must be made about several basic elements of an experiment. One of these
262 issues is:
A. manipulation of the extraneous variable(s).
B. selection and measurement of the dependent variable.
C. control over the dependent variable.
D. selection of assignment of extraneous variables.

B 11. The purpose of most business research experimentation is to measure and compare the
262 effects of the experimental treatments on the variable.
A. independent
B. dependent
C. control
D. extraneous

B 12. One of the following research methods differs from other methods because it gives the
257 researcher a high degree of control over the research situation. Which is it?
A. Surveys
B. Experiments
C. Observation studies
D. Secondary data studies

D 13. If we are experimenting with different forms of advertising copy appeals, the
262 dependent variable might be defined as:
A. a measure of advertising awareness.
B. recall.
C. change in brand preference.
D. any of the above.
Chapter 12 Experimental Research 118

C 14. If subjects in one experimental group are always administered the treatment in the
264 morning, while subjects in a second experimental group are always administered the
treatment in the afternoon, then error has occurred.
A. sampling
B. random experimental
C. constant experimental
D. treatment

A 15. The result of a constant error:

264 A. is to distort the experimental results in a particular direction.
B. is not to distort the experimental results in any particular direction, but to
obscure them.
C. is neither to distort nor obscure the experimental results.
D. has no effect on experimental results.

B 16. In a human resources training study, “manager turnover” is most likely to be a(n):
262 A. independent variable.
B. dependent variable.
C. control group variable.
D. random assignment variable.

D 17. Which of the following is NOT a major factor influencing internal validity?
271 A. History
B. Mortality
C. Instrumentation
D. Experimentation

D 18. The influence of specific events in the external environment that occur between the
272 first and second measurements and are beyond the control of the experimenter is
referred to as:
A. selection.
B. testing.
C. maturation.
D. history.

B 19. If, during the course of an experiment, subjects change in some way that will have an
272 impact on the experimental results, the effect is referred to as:
A. history.
B. maturation.
C. selection.
D. mortality.

A 20. Students taking achievement tests for a second time usually do better than they did the
Chapter 12 Experimental Research 119

272 first time. This is an example of:

A. testing effect.
B. maturation effect.
C. history effect.
D. none of the above.

D 21. In a test marketing study of a new type of instant tea, sales volume (as measured by
262 dollar sales) is most likely to be a(n):
A. random variable.
B. independent variable.
C. control variable.
D. dependent variable.

C 22. In a test marketing study, initial trial purchase is most likely to be a(n):
262 A. independent variable.
B. control variable.
C. dependent variable.
D. random variable.

B 23. If a study of factory workers finds that workers perform better on a job assembly task
if
267 they had been exposed to another type of job assembly task earlier in the study, the
study is said to have which type of bias?
A. Experimenter bias
B. Order of presentation bias
C. Interviewer bias
D. Double-blind bias

A 24. According to the symbols used in describing experimental designs,

275 A. X is the application of the experimental treatment.
B. O is the observation of the independent variable.
C. vertical distances represent movements in time.
D. subscripts refer to different test units.

C 25. In the unit-pricing experiment described in the textbook, the independent or

257 experimental variable was:
A. store type.
B. average-price-paid-per-unit purchased.
C. unit price format (shelf tag vs. list).
D. city.
Chapter 12 Experimental Research 120

B 26. In the unit-pricing experiment described in the textbook, the dependent variable was:
257 A. list format (shelf tag vs. list).
B. average-price-per-unit.
C. stores.
D. the number of weeks.

D 27. In the unit-pricing experiment described in the textbook, if a competitor introduced a

272 two-for-one sale in this mid-western city, the experiment would have had problems of:
A. maturation.
B. testing.
C. mortality.
D. history.

C 28. As experiments _____ in naturalism, they approach pure _____ experiments, and
274 as they become _____ artificial, they approach _____ experiments.
A. increase; field; less; laboratory
B. decrease; field; less; laboratory
C. increase; field; more; laboratory
D. decrease; field; more; laboratory

A 29. In terms of internal validity, a large layoff of middle-level managers during a research
272 study would have what type of extraneous variable?
A. History
B. Maturation
C. Testing
D. Instrumentation

D 30. If the interviewers are changed during a research study, this would have what type of
273 effect on the extraneous variables?
A. Testing effect
B. History effect
C. Selection bias
D. Instrumentation

D 31. A static group design:

277 A. uses two experimental groups.
B. assigns subjects randomly.
C. does not eliminate the “volunteer effect.”
Chapter 12 Experimental Research 121

D. is often applied in "use tests" for new products.

C 32. The before-after design with control group is least susceptible to problems of:
278 A. selection bias.
B. history effect.
C. testing effect.
D. maturation.

C 33. The after-only with control group design eliminates the problems of:
278 A. selection bias and mortality.
B. mortality and testing.
C. testing and instrumentation.
D. selection bias and testing.

B 34. Which of the following requires randomization of subject assignment?

277 A. One-group pretest-posttest design.
B. Pretest-posttest control group design.
C. All of the above.
D. None of the above.

A 35. An automobile dealer, after a very cold winter, found himself with a very large
275 inventory. He experimented with offering a free trip to Chicago (X) with every car
sold. He recorded sales (O) after the promotion. This is an example of:
A. a one-shot (after only) design.
B. a static group design.
C. a completely chance design.
D. a completely randomized design.

C 36. An experiment that investigates long-term structural change is:

280 A. the static group design.
B. the Solomon four-group design.
C. a time series design.
D. a Latin square design.

C 37. An experimental researcher manipulated advertising (high versus low) and price (high
284 versus low). When advertising was high and price was high there was not much of an
influence on sales volume, but when advertising was high and price was low, there was
a great influence on sales volume. This indicates:
A. a main effect for advertising.
Chapter 12 Experimental Research 122

B. a main effect for price.

C. an interaction effect.
D. a factorial effect.

C 38. A researcher wishes to simultaneously investigate the package colors blue and green
284 and price levels of $20, $30, and $40 in a factorial experiment. How many cells or test
markets will be required?
A. two
B. five
C. six
D. three

B 39. A major assumption of the Latin square design is:

287 A. all variables have multiple levels.
B. interaction effects are nonexistent.
C. the number of treatments is limited to three.
D. there can be no more than one confounding extraneous variable.

A 40. Employee turnover during a research study has the most effect on which of the
273 following factors which affect the internal validity of the study?
A. Mortality
B. Testing
C. Instrumentation
D. History

C 41. A major advantage of the is its ability to measure the interaction effect among two
284 or more treatments at various levels.
A. completely randomized design
B. randomized block design
C. factorial design
D. quasi-randomized design

C 42. If the researcher is attempting to answer the question: “Will the test market results in
273 Kansas City hold true in California?”, the researcher is most concerned with:
A. the history effect.
B. internal validity.
C. external validity.
D. mortality.

B 43. If age is expected to influence savings behavior, a bank conducting an experiment may
264 have the most assurance that there are no intersubject differences if subjects in all
experimental groups:
Chapter 12 Experimental Research 123

A. are customers of the bank.

B. are matched on age.
C. are given a pretest.
D. all of the above.

A 44. When a prior measurement affects a test unit's response to the experiment, we might
272 say that occurs.
A. testing effect
B. sensitizing
C. prioritizing
D. maturation

A 45. In general, field experiments have _____ internal validity and _____ external validity
274 than laboratory experiments.
A. less; more
B. less; less
C. more; more
D. more; less

B 46. The characteristic that differentiates true experimental design from quasi-experimental
277 design is:
A. the use of a control group.
B. random assignment of subjects to experimental groups.
C. multiple observations.
D. all of the above.

D 47. The one-shot design may have severe validity problems because:
275 A. random assignment is made.
B. the group is not representative.
C. the experimental variable cannot be manipulated.
D. it provides little control over extraneous variables.

Fill in the Blank

1. If a research study is attempting to measure whether package size affects the amount of
256 product used, the independent variable is ____.

package size
Chapter 12 Experimental Research 124

2. Field experiments are conducted in ______ settings.

270
natural

3. A one-group pretest-posttest design would be diagrammed as: ___ .

276
O1 X O2

4. In an experiment, the group of subjects exposed to an experimental treatment is called

261 the ___ _____.

experimental group

5. In an experiment, the group of subjects not exposed to an experimental treatment is

261 called the ___ _____.

control group

6. The is a subject or entity whose responses to experimental treatments are

262 observed and measured.

test unit

7. A procedure for assigning subjects to groups by ensuring each respondent group is

264 equated on the basis of pertinent characteristics is called .

matching

8. An unintended effect on the results of an experiment, caused by the subjects knowing

266 that they are participants, is called the _____ _____ .

Hawthorne effect

9. The quality of being able to generalize beyond the data of an experiment to other
273 subjects or groups in the population under study is known as ___ ______ .

external validity

10. The testing effect in internal validity is also called the _____ effect.
272
pretesting
Chapter 12 Experimental Research 125

11. A(n) refers to sample attrition that occurs when some subjects withdraw from
273 an experiment before it is completed, affecting its validity.

mortality effect

12. The term refers to the question of whether the experimental treatment was the
271 sole cause in observed changes in the dependent variable.

internal validity

13. Experimental error occurs when variables are allowed to have an influence on the
265 dependent variables.

extraneous

14. An error that occurs in the same experimental conditions every time the basic
264 experiment is repeated is called ___ ______ .

constant error

15. The term refers to experimental design procedures that unintentionally hint to
265 subjects about the experimenter's hypotheses.

demand characteristics

16. During a day-long experiment, subjects may grow hungry, tired, or bored. If this
272 occurs, there is a possible problem with effects due to .

maturation

17. The process of providing subjects with all pertinent facts about the nature and purpose
268 of the experiment after its completion is called .

debriefing

18. A(n) experiment is an experiment conducted in a natural setting, often for a long
270 period of time.

field

19. Diagram a before-after with control group experiment: .

277
Chapter 12 Experimental Research 126

Experimental group: R O1 X O2
Control group R O3 O4

20. A(n) experiment is an experiment conducted in an artificial setting that gives the
270 researcher almost complete control over the research setting.

laboratory

21. In a(n) design, neither the subjects nor the experimenter knows which are the
267 experimental and which are the control conditions.

double-blind

22. The purpose of experimental research is to allow the researcher to control the research
257 situation so that relationships among variables may be evaluated.

causal

23. A(n) refers to specific events in the external environment occurring between the
272 first and second measurements that are beyond the control of the experimenter and that
affect the validity of the experiment.

history effect

24. Diagram an after-only design: .

275
X O1

25. Many business experiments may be conducted in a short period of time (a month or
280 two). However, business experiments that investigate long-term structural change may
require a(n) design.

time series

26. When experiments are conducted over long periods of time, they are most vulnerable
280 to changes.

historical

27. The experiment diagramed as:

280
O1 O2 O3 X O4 O5 O6
Chapter 12 Experimental Research 127

is called a(n) design.

time series

28. In a randomized block design, the researcher attempts to the effects of a single
282 variable by blocking out its effects.

isolate

29. The _____ design is a balanced, two-way classification scheme that attempts to
286 control or block out the effect of two or more extraneous factors by restricting
randomization with respect to the row and column effects. (Answer could be several
words).

Latin square

30. The number of dollars spent on training may be any number of different values. This is
260 an example of an independent variable that is a(n) variable.

continuous

31. A procedure in which the assignment of subjects and treatments to groups is based on
267 chance is called _____ _____ .

random assignment

32. Exposure of the same subjects to all experimental treatments, in order to eliminate
264 problems due to subject differences, is called ____ _______ .

repeated measures

33. An experimental design that investigates the interaction of two or more independent
283 variables is known as a _____ design.

factorial

34. The machine that is utilized to allow researchers to experiment with the visual impact
270 of advertising, packaging, and so on, by controlling the amount of time a visual image
is exposed to a subject is called a(n) .

tachistoscope
Chapter 12 Experimental Research 128