Deck 3: Research Techniques: Experiments

Question 1

Failure of an independent variable to produce changes in a dependent variable is called a

A) statistical error.
B) failure to replicate.
C) null result.
D) confound.

Answer 1

null result.

Question 2

One reason for preferring randomization to matching for establishing group equivalence is that

A) randomization involves more confounding.
B) we do not know all the relevant variables to match.
C) counterbalancing does not require randomization.
D) randomization guarantees group equivalence.

Answer 2

we do not know all the relevant variables to match.

Question 3

General practice effects in within-subjects experiments include

A) boredom.
B) fatigue.
C) learning.
D) all of these.

Answer 3

all of these.

Question 4

A control group or a control condition is included in an experiment in order to

A) evaluate experimenter effects and demand effects.
B) provide a baseline against which the variable of interest can be compared.
C) prevent ceiling or floor effects.
D) increase the generalizability of the results.

Question 5

Random selection refers to

A) the unreliability of two different experiments yielding different results.
B) the matching of participants on several different factors.
C) the process of categorizing continuous variables as dichotomous variables.
D) the technique used to ensure that each participant has an equal chance of being selected for an experiment.

Question 6

In its most basic form, the __________ the levels of interest of the independent variable.

A) experimental group does not receive
B) control group does not receive
C) both the experimental and the control group do not receive
D) both the experimental and the control group receive

Question 7

The influence of an independent variable may not be accurately reflected in a dependent variable if distorted by

A) counterbalancing.
B) randomization.
C) floor and ceiling effects.
D) chair and sofa effects.

Question 8

An interaction occurs when

A) the effect of one independent is different at the different levels of another independent variable.
B) the effect of one dependent variable is different at the different levels of the confounding variable.
C) chance results occur.
D) the effect of the dependent variable is not significant.

Question 9

Which of these is NOT a possible explanation for obtaining null results in a valid experiment?

A) the levels of the independent variable are too similar to each other.
B) the experimenter used a valid and reliable measure of the dependent variable.
C) the manipulation of the independent variable was unsuccessful..
D) the dependent variable is subject to a floor effect.

Question 10

A dependent variable

A) is manipulated by the experimenter.
B) leads to changes in the independent variable.
C) is a hypothetical construct that cannot be measured.
D) is measured by the experimenter.

Question 11

In an experiment testing the effects of alcohol on performance, Group A receives 3 nonalcoholic beers, Group B receives 6 nonalcoholic beers, Group C receives 3 alcoholic beers, and Group D receives 6 alcoholic beers. The independent variable(s) in this experiment is/are

A) amount of alcohol.
B) number of drinks.
C) amount of alcohol and number of drinks.
D) level of performance after drinking.

Question 12

A major disadvantage of between-subjects designs is that

A) the effect of one treatment may alter the effectiveness of later treatments.
B) participant differences may obscure treatment effects.
C) one must employ fewer independent variables.
D) one can use only a single independent variable.

Question 13

In an experiment the __________ variable is a manipulation of the environment controlled by the experimenter.

A) control
B) experimental
C) dependent
D) independent

Question 14

Which of the following is NOT a technique for creating equivalent comparisons between groups and/or conditions in experiments?

A) obtaining a baseline.
B) matching.
C) randomization.
D) using multiple dependent variables.

Question 15

In the preceding example, if only Group A and Group D were included as conditions

A) the effects of both alcohol and number of drinks could be assessed.
B) the effect of alcohol only could be assessed.
C) the effect of number of drinks only could be assessed.
D) alcohol content and number of drinks would be confounded.

Question 16

The Hawthorne effect is an example of

A) experimenter effects.
B) demand characteristics.
C) experimenter bias.
D) ceiling effects.

Question 17

Regression to the mean implies that if extreme scores are obtained for some individuals and then the observations are repeated, the second scores obtained for these same individuals will be

A) closer to the mean of the first scores of the entire group.
B) more extreme than the first scores of the entire group.
C) exactly the same as before.
D) unreliable.

Question 18

So-called critical experiments

A) test two theories that make different predictions.
B) test a theory that makes two different predictions.
C) involve an inductive approach.
D) involve a number of independent and dependent variables.

Question 19

A mixed design is one in which

A) there is one independent and one dependent variable.
B) at least one independent variable is tested within-subjects, and at least one other independent variable is tested between-subjects.
C) at least one independent variable is manipulated, and at least one other independent variable is controlled.
D) each participant receives all levels of each independent variable.

Question 20

In an experiment, the variable that is expected to change as a result of some manipulation is termed the

A) dependent variable.
B) independent variable.
C) control variable.
D) experimental variable.

Question 21

It is likely that you will find an interaction when

A) the dependent variable is affected by an independent variable.
B) the effect of one independent variable is affected by a second independent variable.
C) the effect of an independent variable is affected by the dependent variable.
D) there is a restriction of range problem.

Question 22

In an experiment with two independent variables, the occurrence of an interaction means that

A) the experiment was poorly designed.
B) there was a ceiling effect.
C) the effect of one independent variable was different at different levels of the other independent variable.
D) the effect of one independent variable was the same at all levels of the other independent variable.

Question 23

In order to obtain equivalent groups in between-subjects designs, you can use each participant as his or her own control.

Question 24

A control variable is a potential dependent variable that is held constant during the experiment.

Question 25

A researcher wants to know whether a new drug reduces anxiety. Forty participants who suffer from anxiety disorder are assigned randomly to one of two groups. One group gets the new drug and the other group gets a placebo. Neither group knows which they receive. After 30 days on the drug, their anxiety levels are measured. In this experiment, is the independent variable and is the dependent variable.

A) anxiety, the new drug
B) the new drug, the placebo
C) anxiety, the placebo
D) the new drug, anxiety

Question 26

Counterbalancing reduces the problem of order in a within-subjects design.

Question 27

One technique in psychological experiments is to keep the experimenter from knowing the research hypothesis or the experimental condition of the subjects currently being tested. This helps to prevent the problem of

A) ceiling effects.
B) carryover effects.
C) placebo effects.
D) experimenter effects.

Question 28

The practice of repeating an experiment that has already been done to determine whether the same effect is observed is called

A) regression.
B) replication.
C) randomization.
D) counterbalancing.

Question 29

Which of the following uses a quasi-experimental design?

A) A researcher compares the effects of caffeine on sleep by assigning subjects randomly into one of 3 treatment groups (5, 10 or 20 mg. doses of caffeine) and recording EEGs of each subject
Throughout the night.
B) To see whether the effectiveness of a new method of reading instruction is different for bilingual children vs. children who speak only English, 40 Bilingual and 40 English-Speaking children are
Taught using the bilingual method for one semester and reading scores for the 2 groups are
Compared.
C) A clinical psychologist tests the relative effectiveness of two forms of therapy for treatment of panic attacks by recruiting 20 people diagnosed with panic attacks and treating half with one form
Of therapy and half with the other form of therapy for the same amount of time.
D) In a study of picture memory, 40 subjects are shown 1000 photographs under standard viewing conditions and recognition of the pictures is tested 1 week, 2 weeks, 4 weeks, and 8 weeks later.

Question 30

Automating an experiment can

A) greatly reduce the possibility of experimenter effects.
B) be a source of an experimenter effect.
C) eliminate the need for a control group.
D) prevent the occurrence of a floor effect.

Question 31

An interaction occurs when the effect of one independent variable is not the same at each level of another independent variable.

Question 32

Ceiling and floor effects are caused by careless counterbalancing.

Question 33

One criterion for a good dependent variable is stability in repeated measures.

Question 34

Matching is a technique used to equate experimental and control groups on certain variables.

Question 35

The effect of boredom or fatigue with a task are termed general practice effects.

Question 36

What is a subject variable?

A) A characteristic or feature of a research participant used to assign that participant to a condition of an experiment (e.g., the subject's age).
B) The individual score of a subject on the dependent variable measure.
C) A characteristic or feature of a subject which is expected to change as the result of an experimental manipulation.
D) The degree to which a subject responds to an experimental manipulation.

Question 37

A control variable is

A) under the control of the subject.
B) a potential independent variable that is held constant in an experiment.
C) a potential independent variable that is manipulated by the experimenter.
D) expected to change as the result of an experimental manipulation.

Question 38

When an experiment produces null results

A) it is possible that the experimenter did not use a valid manipulation of the independent variable.
B) it shows that the independent variable manipulation produced a change in the dependent variable.
C) it shows that the dependent variable had no effect on the independent variable.
D) it is possible that the effect of the independent variable occurred by chance.

Question 39

Randomization is an old and rarely used technique for creating equivalent groups in an experiment.

Question 40

A weak manipulation of the independent variable may result in null results.

Question 41

Changes in the independent variable are presumably caused by changed in the dependent variable.

Question 42

Describe three techniques used to ensure equivalent groups in an experiment. Outline one advantage and one disadvantage of each technique.

Question 43

Describe the rationale for counterbalancing in an experiment.

Question 44

In a between-subjects design, each participant receives one level of the independent variable.

Question 45

In a quasi-experiment, participants can be randomly assigned to all experimental conditions.

Question 46

An experiment needs a dependent variable.

Question 47

A dependent variable is manipulated by the experimenter.

Question 48

Discuss the advantages and disadvantages of within-subjects and between-subjects designs? Describe a situation where one would prefer the use of one design over the other.

Question 49

Counterbalancing is a common way to control for general practice effects.

Question 50

Age, sex, and race are NOT examples of variables that would be used in a quasi-experiment.

Question 51

What is counterbalancing? Provide two examples.

Question 52

One way to protect against experimenter effects is to conduct a double-blind experiment.

Question 53

The following result is an example of an interaction: With normally active children, the stimulating effect of amphetamines increases as the dosage increases, but with hyperactive children, the greater the dosage of amphetamines, the calmer the children.

Question 54

An experimenter can influence the outcome of an experiment by providing subtle cues to the participant.

Question 55

Demand characteristics are those factors that are necessary to conduct an experiment.

Question 56

Matching cannot guard against confounding variables.

Question 57

When data are presented in figures, the dependent variable is represented on the ordinate, and the independent variable is presented on the abscissa.

Question 58

If an experimenter suspects that the effects of one treatment may linger on to alter a later treatment, then that experimenter should use a within-subjects design.

Question 59

What are null results? What are three possible explanations for null results?

Question 60

In order to have an experiment, it is necessary to have at least two levels of the dependent variable.

Question 61

Imagine that you are testing the relative effectiveness of a new reading program for English-speaking vs. bilingual children using a quasi-experimental design. What are some of the steps you would take to obtain equivalent groups for your experiment?

Question 62

Describe the advantages of experimentation over the other research techniques. What are the potential problems with experimentation? How are these problems avoided?

Question 63

Describe an experiment in which a practice effect might be observed.