Deck 5: Survey Research

Question 1

Figure:
A researcher was interested in developing a reliable and valid measure of friendliness.That is,she wanted to distinguish between friendly and unfriendly people using her questionnaire.To determine whether she developed a good measure (called "Friend"),she administered her "Friend" questionnaire to a group of 300 people twice (Friend-Time1 and Friend-Time 2),separated by six months.At the second testing session she also administered a similar questionnaire called "Nice," which previous research has shown to be a valid measure of people's ability to get along with people and to make friends.She also administered a questionnaire designed to measure art ability,which she predicted to be unrelated to friendliness.She observed the following correlations among the measures:
What correlation coefficient value indicates the "Friend" measure is reliable?

Answer 1

The test-retest reliability is .85.

Question 2

A professor wants to determine if students' final exam scores in a course can be predicted based on their scores on the first test.The quantitative index the professor is likely using in this situation is the

A)prediction product.
B)contingency score.
C)correlation coefficient.
D)covariation cross product.

Answer 2

correlation coefficient.

Question 3

Surveys play an important role in a type of research that is intended to assess the covariation of naturally occurring variables with the goal of identifying predictive relationships.This general type of research is called

A)experimental research.
B)analytical research.
C)qualitative research.
D)correlational research.

Answer 3

correlational research.

Question 4

The quantitative index of the direction and magnitude of a predictive relationship is called a

A)convergent validity.
B)correlation coefficient.
C)contingency variable.
D)discriminant statistic.

Question 5

Surveys play an important role in the type of research that is intended to assess the covariation between naturally occurring variables.This type of research is called

A)correlational research.
B)qualitative research.
C)causal research.
D)experimental research.

Question 6

Self-report measures of people's attitudes and feelings are reliable if they consistently yield similar results.How is reliability typically measured?

Question 7

Why is a longitudinal design preferred over a study using successive independent samples to assess the direction and extent of attitude change for individual respondents?

Question 8

Which of the following pieces of information is the least useful when evaluating whether the results of a survey are biased?

A)whether an organization sponsored the survey
B)whether the survey involves a biased sample
C)whether the wording of the questions on the questionnaire were slanted
D)whether the survey data have been selectively analyzed or reported

Question 9

Figure:
Suppose a researcher wishes to use stratified random sampling to select a representative sample of 100 people from a population that has the following characteristics:
Describe the sample in terms of how many single and married women and men the researcher would need to represent the population.

Question 10

In sampling,the specific list of the elements of the population is called the

A)sample.
B)sampling frame.
C)survey set.
D)representative sample.

Question 11

The goal of sampling is to identify a _______ that will be representative of the ________.

A)population,sample
B)sample,sampling frame
C)sampling frame,sample
D)sample,population

Question 12

Figure:
A researcher was interested in developing a reliable and valid measure of friendliness.That is,she wanted to distinguish between friendly and unfriendly people using her questionnaire.To determine whether she developed a good measure (called "Friend"),she administered her "Friend" questionnaire to a group of 300 people twice (Friend-Time1 and Friend-Time 2),separated by six months.At the second testing session she also administered a similar questionnaire called "Nice," which previous research has shown to be a valid measure of people's ability to get along with people and to make friends.She also administered a questionnaire designed to measure art ability,which she predicted to be unrelated to friendliness.She observed the following correlations among the measures:
What correlation coefficient value indicates the "Friend" measure has good convergent validity?

Question 13

A sampling frame in survey research could be considered a(n)__________ of the population.

A)representative sample
B)biased sample
C)operational definition
D)probability sample

Question 14

Identify the key features of a good questionnaire item.

Question 15

Figure:
A researcher was interested in developing a reliable and valid measure of friendliness.That is,she wanted to distinguish between friendly and unfriendly people using her questionnaire.To determine whether she developed a good measure (called "Friend"),she administered her "Friend" questionnaire to a group of 300 people twice (Friend-Time1 and Friend-Time 2),separated by six months.At the second testing session she also administered a similar questionnaire called "Nice," which previous research has shown to be a valid measure of people's ability to get along with people and to make friends.She also administered a questionnaire designed to measure art ability,which she predicted to be unrelated to friendliness.She observed the following correlations among the measures:
Is the researcher able to conclude that she has developed a reliable and valid measure of friendliness?

Question 16

Which of the following characteristics do most surveys have in common?

A)direct observation and inference
B)sampling and a set of predetermined questions
C)limited scope and pragmatic purpose
D)interviewing and the use of focus groups

Question 17

Distinguish between nonprobability sampling and probability sampling.Name a type of nonprobability sample and two types of probability samples.

Question 18

Figure:
Suppose a researcher wishes to use stratified random sampling to select a representative sample of 100 people from a population that has the following characteristics:
Suppose the researcher mails the survey to the 100 people in the representative sample and only 50 of the people return a completed survey.Can you state that this sample of 50 represents the population? Why or why not?

Question 19

Figure:
Suppose a researcher wishes to use stratified random sampling to select a representative sample of 100 people from a population that has the following characteristics:
Dual-income couples now outnumber those with single incomes,and this has led researchers to examine how the demands of maintaining dual careers affect relationships.The researchers found that among dual-career couples there was a positive correlation such that couples who spent greater amounts of time together found their relationship more satisfactory.Presume that this finding is a reliable one.What mediating variables would you propose to explain why there is a positive correlation between the time dual-career couples spend together and their degree of satisfaction with their relationship?

Question 20

Figure:
A researcher was interested in developing a reliable and valid measure of friendliness.That is,she wanted to distinguish between friendly and unfriendly people using her questionnaire.To determine whether she developed a good measure (called "Friend"),she administered her "Friend" questionnaire to a group of 300 people twice (Friend-Time1 and Friend-Time 2),separated by six months.At the second testing session she also administered a similar questionnaire called "Nice," which previous research has shown to be a valid measure of people's ability to get along with people and to make friends.She also administered a questionnaire designed to measure art ability,which she predicted to be unrelated to friendliness.She observed the following correlations among the measures:
What correlation coefficient value indicates the "Friend" measure has good discriminant validity?

Question 21

The decision of how many people needed in a random sample to represent the population should be based on the

A)number of people available and willing to participate in the survey.
B)financial resources available to the researcher for completing the survey.
C)degree of variability in the population.
D)the availability of an adequate sampling frame.

Question 22

Techniques to increase the return rate for a survey include

A)preserving a sense of confidentiality by addressing the survey to no one in particular ("Dear occupant").
B)making sure the survey requires minimal effort from the respondent.
C)minimizing the respondents' likely interest in the survey or sponsoring organization.
D)all of these

Question 23

A TV reporter interviewed several dozen people at the scene of a political demonstration about their views of the people doing the demonstrating.In the terminology of survey research,the reporter has gathered a(n)

A)purposive sample.
B)convenience sample.
C)random sample.
D)stratified sample.

Question 24

When the distribution of characteristics in a sample is systematically different from that of the target population,the sample is called a

A)distorted sample.
B)selected sample.
C)skewed sample.
D)biased sample.

Question 25

Stratified random sampling is especially useful when the researcher is interested in

A)making specific statements about individual respondents.
B)making general statements about the population as a whole.
C)making comparisons with previous studies that have used simple random sampling.
D)making general statements about specific portions of the population that has been sampleD.

Question 26

A researcher adjusted the wording of a question to fit the respondent or records only selected portions of the respondent's answers.What survey problem does this represent?

A)respondent bias
B)response bias
C)selection bias
D)interviewer bias

Question 27

Which of the following samples would be representative of a population that is 75% women and 25% men?

A)60 women,20 men
B)75 women,25 men
C)225 women,75 men
D)all of these

Question 28

Which of the following survey methods is best for dealing with highly personal or embarrassing topics,especially when the anonymity of respondents is preserved?

A)mail survey
B)personal interview
C)telephone interview
D)cluster survey

Question 29

The best approach for obtaining a representative sample when using stratified random sampling is to

A)sample so that there is an equal number of respondents in the different stratA.
B)sample so that the proportion of respondents in the different strata represents the proportions in the population.
C)use random-digit dialing.
D)eliminate people from the sampling frame who are unlikely to return a completed survey.

Question 30

A researcher decides to conduct a survey using telephone interviews and has access to random-digit dialing.Because some people have multiple phone numbers and therefore,a greater likelihood of being surveyed,the researcher should be aware of the potential problem of

A)response rate bias
B)interviewer bias
C)selection bias
D)telephone bias

Question 31

Which of the following samples would represent a biased sample if a population comprised 40% freshmen,30% sophomores,15% juniors,and 15% seniors?

A)200 of each class year (a total sample of 800 students)
B)80 freshmen,60 sophomores,30 juniors,30 seniors
C)40 freshmen,30 sophomores,15 juniors,15 seniors
D)8 freshmen,6 sophomores,3 juniors,3 seniors

Question 32

Survey researchers are most interested in

A)the responses of individuals in their sample.
B)obtaining a very large sample size.
C)predicting which individuals in the sample will return a completed survey.
D)generalizing responses from a sample to the population.

Question 33

A survey researcher chose to administer a survey using the Internet.This sample likely overrepresents people who own a computer and underrepresents people who don't own a computer.This problem in the researcher's survey is best described as

A)low response rate.
B)selection bias.
C)social desirability.
D)a spurious relationship.

Question 34

The problem of selection bias in survey research is similar to the problem of ________ in research using archival data.

A)selection threat to internal validity
B)spurious relationships
C)selective survival
D)selective deposit

Question 35

The ability to generalize the results from a survey sample to the population of interest depends most critically on the sample's

A)representativeness.
B)stratification.
C)homogeneity vs.heterogeneity.
D)correlations.

Question 36

Suppose a recent issue of the campus newspaper reported a survey about students' attitudes toward proposed changes in the required curriculum.Which of the following would you need to focus on most to decide whether the survey results reported in the paper can be generalized to the entire population of students?

A)number of students in the sample
B)distinctiveness of the sample
C)representativeness of the sample
D)variability among students in the sample

Question 37

In a simple random sample,every individual in the population

A)completes the survey.
B)decides to participate in the survey if it is convenient for them.
C)has an equal chance of being selected to be in the sample.
D)none of these

Question 38

Mail surveys are particularly vulnerable to a problem that arises when not all respondents complete and return the survey.This problem is called

A)selection bias.
B)reactivity.
C)complete bias.
D)low response rate.

Question 39

A survey using the Internet to ask people about attitudes toward mental health services would most likely be criticized for

A)selection bias.
B)population bias.
C)interviewer bias.
D)response rate bias.

Question 40

Low response rates that result when respondents fail to return a completed mail survey are a major factor leading to response rate bias.These low response rates pose a potential problem because

A)segments of the population may be overrepresented or underrepresented in the final sample.
B)a carefully selected convenience sample becomes a probability sample.
C)only surveys with a 100% response rate can be interpreted.
D)the sample size becomes smaller than the researcher intendeD.

Question 41

Attrition in survey research corresponds most closely to which of the following problems in the use of unobtrusive measures (physical traces,archival records)?

A)natural treatments
B)selective survival
C)selective deposit
D)spurious relationships

Question 42

A major threat to the interpretation of results from a successive independent samples design is

A)noncomparable successive samples.
B)attrition from the study.
C)stratified random sampling from the population.
D)interviewer bias.

Question 43

Construct validity can be assessed by determining the degree to which similar measures

A)are not correlated.
B)converge on a construct.
C)show consistency over time.
D)all of these

Question 44

studied changes in the values and goals of college freshman from 1966 to the present.They drew random samples from each year's incoming class.Which survey research design did these researchers use?

A)convenience design
B)cross-sectional design
C)successive independent samples design
D)longitudinal design

Question 45

Scales that measure preferences and attitudes on a dimension such as life satisfaction are intended to identify

A)respondents showing unique scores on the dimension.
B)respondents who consistently choose the middle response on a scale.
C)differences among respondents on the dimension.
D)respondents showing extreme scores on the dimension.

Question 46

A university alumni office conducts a survey of their alumni to assess changes in attitudes toward the university over time.In the first year of the survey,the alumni office surveys graduates from the year 2007.In the second year of the survey,graduates from the year 2009 are surveyed.The results of the survey indicated very different attitudes of the alumni.The alumni office concluded that attitudes toward the university changed over time.This conclusion

A)fairly describes the differences between 2007 and 2009 graduates.
B)fairly describes changes in attitudes over time among the alumni.
C)may be inaccurate because of the problem of attrition.
D)may be inaccurate because of the problem of noncomparable samples.

Question 47

The distinguishing characteristic of the longitudinal design is that

A)several samples from the same population are tested once.
B)samples from several populations are compared.
C)a sample of individuals is asked to complete an especially long survey.
D)the same sample of respondents is surveyed more than once.

Question 48

Convergent validity in survey research,used to establish the construct validity of a measure,is similar to the use of _________ to establish the validity of findings from research employing unobtrusive measures.

A)converging evidence
B)spurious relationships
C)natural treatments
D)archival data

Question 49

The reliability of a survey or test can be increased by

A)identifying a sample of individuals who are very similar to each other in terms of the construct of interest.
B)asking people to respond to a few brief questions about the concept of interest.
C)asking people to answer many questions about the concept of interest.
D)none of these

Question 50

The potential problem of noncomparable successive samples can arise in the successive independent samples design.Samples are considered noncomparable when

A)they differ in sample size by more than 50 respondents.
B)they are not representative of the same population.
C)there is a large difference among the samples on the average survey response.
D)there is no standard value in the population on which the different samples can be compareD.

Question 51

A survey researcher randomly selected individuals in Canada,Mexico,and the United States and mailed them a survey about North American trade policies.The researcher compared individuals' survey responses in these countries.This researcher used a

A)cross-sectional survey design.
B)successive independent samples design.
C)longitudinal survey design.
D)observational design.

Question 52

A nationwide study was done in which samples of students were asked their opinions about the quality of their college education.The results of the survey were analyzed for differences among students at the various colleges and universities.What survey design was used in this study?

A)correlated samples design
B)cross-sectional design
C)successive independent samples design
D)longitudinal design

Question 53

Unless all the respondents in the original sample complete all phases of a longitudinal design,there is a possible problem due to

A)selection bias.
B)projected data.
C)attrition.
D)data erosion.

Question 54

Which of the following statements regarding Internet surveys is false?

A)Selection bias is less problematic for Internet surveys because of the large samples that are typically obtained.
B)Researchers have little control over the research environment when conducting Internet surveys.
C)Response rate bias due to nonresponding leads to relatively low response rates for Internet surveys.
D)Internet surveys represent a low-cost,efficient method for obtaining a large,potentially diverse sample of respondents.

Question 55

Which of the following is not a potential limitation of the longitudinal design?

A)It takes a massive effort to complete a longitudinal design successfully.
B)It can be difficult to obtain a sample of respondents willing to participate in a longitudinal design.
C)It is not possible to determine whether the final sample is comparable to the original sample in a longitudinal design.
D)It is difficult to pinpoint the exact causes for changes in individuals' attitudes that are identified using a longitudinal design.

Question 56

A researcher is interested in studying the developmental changes in social and political attitudes that occur during the time a person is in college.The researcher is interested in both overall trends and in the changes that individual students undergo.Which survey research design will this researcher need to use?

A)observational design
B)cross-sectional design
C)successive independent samples design
D)longitudinal design

Question 57

Which of the following factors is likely to increase the reliability of a test or of a questionnaire?

A)distractions in the testing situation
B)variability on the factor being tested among those individuals being tested
C)ambiguous instructions for completing the test
D)fewer items on a test

Question 58

Which of the following must be true for a test or questionnaire to be considered reliable as measured by the computation of test-retest reliability?

A)People would have to have identical scores on the two administrations of the test.
B)People would have to hold very different relative positions in the distributions for the two test times.
C)People would have to hold similar relative positions in the distributions for the two test times.
D)People would have to have scores on the two administrations of the test that were uncorrelateD.

Question 59

In order to assess the test-retest reliability of a measure the researcher should

A)administer the measure at two different times to the same people and compute the correlation between the two sets of scores.
B)administer the measure at two different times to different people and compute the correlation between the two sets of scores.
C)correlated individuals' scores on the measure with a different measure designed to measure the same construct.
D)ask two people to complete the measure several times and compute the correlations across the many time periods.

Question 60

Reliability refers to the ______ of a measure,and validity refers to the ______ of a measure.

A)consistency,design
B)truthfulness,design
C)truthfulness,consistency
D)consistency,truthfulness

Question 61

Survey questions that are ordered from the most general to the most specific are

A)filter questions.
B)leading questions.
C)double-barreled questions.
D)funnel questions.

Question 62

When a questionnaire item implies a preferred response within the wording the question,it is called a

A)loaded question.
B)double-barreled question.
C)leading question.
D)filter question.

Question 63

A researcher develops a measure of the "self-esteem" construct.One item on the questionnaire states,"I generally feel good about myself." Another item states,"I think I'm not a very good person." The researcher likely uses these two items to protect against possible problems associated with

A)leading questions.
B)response bias.
C)selection bias.
D)the order of questions.

Question 64

One way to minimize the likelihood that respondents might view questions as "loaded" for a survey on controversial topics is to

A)use funnel questions to direct people to survey questions that are not controversial.
B)ask people with a range of perspectives on the topic to review and edit the questions for potentially offensive wording.
C)use leading questions first to make sure the respondents agree with the survey topiC.
D)use emotional words that represent both sides of the controversial topic so that the survey is not biaseD.

Question 65

An important initial step when developing a questionnaire is to

A)ask a wide variety of people to submit possible questions.
B)pretest the questionnaire using a sample similar to the planned survey.
C)write the questions as quickly as possible because a person's first ideas about a topic are usually the best.
D)test only one or two items and then add more for the planned survey.

Question 66

The principle advantage of free-response or open-ended questions is that they

A)are easier to score and code.
B)provide greater comparability of answers across respondents.
C)minimize the problem of expressiveness and spontaneity.
D)offer the respondent greater flexibility.

Question 67

A student is asked in a survey question about whether he would help another person who is in trouble.The student thinks that he probably would not help because he would be hesitant to get involved in a potential dangerous situation.The student decides,however,to respond that he would help.The student is likely responding to the pressure of

A)sensitivity to conformity.
B)social desirability.
C)response bias.
D)consensus responding.

Question 68

A respondent completes items on a closed-response survey by agreeing with each statement on the survey.This might represent a problem of

A)response bias.
B)selection bias.
C)filter questions.
D)loaded questions.

Question 69

A researcher wants to assess the construct validity of a measure of life satisfaction.Which of the following dimensions would likely be most useful in determining the discriminant validity of the measure?

A)happiness
B)well-being
C)self-esteem
D)artistic ability

Question 70

Which of the following standards is typically applied when judging the truthfulness of the responses people make to survey questions?

A)generally accept people's responses as truthful unless there is reason to do otherwise
B)generally be skeptical of people's responses as probably not truthful and statistically adjust the responses
C)generally be skeptical of people's responses as probably not truthful unless there is strong evidence that they are truthful
D)generally accept people's responses as truthful once the survey responses have been collected

Question 71

The primary disadvantage of closed-response questions is that

A)they allow respondents too much flexibility in their responses.
B)responses are very difficult to code and to summarize.
C)respondents may be forced to choose among response alternatives that don't describe their true opinion.
D)all of these

Question 72

Correlational research is potentially limited when it comes to

A)applying the results of correlational studies to predict such issues as physical and psychological health-related problems.
B)interpreting causal relationships based on correlational studies alone.
C)establishing quantitative relationships between two sets of responses obtained in a survey.
D)using sophisticated statistical procedures in analyzing correlational studies.

Question 73

A survey on consumer products has questions regarding automobile repair,home electronics,and travel destinations.The method for allowing respondents to skip the section on automobile repair if they don't own a car is to use

A)leading questions.
B)funnel questions.
C)filter questions.
D)loaded questions.

Question 74

A survey question that asks,"Are you worried about global climate change and nuclear weapons?" is an example of a

A)double-barreled question.
B)loaded question.
C)leading question.
D)funnel question.

Question 75

There is a reliable correlation between being outgoing and being satisfied with one's life.One possible way to account for this relationship is that a third variable (number of friends)leads people to be more outgoing and more satisfied with their lives.A correlation that can be accounted for by a third variable such as number of friends is called a

A)chaotic relationship.
B)causal relationship.
C)spurious relationship.
D)co-dependent relationship.

Question 76

If the relationships between measures of poverty,chaos,and psychological distress differed depending on whether families lived in rural or urban areas,this variable of population density (rural vs.urban)would be considered a

A)moderator variable.
B)mediating variable.
C)causal variable.
D)spurious variable.

Question 77

When individuals report on a questionnaire that their behavior is more favorable than it actually is,researchers need to be concerned about the _____ of the questions.

A)morality
B)selection bias
C)demographics
D)social desirability

Question 78

Research findings indicate that people's verbal reports of their behavior and observations of their actual behavior do not always match.These findings provide a good rationale for a(n)__________ approach to psychological research.

A)multimethod
B)observational
C)survey
D)correlational

Question 79

Using path analysis,the correlation between poverty and psychological distress can be partially explained by the level of chaos in the home.The chaos variable in this example represents a(n)

A)moderator variable.
B)mediating variable.
C)proxy variable.
D)spurious variable.