Observations and Surveys (Pt.2) Flashcards
Open-ended questions
Allows respondents to say what they like, instead of choosing a response option (EX: What is helping you learn in this class? What is not working well for you in this class?)
What are the Pros and Cons of open-ended questions
Pros: Provide researchers w/spontanous, rich information
Cons: Responses must be coded (time-consuming), need knowledge on appropriate coding techniques
Forced-Choice Questions
Ask participants to choose the best of two (or more) options. Usually true/false or yes/no questions
(EX: I really like to be the center of attention, it makes me uncomfortable to be the center of attention, have you ever had a doctor make you feel unheard, etc.)
Likert Scales
Series of statements w/a rating scale participants use to indicate their agreement (EX: please rate the following by how much you agree)
Semantic Differential Format
Respondents rate a target using a numeric scale anchored w/ adjectives
(EX: Running is - then they’re given a scale of 1-5)
Well-Worded Questions
How we word items can impact the way someone responds
Leading questions or using “loaded” words
Wording leads people to pick one particular response over another (suggests a particular viewpoint), or simply using emotional terms
Avoid double-barreled questions
Asking 2 or more questions in one sentence
Avoid double negatives
When a question contains a double negative that leads to confusion
Avoid making assumptions about the respondent
“What kind of vehicle do you own”
Avoid non-inclusive reponse options
When the option you give participants don’t include all of the possibilities
(EX: How likely are you to get cancer in your lifetime - then given a very likely to very unlikely scale)
Be specific and precise in your phrasing and wording
(EX: “Do you support Biden?” to “Do you support President Biden’s student loan forgiveness plan?”)
Pretesting Questions
Run a pilot study to test items, calculate reliability estimates, ask participants for feedback, anything confusing or unclear
Question order
Questions administered earlier can affect the way participants respond to subsequent questions (AKA order effects) - Want to avoid starting w/sensitive topics, make sure to measure “stable” traits at the end, and be sure to measure the most important things first
Problems with self-report
People may not give completely accurate info because of many things: Minimizing memory errors, socially desirable responding, minimizing yea and nea saying, and fence-sitting
Minimizing Memory Events
Experience Sampling methods (ESM) - Participants report what they are thinking, feeling, and doing at multiple points over time
Socially Desirable Responding
When participants respond in a way that makes them look better than they really are - Minimizing this? Neutral questions (no right answers), Anonymous responding
Minimizing yea and nea-sayin
Include an equal number of true/yes and false/no questions for the same issue. Keep survey short
Fence-Sitting
Tendency to respond in the middle, such as “neither agree nor disagree” - Minimize this? No neutral response option, used forced-choice format, keep survey short
