Chapter 7

Question 1

A survey is a _________ way to ask close-ended questions with fixed response categories (variables) and should be _______ and _______

Answer 1

highly structured; mutually exclusive (2 things can’t happen at the same time); exhaustive (cover all bases)

Question 2

A survey requires ______ sample sizes and _______ number of questions

Answer 2

larger; larger

Question 3

benefits of a survey

Answer 3

1. documents change over time
2. time & cost effective
3. breadth of topics

Question 4

Types of surveys (timing)

Answer 4

• cross sectional
• longitudinal
• repeated cross sectional (limit; no cause-effect or analyzing behaviour overtime)
• panel (limit; attrition)

Question 5

you must ask the same question ____ to compare people overtime

Answer 5

everytime

Question 6

How do surveys document change overtime and what types?

Answer 6

• within-person (panel design)
• societal trends (repeated cross-sectional)

Question 7

How are surveys time and cost effective

Answer 7

• standard formats
• short questions & multiple choice
• a large sample is possible if sample is representative of the population or have representativeness

Question 8

how can a large sample size be representative and what does it need?

Answer 8

• no selection bias (random selection)
• stratified sampling (same proportions of population)
• weighting
• NEEDS high external validity

Question 9

the quality of a survey depends on _____.

Answer 9

sampling

Question 10

how can surveys cover a breadth of topics

Answer 10

• data for many studies
• primary data
• secondary data

Question 11

What is primary data?

Answer 11

create and conduct survey

Question 12

what is secondary data?

Answer 12

survey data collected by others and published so others can use it (GSS)

Question 13

Modes of Surveys

Answer 13

face-face
telephone
mail or self-administered questionnaires
online survey

Question 14

Face-face survey strengths and weaknesses

Answer 14

high: cost, response rate, researcher control over interview, interviewer effects

Question 15

what are interviewer effects?

Answer 15

social desirability bias

Question 16

telephone survey strengths and weaknesses

Answer 16

moderate cost
high response rate
moderate research control and interviewer effects

Question 17

mail or self-administered questionnaires strengths and weaknesses

Answer 17

low cost, response rate, researcher control over interview, interviewer effects

Question 18

online surveys strengths and weaknesses

Answer 18

low: cost, interview effects
moderate: response rate, researcher control over interview

Question 19

mixed-mode surveys can help ________ and _______

Answer 19

assess quality; potential bias

Question 20

mixed-mode surveys rely on

Answer 20

logical reasoning and existing research

Question 21

Survey samples need a ________

Answer 21

sampling frame (random selection, stratified sampling)

Question 22

screening surveys create issues of _______.

Answer 22

non-response

Question 23

issues with survey samples

Answer 23

sampling frames sometimes don’t exist
characteristic you want may not be listed

Question 24

you can use ______ to calculate margin of error

Answer 24

statistics

Question 25

non-probability samples have ________.

Answer 25

systematic errors (convenience sample, can't estimate)

Question 26

Ethical considerations

Answer 26

- minor psychological distress - indirect harm - most surveys aren't sensitive - confidentiality - anonymity - respect, beneficence, justice

Question 27

what can go wrong in a survey?

Answer 27

1. survey content can lead to measurement error (validity and reliability) 2. organization 3. order effect 4. priming effect

Question 28

the organization of a survey uses what

Answer 28

screener questions (questions that determine if they go to the next round of questions) skip patterns (ask a respondent and if they answer a question well you go onto more) avoid monotony (similar questions)

Question 29

how to avoid monotony in a survey

Answer 29

switch type of questions (ask question then the opposite to see if answers conflict)

Question 30

how to fix the order effect (placement of questions creates logical answer for next question)

Answer 30

put them in different sections (demographic)

Question 31

priming effect

Answer 31

first question influences second answer (putting words in someones mind)

Question 32

close-ended vs. open-ended

Answer 32

close-ended: mutually exclusive, exhaustive open-ended: qualitative data

Question 33

what is a partially open-ended question?

Answer 33

other, specify cons: might not answer the question

Question 34

what to avoid in survey questions

Answer 34

double negative double barreled leading questions acquiescence bias

Question 35

what is a double negative and how to fix it?

Answer 35

(option of yes or no, not understanding what answer means) Fix: question simple to answer

Question 36

what is a double-barreled question and how to fix it?

Answer 36

(having two variables that can have different answers. E.g. how much confidence do you have in president’s ability to handle domestic & foreign policy) Fix: split question apart

Question 37

what is a leading question and how to fix it?

Answer 37

(people at UWO love their jobs. How do you feel about your job at UWO?) Fix: only ask the question, do not guide them to answer

Question 38

what is acquiescence bias and how to fix it?

Answer 38

(tendency for people to agree with everything; lower confidence leads to agreeing to everything) Fix: forced-choice version (ask respondent to choose, don’t ask them an agree-disagree question)

Question 39

sensitive words like rape require _______

Answer 39

different phrasing; use sexual relations against your will; but DO NOT use lengthy complex terms

Question 40

for surveys you should use ______ and ______ methods

Answer 40

tried; true

Question 41

single item vs. composite

Answer 41

single: 1 question 1 variable (household income) composite: identify all survey questions, tell them how you combine them (more yes's leads to overall answer)

Question 42

how do you assess measurement errors?

Answer 42

validity and reliability internal and external

Question 43

what other errors besides measurement error can you encounter in surveys and how do you fix them?

Answer 43

nonresponse fix: maybe its the wrong method to find ppl, post-survey weighting, drop useless respondents coverage fix: add another screening question sampling fix: nonprobability sample (won't be representative)

Question 44

what is pre-testing and why do you use it?

Answer 44

recruiting people to answer the survey that are similar to the target population; used to identify errors

Question 45

how to conduct pre-testing

Answer 45

cognitive interview (get their input on survey) preliminary data analysis (statistical, distribution, average value, detect anomalies)

Question 46

preliminary data analysis tests what and what makes someone need to do more than 1 test?

Answer 46

statistical test, distribution, average value, detect anomalies - if survey is mostly tried and true only 1 is needed - if new measures, need more than 1 test

Question 47

can the data from pre-testing be used in the actual survey?

Answer 47

NO (only helps develop survey)

Question 48

a sampling frame should include....

Answer 48

Who will you interview and why? How will participants be chosen? What safeguards will you take to ensure that your sample is not biased?