Chapter 6: Surveys & Observations Flashcards
Surveys
People are asked about a consumer product
On the phone, in person interviews, paper/pencil questionnaire, online, etc
Asking a sample of people to answer questions
Poll
People are asked about social or political opinions
Open Ended Questions
Respondents may answer any way they like
E.g. what public figure do you admire most, your experience at a hotel
Advantages: provide spontaneous and rich info, may not have thought of the response yourself
Drawback: responses must be coded and categorized (difficult and time consuming)
Forced Choice Questions
People give their opinion by picking the best of two or more options
Political polls asking who you will vote for
Choose the statement you agree with (Narcissistic personality inventory, Likert-type scale)
Yes/no questions (adverse childhood experiences survey)
Present a statement and rate with a scale (Likert scale…strongly agree, agree, etc)
Semantic differential format: rate a target object using numeric scale anchored with adjectives
E.g. show up and pass 1 2 3 4 5 hardest thing I’ve ever done
E.g. five star reviews on google
Advantages: reduces the extra noise/details you don’t want, simplifies statistical analysis
Leading Questions
Wording leads people to a particular response
Word questions neutrally, avoid emotional terms
Avoid biased language
Ask multiple questions on the same topic
Double Barrelled Questions
Double barreled question: asks two questions in one
Complicated wording makes it difficult to respond in ways that accurately reflect opinions
Poor construct validity, could respond to first half, second half, or both questions
E.g. I look for main ideas and I formulate answers to questions I have as I read an assignment
Negative Wording/Double Negatives
Negatively worded questions: negative phrasing can cause confusion and reduce construct validity
Can ask the question both ways to study internal consistency
Might capture people’s ability to figure out the question and not their true opinions
E.g. do you oppose not allowing the board to pass article 10 of the ballot
Question Order
Previous question responses can influence later responses
E.g. do you favour or oppose affirmative action programs for women, do you favour or oppose affirmative action programs for racial minorities
If asked about women first, more likely to say yes to minorities
Should prepare multiple versions of survey, report results separately
Response Sets (non differentiation)
Type of shortcut people can take when answering survey questions
Do not cause problems for single questions
People may adopt consistent ways of answering all the questions in a long questionnaire
Reverse coding is good solution
Acquiescence (yes saying)
When people say yes to every item instead of thinking carefully about each one
Instead of measuring construct of true feelings, survey could be measuring tendency to agree/lack of motivation to think carefully
Include reverse worded items to see who really agrees with the items
Reverse wording slows people down to answer more carefully
Drawback: can result in negatively worded items which can be more difficult to answer
Fence Sitting
Playing it safe by answering in the middle of the scale (especially controversial items)
May occur when the question is unclear
Can weaken construct validity when middle scores suggest some people don’t have an opinion even though they do
Can be combated by removing the neutral option (e.g. scale of 4 instead of 5)
But some people truly don’t have an opinion so could invalidly represent their neutral stance
Forced choice questions (pick one of two) have same disadvantages as removing neutral option
Socially Desirable Responding/Faking Good
Giving answers that make respondents look better than they are
Faking bad: less common
Reminding participants of anonymity helps, but may cause respondents to take it less seriously
Minimize problem by including items to identify socially desirable responders
People who agree with these items, researchers can discard their data
Ask people’s friends to also rate them
Conscientious Responders Scale
To answer this question please choose “___”
Use questions that are obvious (e.g. I can read English)
Observational Research
When researcher watches people or animals and systematically records how they behave or what they do
Can be better than self reports (removes biases, influences on results, etc)
May get more objective data (e.g. self reports can be reported as more socially acceptable than were)
May not always be able to observe what you’re interested in (e.g. opinions, internal thoughts, feelings)
Researcher may be biased by what they expect to find
Observer Bias
When observers see what they expect to see
Observers expectations can influence interpretation of participants behaviours or outcome of the study
Show therapists the same tape of a man speaking with a prof, one condition say he is a patient and other condition say he is a job applicant
Job applicant condition described him as candid, attractive, innovative
Patient condition described him as tight, defensive
