Lecture 2: Studying Well-Being Flashcards
How is happiness and well-being related?
They are latent constructs
Theoretical, abstract concepts that can’t be directly observed & measured
How do we measure happiness and well-being?
Operationalization - process of expressing theoretical variables into measurable variables
We meausure are indicators - measured variables that are used to infer information about latent constructs
What does each variable stand for? X = T +- E
X = happiness score
T = true happiness
E = Measurement Error - will cause x to be greater or less than your true level
What are the 2 types of Measurment Eror?
Random - random fluctuations in score, influence score in unpredictable ways, unrelated to construct of interest
Systematic - non random fluctuations in score, influence scores in predictable ways, related to construt of interest
What are the 2 criteria’s for measurement?
Reliability - reflects how much random measurement error there is in an individual’s score
- consistently measuring some construct
Validity - captures how much an individual’s score reflects their true score and systematic measurement error
- measures what it’s supposed to
List the ways of asssessing reliability.
- Split-half reliability
- Test-retest reliability
- Inter-rater reliability
- Internal consistency
What is split-half reliabilty?
Randomly create 2 subsets from a single measure
& correlate avg score of subset 1 with avg score of subset 2 (software will mix/match the questions in 2 groups)
- the 2 halves should be consistent with each other
If there is NO RANDOM MEASUREMENT ERROR then: correlation (r) should be pos
- close to +1 is good but at least 0.7 would be okay
What is test-retest availability?
Administer measure at diff time points & find correlations between the time points - look at correlation between avg at time 1 & time 2 (r= +1)
Ex. give a questionnaire now then give the same one a week from now
Short-term intervals - tells us about reliability (personality trait like self-esteem shouldn’t change in a day)
Long-term intervals - tells us about stability
What is Inter-rater reliability
Different people make simultaneous ratings of the same target & find agreement across raters
If there is no random error, KAPPA = 1 (0.7 is good)
What is internal consistency?
Using multiple items in the measure, look at how items correlate with each othe
Chronbach’s a - is the avg of all the correlations on the questionnaire
No random error = Chronbach’s alpha = 1
What are different ways of assessing Validity?
- Content validity
- Construct validity
- Criterion validity
- Concurrent (convergent and discriminant) and predictive validity
- Internal/external validity
What is content validity?
looking at the content of your measure and deciding if that’s mesuring what u want it to measure (Ex. looking at the questionnaires, face validity)
- Can be biased about it
What is Construct validity?
Something other than our own opinion, can use theory for guidance, can use a criteria that should or shouldn’t be related
Ex. factor structure
- if you want to measure personality, use the 5 thoeries: Extraversion, neuroticism, openness, agreebleness, conscientiousness.
- each group should have qs from each theory - see if responses match up with theories
What is criterion validity?
Measuring all variables at the same time
Concurrent validity includes convergent and divergent validity
Convergent validity - how much scores on your measurement correlate with the scores on a diff measure of the same construct - want to see the same correlation
Divergent - measuring something we don’t want to measure, we want to discriminate it
Ex. so if correlation for measuring self-esteem is 0, doesn’t tell u, your measuring happiness but atleast tells u your not measuring self-esteem
What is predictability?
If there is a strong correlation between your scores on my questionnaire and what you votes for
What is Internal Validity?
Is it my independent variable causing the changes to my dependent variable(s)
so random assignment and control
What is External validity?
Do the results of the study generalize to:
- the wider population
- real world situations - (research done in the lab to real world situations), MUNDANE REALISM - ex. basment converted to prison
- Real world psychological experiences, Psychological REALISM - somethings we can’t measure
Replicability is to consistency as…
Validity is to Accuracy
What are close-ended responses
Experimentor provides response options like rating scale
Advantage: not time consuming and inexpensive
Disadvantage: we get limited info, can’t capture the underlying dynamics
- no. of responses, order of responses, & response labels influence reliability & validity
What are open-ended questions
Free to respond how they want, usually through a list or describe their thoughts and feelings about their own well-being. Ex questionnaire or focus group
Advantages: captures the underlyin dynamics, more flexibility
Disadvantages: Interpretation of those responses may be subject to bias, need to use a software to code those responses
- labour extensive, time consuming & expensive
What 2 things are necessary for a rating scale?
Translational ease - no. of points should match participants psychological distinctions (like, neutral, dislike)
Clarity of Meaning - clarity may be compromised when a large no. of points are used (Ex. on a scale of 101, what does 76 mean to u?)
How does wording of questions/responses influence people’s evaluations?
- The MEANING of the question can be different for diff people (make sure meaning for the participants is the same as what u intent to be)
- People are MOTIVATED to response to a question/statement or response in a certain way (can have response biases)
What is the Satisficing respondent motive? why does it occur?
Tendency to look for cues in the qs or response options that point to a reasonable answer (what’s the most reasonable answer and how do I get it quickly and easily)
Why? We want to put in little effort thinking about an accurate answer (less time and energy)
- we don’t have sufficient cognitive resources or skills to think about an accurate answer (tired)
Avoiding this? Instead of confirming them to think how u already think, ask general questions (how easy did u find this task vs. how they found the task and putting very difficult on the scale first)
What is the Acquiescence respondent motive? why does it occur?
Tendency to agree with an assertion made in a question/statement regardless of the assertion (sometimes we just want tosay yes to everything)
Why? want to be polite and avoid confrontation, difference to higher social status individuals, satisficing
Avoiding this? vary the questions, change the wording. Ex. asking about the same concepts by neg wording them
