Week 3 Flashcards
Issues in Survey RM (21 cards)
What type of research aims to describe individuals (thoughts, opinions/feelings) + offer predictions about them based on demographics, and involves asking questions and/or observing individuals directly.
used to study associations (correlations) but not causations.
Survey research
Is there manipulation of variables involved in surveys?
No
This is a description of what type of question: pp is free to respond how they want, higher ecological validity, potentially more representative of person’s true opinions, difficult to analyse.
Open questions.
This is a description of what type of question: pps have to select a pre-specified response (e.g. 1-7), useful for statistical analysis, easier to score/analyse, more susceptible to designer bias
Closed questions.
Semantic differential scale:
Developed by Osgood (1952), asks people to rate something on a number of paired dimensions (1-10 inferior/superior, unlikeable/more likeable). Takes advantage of people’s ability to think metaphorically, more indirect.
Likert scale
Used to express extent of agreement, frequency etc. Offer different strengths of opinion, often used in standardised measures to produce scores. Direct form of questioning.
Categorical answers
Short questions offering a range of options or categories. Often used for demographic information. Not typically used on standardised measures (with some exceptions - yes/no format scales)
Ambiguous questions, technical terminology, leading questions, hypothetical questions and double/barrelled questions, are examples of what?
Common errors when writing questions.
Measurement invariance (equivalence) captures the degree to which
Your measure is testing the same thing across conditions. useful for if you’re testing for differences across time or across conditions.
If I measure your height today, will it be the same as yesterday? is an example of what
Test-retest reliability
If I measure you, will your height be the same as when Scott measured you? (cohen’s kappa) is an example of what
Inter-rater of reliability
Will your height be the same if I use a tape measure as when I use a ruler? is an example of what
Inter-method (or parallel-forms) reliability
Did you respond to similar questions in a similar way (e.g. when measuring extroversion) is an example of what
Internal consistency reliability (split-half technique, Cronbach’s alpha or McDonald’s Omega)
Is this a valid means of tapping into the construct you are attempting to measure? What type of validity
Construct validity
The extent to which the domain of interest is adequately represented by the scale items. What type of validity
Content validity
What validity is able to predict behaviour in the future?
Predictive validity
What validity assesses if the score is related to another criterion of this construct tested at the same time?
Concurrent validity
What validity shows if the score related to other measures of the same construct
Convergent validity
What validity determines if the score is different from measures that theoretically measure something else?
Discriminant validity
Strengths of questionnaires are
Access to large numbers of people with different demographics
Can gather very large sample sizes
Efficient and inexpensive to administer
Weaknesses of questionnaires are
No, or limited, explanation (follow-up research needed)
Susceptible to bias (respondents and researchers)
Have to be very carefully designed
Analysis = simple, interpretation = difficult