Deck 13: Multiple Methods

Question 1

Which of the following is not an example of triangulation?

A) using three different questions to measure the same concept
B) replicating an experiment with a different manipulation of the independent variable
C) asking the same question to a large sample of respondents
D) using both experimentation and the analysis of available data to test the same hypothesis
E) using a team of investigators in field research

Answer 1

C

Question 2

Which of the following is not an advantage of composite measures over single-item measures?

A) increased reliability
B) lower measurement error
C) greater variability
D) easier to interpret research results

Answer 2

D

Question 3

Indexes are similar to scales in that both

A) are composite measures of concepts.
B) reflect an arbitrary combination of items.
C) are designed and tested for unidimensional measurement.
D) rely exclusively on survey data.

Answer 3

A

Question 4

Scales differ from indexes in terms of

A) what they are used to measure.
B) how items are selected and combined to create the measure.
C) the type of indicators or survey items included in the measure.
D) how their relationships with other variables can be analyzed.

Question 5

If a Guttman scale contains six items, we can conclude that

A) there are seven perfect response patterns possible.
B) the percentage of nonscale patterns will be lower than in a scale containing fewer items.
C) the percentage of nonscale patterns will be lower than in a scale containing more items.
D) the researcher probably discarded one or two items in the process of selecting scalable items.

Question 6

A researcher is able to figure out a person's responses to each of four items measuring "attitude toward divorce" by knowing his or her total score. This indicates that the researcher has developed

A) an index.
B) a Guttman scale.
C) a Likert scale.
D) a structural equation model.

Question 7

In contrast to a Guttman scale, in a summated ratings or Likert scale (Box 13.1),

A) scale items are added together to arrive at a total score.
B) total score on the scale indicates the values or responses for every scale item.
C) each scale item reflects the underlying continuum of the concept to the same degree.
D) the set of items comprising the scale is ordered and cumulative.

Question 8

A researcher creates a composite measure of attitudes toward research methods consisting of twenty statements with which respondents indicate their level of agreement or disagreement ("strongly agree" to "strongly disagree"). The researcher's measure is an example of

A) an index.
B) a Guttman scale.
C) a Likert scale.
D) a structural equation model.

Question 9

Structural equation modeling differs from composite measurement in that it

A) relies on a single-item measure of the dependent variable.
B) simplifies the measurement process.
C) assesses reliability.
D) maintains the separate identity of each indicator throughout the analysis.

Question 10

Which of the following is true about replications in social research?

A) They are frequently reported in the social science literature.
B) They are most commonly found in experimental research.
C) They duplicate the exact conditions of the original study.
D) They are regarded as cost-effective investments of time and money.

Question 11

Jean Twenge and associates' studies of social exclusion and helping showed that

A) people who feel excluded are more likely to help others.
B) people who feel excluded are less likely to help others.
C) the effect of social exclusion on helping depends on the type of help that is needed.
D) the effect of social exclusion on helping depends on how people are made to feel excluded.

Question 12

How was social exclusion manipulated in the social exclusion and helping experiment?

A) By telling participants that a personality test showed "they were likely to be alone later in life."
B) By asking participants if they have ever been excluded from the activities of a friendship group.
C) By asking participants to think about a time in their lives when they had felt excluded.
D) By pointing out several personal traits that are likely to lead to social exclusion.

Question 13

Jean Twenge and associates' studies of social exclusion and helping

A) replicated the same basic experiment at different times of the day.
B) operationalized social exclusion and helping in different ways.
C) drew subjects from diverse populations.
D) produced widely varying results.

Question 14

In comparison with the original Minneapolis Spouse Abuse Experiment, Berk and Newton's replication

A) found that mediation was more likely than arrest to deter domestic violence.
B) had greater internal validity.
C) involved a different setting and no manipulation of the independent variable.
D) used probability sampling as well as random assignment.
E) involved different measures of the dependent variable.

Question 15

Andrew Cherlin and associates applied the triangulation concept in their study of the influence of abuse on marriage and cohabitation by

A) integrating field research methods with survey research methods.
B) comparing surveys of respondents in low-income and high-income areas.
C) replicating their survey five times within a period of ten years.
D) incorporating an experimental design within their surveys of three communities.

Question 16

To understand the effect of abuse on marriage and cohabitation, Andrew Cherlin and associates

A) conducted a survey after using field research to identify victims of abuse.
B) did field research on single-parent households that were identified in a survey.
C) used survey data to test hypotheses developed from field research.
D) used field research to make sense of conflicting survey results.

Question 17

Which basic approach to social research produces the strongest inferences about causal relationships?

A) experimentation
B) survey research
C) field research
D) available-data research

Question 18

Which basic approach to social research is best for obtaining precise estimates of population characteristics?

A) experimentation
B) survey research
C) field research
D) available-data research

Question 19

Which basic approach to social research lends itself best to examining situations and experiences from the viewpoint of the research participant or actor?

A) experimentation
B) survey research
C) field research
D) available-data research

Question 20

Which basic approach to social research affords the fewest controls on extraneous variables?

A) experimentation
B) survey research
C) field research
D) available-data research

Question 21

Which basic approach to social research lends itself best to studies of long-term social change?

A) experimentation
B) survey research
C) field research
D) available-data research

Question 22

Which basic approach to social research tends to be guided by very broad theoretical and substantive questions that de-emphasize research design?

A) experimentation
B) survey research
C) field research
D) available-data research

Question 23

Which basic approach to social research is least likely to involve reactive measurement effects?

A) experimentation
B) survey research
C) field research
D) available-data research

Question 24

Which basic approach to social research affords the greatest control over error and bias associated with sampling?

A) experimentation
B) survey research
C) field research
D) available-data research

Question 25

Which basic approach to social research is most likely to use college students as the units of analysis?

A) experimentation
B) survey research
C) field research
D) available-data research

Question 26

Which basic approach to social research generally is most difficult to replicate?

A) experimentation
B) survey research
C) field research
D) available-data research

Question 27

Which basic approach to social research optimizes internal validity?

A) experimentation
B) survey research
C) field research
D) available-data research

Question 28

Probability sampling from a specified target population is common in __________ but rarely found in __________.

A) experiments; field research
B) field research; available data research
C) field research; experiments
D) available data research; surveys
E) surveys; experiments

Question 29

Meta-analysis is a form of which approach to social research?

A) experimentation
B) survey research
C) field research
D) analysis of available data

Question 30

In a meta-analysis, the researcher

A) obtains the raw data from and replicates the data analysis in selected previous studies.
B) uses all four approaches to test the same hypothesis.
C) synthesizes the results of previous studies on a research question.
D) investigates two or more hypotheses regarding the same social phenomenon.

Question 31

The studies included in a meta-analysis

A) usually consist of the complete target population.
B) may be biased because of the inability to identify unpublished research.
C) usually consist of a random sample of the target population.
D) are called "file drawer" studies.

Question 32

Which of the following is not true of meta-analysis?

A) Variables are measured by coding different characteristics of studies.
B) Variation in study quality may explain differences in study results.
C) Researchers use their personal judgment to synthesize the results of previous studies.
D) It tends to share the weaknesses and strengths of the studies included in the analysis.

Question 33

The fixed-effects and random-effects frameworks for meta-analysis differ with respect to

A) assumptions made about the sources of variability between studies.
B) the type of studies-experiments or surveys-that are analyzed.
C) the appropriateness of conducting tests of statistical significance.
D) the presence or absence of random processes such as random sampling and random assignment in the studies that are being analyzed.

Question 34

Triangulation strengthens the conclusion that a finding is not due to measurement error or uncontrolled extraneous variables.

Question 35

The key to triangulation is to use several dissimilar methods or measures to address a research question.

Question 36

Triangulation is implicit in studies based on cross-cultural data archives.

Question 37

Indexes and scales are types of composite measures.

Question 38

Indexes are generally superior to scales.

Question 39

One problem with composite measures is that adding items together can magnify random measurement error.

Question 40

The items included in an index are always assumed to be of equal importance.

Question 41

Guttman scaling is a method of constructing a unidimensional scale.

Question 42

A key to Likert scaling is to select items that differentiate between high and low scorers on the dimension being measured.

Question 43

The items in Likert scales are coded as 0 or 1.

Question 44

Structural equation modeling is a scale construction technique.

Question 45

All replications are dissimilar to some degree.

Question 46

Despite their value, replications are infrequent in social research.

Question 47

Experimental replications usually vary certain features of the original study.

Question 48

The Minneapolis Spouse Abuse Experiment used a quasi-experimental nonequivalent control group design.

Question 49

The study of the effects of abuse on marriage and cohabitation combined survey research with the analysis of available data.

Question 50

The choice of a research strategy should be determined primarily by what the researcher has been trained best to do.

Question 51

Surveys are the strongest approach in terms of internal validity.

Question 52

Laboratory experiments and surveys are more susceptible to reactive measurement effects than are other research approaches.

Question 53

Sampling error and bias are best controlled in surveys and most problematic in experiments and field research.

Question 54

Generalization of study results is most problematic in experiments and field research.

Question 55

Checks on reliability and validity are more readily available in field research than in other research approaches.

Question 56

Field research is generally the easiest type of research to replicate.

Question 57

The units of analysis in meta-analysis are reports of research studies.

Question 58

Meta-analysis may be used to assess the external validity of a finding or to test a new hypothesis.

Question 59

The "file drawer" problem in meta-analysis refers to the overabundance of previous studies on a research question.

Question 60

Because of suspicions about unpublished research, meta-analysis is restricted to studies identified in computer searches of standard literature databases.

Question 61

Meta-analysis uses a statistical approach to draw inferences about a set of independent studies.

Question 62

Explain how the concept of triangulation is applied in measurement with respect to (a) operational definitions, (b) reliability assessment, and (c) validity assessment.

Question 63

Sociologists Richard Schwartz and James Miller proposed that changes in societies' legal systems parallel changes in societal complexity (or, in Durkheim's terms, the societal "division of labor"). Fully developed legal systems are seen to have the following three characteristics:
i) Mediation: regular use of third parties to settle disputes
ii) Police: use of specialized, armed force to enforce laws
iii) Counsel: regular use of specialized, nonkin advocates (e.g., lawyers)
Assuming that these three characteristics, in order, represent a progression of complexity, present a table that shows the response pattern of a perfect Guttman scale. Then give an example of a "nonscale" response pattern.

Question 64

Some research indicates that people tend to associate status and height. As a consequence, we tend to perceive persons with prestigious positions or titles as taller than they actually are. Briefly describe how you could test this hypothesis with three different approaches: an experiment, a survey, and the analysis of existing data. Then, compare the relative advantages and disadvantages of using each approach.

Question 65

Discuss the advantages and disadvantages of meta-analysis with respect to internal and external validity.