Chapter_15_Survey Research

Question 1

Question Wording

Answer 1

1. Use the LANGUAGE of the Research Participants
2. Avoid Unnecessary NEGATIVES
3. Ask Only ONE Question at a Time
4. Avoid LEADING and Loaded Questions
5. Be SPECIFIC
6. Do Not Make ASSUMPTIONS
   - familiarity with the topic
7. Address SENSITIVIE Topics Sensitively

Question 2

Leading question

Answer 2

implies that a certain response
is desired

Question 3

Loaded question

Answer 3

uses emotional content to evoke a desired response
- e.g. appealing to social values, such as freedom, and using terms with strong positive or negative connotations

Question 4

Sensitive Questions

Answer 4

1. Intrusiveness
2. Threat from disclosure
3. Social sensitivity

Question 5

Examples of Intrusiveness

Answer 5

personal or household income

Question 6

Examples of Threat from disclosure

Answer 6

criminal behavior

Question 7

Examples of Social sensitivity

Answer 7

under report
- Alcohol consumption
- Smoking
over report
- Energy conservation
- Physical exercise

Question 8

ways to avoid biased responses

Answer 8

• Imply that the behavior is COMMON
• Assume the behavior and ask about FREQUENCY or other details
• Use AUTHORITY to justify the behavior

Question 9

To avoid over-reporting of socially desirable behaviors

Answer 9

• Be casual
• Justify not doing something

Question 10

Measurement Levels

Answer 10

1. Nominal
2. Ordinal
3. Interval
4. Ratio

Question 11

Considering Measurement Levels

Answer 11

1. Information Content
2. Statistical Tests
3. Ecological Validity

Question 12

Information Content

Answer 12

1. Higher level measures contain more information
2. invalid conclusion if drawn on the basis of a lower level
3. more sensitive to the effects of IVs

Question 13

Statistical Tests

Answer 13

robustness
- assumptions underlying a particular statistic can be violated without leading to erroneous conclusions

Question 14

Ecological Validity

Answer 14

the natural level of measurement of a variable may not contain enough
information to answer the research question or may not be appropriate to the statistics

Question 15

Response Formats

Answer 15

1. Comparative Rating Scales
2. Itemized Rating Scales
3. Graphic Rating Scales
4. Numerical Rating Scales

Question 16

Comparative Rating Scales

Answer 16

• all possible pairings
  con
• too many pairs
  solution
• rank order
  conditions
• familiar
• unidimensional
• understand the meaning

Question 17

Itemized Rating Scales

Answer 17

• multiple-choice questions
• classification
• assess hypothetical constructs
• nominal/ordinal
  cons
• skip the item that doesn’t include
• force to choose even no preferred choice

Question 18

2 Factors determine the number of points to use on Numerical Rating Scales

Answer 18

1. sensitivity of measurement
   - detect small differences in the level
2. usability of the scale
   - A very large number of scale points can be counterproductive

Question 19

Choosing a Response Format

Answer 19

• different formats are highly correlated
• format might not have a large effect

Question 20

Advantages of Multi-Item Scales

Answer 20

1. assess multiple aspects
   of a construct
2. total score on a multi-item scale has greater reliability and validity
   - Adding an invalid or unreliable item decreases the scale’s reliability and validity
3. greater sensitivity of measurement

Question 21

Types of Multi-Item Scales

Answer 21

1. Likert Scales
2. Thurstone Scales
3. Guttman Scales
4. Semantic Differential

Question 22

Likert Scales

Answer 22

1. Write a large number of items
2. Administer the items to a large number of respondents
3. Conduct an item analysis
   - internal consistency
4. The items with the highest CITC comprise the final scale

Question 23

Thurstone Scales

Answer 23

Items represent attitudes from highly positive through neutral to highly negative
criteria
1. They must represent the entire range of attitudes
2. they must have very low variance in their judged favorability

Question 24

Disadvantages of Thurstone Scales

Answer 24

1. difficult to create
2. assumes unidimension
3. raters’ attitudes influence their assignment of scale values to items
4. the scale values that judges assign to items can change over time

Question 25

Guttman Scales

Answer 25

- a set of ordered attitude items - a respondent will agree with all items up to a point - and disagree with all items after that point

Question 26

Semantic Differential

Answer 26

rate the concept on sets of bipolar adjective pairs cons ensure that the adjective pairs one chooses for scale items are relevant to the concept being assessed

Question 27

Response Biases

Answer 27

1. Question-Related Biases 2. Estimation Biases 3. Ecological Validity

Question 28

Question-Related Biases

Answer 28

1. Scale AMBIGUITY - label as "frequency" -> numbers 2. Category ANCHORING - respondents use the amounts provided in the response options as cues for what constitutes an appropriate response 3. option PHRASED - in terms of “more than” a certain amount as representing an excessive amount 4. ESTIMATION Biases 5. Respondent INTERPRETATION of Numeric Scales 6. PRIMACY and Recency Effects

Question 29

Estimation Biases

Answer 29

Open-ended questions asking for frequency and amount - how often it occurs rather than count specific instances of the behavior

Question 30

Respondent Interpretation of Numeric Scales

Answer 30

‘0,’ respondents interpreted it to reflect the absence - with the numeric value ‘−5,’ and the scale offered ‘0’ = explicit failures

Question 31

Primacy and Recency Effects

Answer 31

- written questionnaires: people tend to choose an item from near the top of a list of response options - during interviews: respondents are more likely to choose items near the end of the list solution - make simple - random order

Question 32

Person-Related Biases

Answer 32

1. Social Desirability 2. Acquiescence 3. Extremity

Question 33

2 Forms of Social Desirability

Answer 33

1. Self-deceptive enhancement - unknowingly - part of personality 2. impression management - intentionally - depends on motivation

Question 34

Acquiescence

Answer 34

tendency to agree or disagree with statements - lack the skill or motivation - requires a great deal of thought - unsure about how to respond

Question 35

Controlling for acquiescence

Answer 35

balanced measure - half “positive items” - half “negative items”

Question 36

Extremity

Answer 36

give extreme responses on a measure such as by using only the end points

Question 37

Cultural Response Sets

Answer 37

a cultural tendency to respond in a certain way on tests or response scales - measurement confound - systematic error

Question 38

Collectivists

Answer 38

- stronger social desirability - higher on impression management - high acquiescent - prefer using the middle range

Question 39

Individualists

Answer 39

- higher on the self-deceptive positivity

Question 40

literal interpretation fallacy

Answer 40

taking the meanings of scale anchors at face value - unwilling to go any lower

Question 41

Question Order

Answer 41

1. Question Sequencing 2. Context Effects

Question 42

Question Sequencing

Answer 42

- start with easy, interesting, important, and related to the stated purpose - end with demographic - general to the specific

Question 43

Context Effects

Answer 43

responding to one question affects the response to a later question - thoughts or emotions aroused

Question 44

Constructing Questionnaires

Answer 44

1. Use closed-ended questions as possible 2. Use a consistent item format 3. Use a vertical item format whenever possible 4. indicate where one question ends 5. Do not split questions or response options between pages

Question 45

Using Existing Measures

Answer 45

1. use a single response format for all items 2. Context Effects

Question 46

Context Effects

Answer 46

occur when research participants fill out two (or more) questionnaires they believe are related solutions - having two data collection sessions - presenting them as belonging to different studies - counterbalance the order