Week 4: Psychological Measurement Flashcards
Assignment of scores to individuals so that the scores represent some characteristics of individuals:
Measurement
What is a psychological measurement called?
Psychometrics
Psychological variable that represents an individual’s mental state of experience, often not directly observable:
Psychological Construct
What are personality traits, emotional states, attitudes and abilities?
Psychological constructs
Why are psychological constructs not observable?
- Represent general tendencies, not true 100% of the time.
- Often involve internal process which are unseen
What are the big five personality dimensions? OCEAN
- Openness to experience
- Conscientiousness
- Extraversion
- Agreeableness
- Neuroticism
Describes behaviours and internal processes that make up a psychological construct, along with how it relates to other variables:
Conceptual definition
Are conceptual definitions concrete?
No. They are proposed, tested empirically and revised as necessary by researchers
Definition of a variable in terms of precisely how it is to be measured:
Operational definition
What are the three categories of operational definition?
- Self report
- Behavioural measures
- Physiological measures (eg. heart rate)
Can constructs have multiple operational definitions?
Yes
When psychologists use multiple operational definitions of the same construct - either within a study or across studies:
Converging operations
When converging operational definitions produce same pattern of results:
Good evidence the construct is being measured effectively and is useful
What are the four levels of measurement?
- Nominal
- Ordinal
- Internal
- Ration
Measurement level for categorical variables with no order and assigning scores to a category:
Nominal level
Measurement level for lowest level of measurement:
Nominal level
What kind of measurement level is martial status?
Nominal level
Measurement level that assigns scores in rank order:
Ordinal level
Measurement level that is ordered:
Ordinal level
What is the issue with ordinal level of measurement?
Difference between scores not necessarily equal
Rankings of runners as first, second third etc is what kind of measurement level?
Ordinal level
Measurement level where scores on the numerical scale are even throughout but has no zero point:
Interval level
IQ test scores and Fahrenheit are what kind of measurement level?
Interval level
Measurement level which has a zero point on a numerical scale:
Ratio level
Measurement level which covers height or weight?
Ratio level
What refers to the consistency of a measure?
Reliability
What are the three types of consistency?
- Overtime (test-retest reliability)
- Across items (internal consistency)
- Across different researchers (inter-rater reliability)
What measure of consistency goes over time?
Test-retest reliability
What kind of consistency goes across items?
Internal consistency
What kind of consistency goes across different researchers?
Inter-rater reliability
When the scores are consistent across time, the construct has:
Test-retest reliability
What test-retest correlation score indicates good reliability?
+.80 or more
Trest-retrest reliability is measured using:
Correlation between two scores sets
Internal consistency is measured using:
- Split-half correlation
- Cronbach’s alpha
Consistency of people’s responses across the items on a multi-item measure:
Internal consistency
Splitting test items into two and examining the relationship between two sets of scores in order to assess the internal consistency of a measure :
Split-half correlation
What split-half correlation score indicates good internal consistency?
+.80 or more
The later developed statistic that measures the internal consistency among items in a measure:
Cronbach’s alpha
Mean of all possible split-half correlations for a set of items:
Cronbach’s alpha
What Cronbach’s alpha value indicates good internal consistency?
+.80 or more
Extent to which different observers are consistent in their judgements:
Interrater reliability
How is interrater reliability accessed?
- Cronbach’s alpha for quantitative
- Cohen’s K for categorical
The extent to which the scores from a measure represent the variable the way they are intended to:
Validity
A measure can be _________ but have no ________
A measure can be reliable but have no validity
What are the three types of validity:
- Face validity
- Content validity
- Criterion validity
Extent to which a measurement method appears superficially to measure the construct of interest?
Face validity
What is the weakest evidence of validity?
Face validity
Extent to which a measure covers the construct of interest:
Content validity
Is content validity assessed quantitatively?
No
Assed by checking the measurement method against the conceptual definition of the construct:
Content validity
Extent to which people’s scores on a measure are correlated with other variables (criterion) that one would expect them to be correlated with:
Criterion validity
Scores on a measure of text anxiety should be negatively correlated with performance in school tests:
Criterion validity
A variable that theoretically should be correlated with the construct being measured:
Criterion
Form of criterion validity where the criterion is measured at the same time as the construct:
Concurrent validity
Form of validity where the criterion is measured at some point in the future after the construct has been measured:
Predictive validity
Form of validity whereby new measures are correlated with existing established measures of the same construct:
Convergent validity
Extent to which scores on a measure of a construct are not correlated with a measure of other conceptually distinct constructs, and thus discriminate between them:
Discriminant validity
People’s scores on a measure of self-esteem should not be very highly correlated with their mood:
Discriminant validity
What is the four-step measurement process?
- Conceptually define the construct
- Operationally define the construct
- Implement the measure
- Evaluate the measure
What are the two ways a construct can be operationally defined?
- Use existing measure:
2. Create own measure: only if no existing measure exists
Participants responding in a way that is appropriate but not true when taking the test:
Socially desirable responding
Subtle cues that reveal how the researcher expects participants to respond to measure:
Demand characteristics
How to minimise false answers on a measure:
- Make procedure clear and brief
- Guaranteed anonymity
- Have a blind helper administer test
- Standardise all interactions
