Week 4 - Identifying good measurement Flashcards
List the three main ways to create measurements?
Self-report
Observation
Psychological
Define observation as a measurement tool
Recording observable behaviours or traces of behaviours
Observational measure - measuring observable behaviour
Observational study - another word for correlational study
In self-report studies what are combined results referred to as?
how bout if the results are in the past?
Summary score or composite score
Retrospective report
What do psychological measures include?
what are some pros and cons of this method?
Brain activity, hormone levels, heart rate, eye-movement
Pros - less able to mislead, advances in tech make more accessible, provide insight into subtle effects that are harder to see with observations
Cons - time consuming and expensive, some techniques require million of dollars to operate, questionable links
Define categorical variables?
Levels are categories that have no particular order (ex: gender/political party)
Define ordinal scale?
Levels of a variable are in ranked order - don’t know how much different each level is from the other (intervals are not equal)
Define an interval scale?
Numbers represent equal intervals between levels but there is no true zero (having 0 on something does not mean nothing)
ex: IQ tests - 0 does not mean 0 intelligence
Define a ratio scale?
Numbers have equal intervals and a true 0
lol if you can’t count it, its prob ur mom.. jk a ratio scale
What is inter-rater reliability? how is this related with correlations?
Consistency between observers findings
We quantify test-retest reliability with correlations
Two raters should correlate positively with each other
How does internal reliability involve itself with multi-item self-report measures?
In multi-item self report measure, how consistent are participants responses to similar items
Example: Participants who say “I am satisfied with my life” should also agree with “the conditions of my life are excellent”
Define face validity and content validity?
Face validity - does it look like a good measure?
Content validity - does the measure assess all aspects of the theorized construct?
Define convergent and discriminant validity?
Convergent - does it correlate with similar measurements?
Discriminant - does it show no correlation with measures that are not similar?
Define criterion validity?
Criterion validity - Does the measure correlate with key behaviours/real world outcomes?
Describe the process of designing and evaluating a measure?
Hint: 7 steps
1 - conceptual and operational definition
2 - face and content validity
3 - internal reliability
4 - test-retest reliability
5 - interrater reliability
6 - convergent and discriminant validity
7 - criterion validity
Ordinal, interval and ratio scales are all examples of what kind of variable?
What is its opposite?
Quantitative variable - coded with meaningful numbers
Categorical variable - nominal, categories assigned with a number (ex: age, sex)