Constructing Surveys Flashcards

1
Q

Why is representative sampling important in surveys?

A

In survey research representative sampling is essential to permit generalization from a sample to a population.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

How can you do questionnaire pretesting?

A

Questionnaire pretesting identifies questions that respondents have difficulty understanding or interpret differently than the researcher intended.

Based on an interview (debriefing) with a pilot group.
- Conventional pretesting also detects questions that respondents have difficulty understanding or interpret differently, but less reliably.

Behavior coding: Questions that elicit frequent deviations are presumed to require modification.
- Behavior coding is quite consistent in detecting apparent respondent difficulties and interviewer problems.

Cognitive pretesting, which involves asking respondents to think aloud while answering questions.
- tend to exhibit low reliability across trials and to detect respondent difficulties almost exclusively.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Define Open versus Closed Questions

A

closed-ended questions require people to choose among a set of provided response alternatives.

In open-ended questions, respondents answered in their own words

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

what is social desirability bias?

A

overreporting of admirable attitudes and behaviors and underreporting those that are not socially respected.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

what is meant by Optimizing?

A

A great deal of cognitive work is required to generate an optimal answer to even a single question.

To the extent that a person is encouraged to perform the necessary cognitive tasks in a thorough and unbiased manner, a person may be said to be optimizing.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

what is meant by satisficing?

A

Satisficing is rather than expending the effort to generate optimal answers, respondents may compromise their standards and expend less energy.
o respondents may be less thorough in comprehension, retrieval, judgment, and response selection -> Weak satisficing.
o A more dramatic approach is to skip the retrieval and judgment steps altogether -> strong satisficing.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What CONDITIONS FOSTER SATISFICING?

A

Satisficing is more likely to occur:
o (a) the greater the task difficulty,
o (b) the lower the respondent’s ability,
o (c) the lower the respondent’s motivation to optimize.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is response order effects?

A

response order effects refer to the effect connected with the order in which response alternatives are presented.
- Written format: primacy effects, respondents want to answer as fast as possible
- Interviewer administered format: Cannot predict whether primay or recency effects.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is the perceptual contrast effect?

A

Perceptual contrast effects: it may cause a moderately plausible alternative to seem less plausible if considered after a highly plausible one or more plausible if considered after a highly implausible one.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is Acquiescence?

A

The tendency to endorse any assertion made in a question, regardless of its content, perhaps due to human tendency to appear agreeable or due to human cognitive biases

People who acquiesce are likely to manifest other forms of satisficing.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Explain the discrepancy between ratings and rankings.

A

a number of studies indicate that rankings yield higher-quality data than ratings.

Respondents are more likely to make mistakes when answering rating questions, failing to answer an item more often than when ranking

For rating questions, respondents who are inclined to implement strong satisficing can simply select a reasonable point on the scale and place all the objects at that point.
- called nondifferentiation.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Explain the SELECTION OF NO-OPINION RESPONSE OPTIONS

A

Many more respondents say they have no opinion on an issue when this option is explicitly offered than when they must volunteer it on their own.

By offering a NO option, respondents would be discouraged from offering meaningless opinions.

Although NO responses sometimes occur because people have no information about an object, they occur more often for a variety of other reasons.
o i.e., because people feel ambivalent about the issue or because they do not understand the meaning of a question.
o offering a NO option may discourage respondents from providing thoughtful answers -> result from satisficing.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Problems related to QUESTIONS ABOUT EVENTS AND BEHAVIORS

A

Lack of encoding (simply not being aware)  Is the biggest obstacle

For events that respondents do encode, two major types of errors affect self-reports.
o Omission result when individual events are forgotten because of dating errors, because similar events become conflated in a “generic memory”, or because the wording of a question leads the respondent to search some areas of memory while neglecting others.
o Intrusions result when events are telescoped forward in time or when memories are altered by scripts, schemata or embellishments from retellings over time.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Problems related to QUESTIONS ABOUT SUBJECTIVE PHENOMENA

A

Subjective phenomena can be related to problems such as, lack of motivation to answer carefully, and response sets such as acquiescence.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is meant by unipolar/bipolar rating questions?

A

Rating questions can be structured as either bipolar (ex. Extremely boring to extremely interesting) or unipolar (ex. Not at all interesting to extremely interesting).

The bipolar scale has a midpoint at which there is a transition. This can be perceived as either indifference or ambivalence.

Unipolar items make fewer assumptions, but they risk irritating respondents who see questions that present negative and positive dimensions separately as repetitive or inappropriate.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Why is it problematic to ask one question only for a dependent variable if it is on a Likert scale?

A

When reporting data through Likert-scale it is important that we ask more questions for the dependent variable, because it will most likely not be normally distributed.
o At least not with a small sample size

Instead, ask more questions and average them into a sum scale to gain a continuous variable with a better distribution.

17
Q

Mention some probability sampling methods

A

o Simple random sampling
o Systematic
o Stratified
o Multi-stage cluster

18
Q

What can be done to correct bias arising from nonresponse?

A

If nonresponse creates a bias, we could try to correct with weights (if the sample was probabilistically selected)

Figuring out discrepancies between population and sample can produce correcting weight
=>apply the weight in the analysis, which basically means that some individuals will “count for more and some for less” in calculation of sample characteristics.

19
Q

What are some pro/cons of interview surveys?

A

PROS:
o Can clarify questions (rigid/conversational interviewing)
o Encourages participation/completion

CONS
o Bias- interviewer effects, higher social desirability
o May not be seen as anonymous

20
Q

What are some pro/cons of self-administered surveys?

A

PROS:
o Easier access
o Efficient, low cost, fast, anonymous
o Allows for branching and piping, feedback, error, and consistency checking, and random order of questions

CONS
o Low response rates
o Low involvement from respondents
o Ambiguities not detected- false responses

21
Q

What is common method variance bias?

A

The covariance between two constructs/measures might be caused by utilizing the same method to collect data on both constructs rather than true relationships that we want to measure

When we are creating measures that are supposed to capture systematic variance between items, any systematic variance will be clumped together with the measure, that means also systematic variance caused by common method.

This systematic error in one or more variables will pollute results in a regression- pollutes estimated of relationship between two constructs