Week 5 (Standardised Measures)

Question 1

What does reliability mean psychometrics

Answer 1

Is the measure as free as possible from random error

Question 2

What does validity mean psychometrics

Answer 2

Are we measuring what we think we’re measuring

Question 3

What does unidimensionality mean psychometrics

Answer 3

Are we measuring just the one thing we want to measure, or have we ended up measuring other things too

Question 4

What does discrimination mean psychometrics

Answer 4

How well do our items distinguish between levels f the thing we’re measuring

Question 5

What does equivalence mean psychometrics

Answer 5

Does this measure perform the same way for different groups of people (e.g, teenagers and adults)

Question 6

What does norm-referencing mean psychometrics

Answer 6

How are scores distributed the population.

Question 7

Difference between random and systematic errors

Answer 7

Systematic errors do not cancel each other out, they accumulate over time

Question 8

How to reduce random error

Answer 8

-Repeat measurements and average them (though isn’t simple for psychological variables)

Question 9

How to reduce systematic error

Answer 9

-Use multiple measurements, each with different downsides
-That way the variable of interest is measured consistently, but the ‘noise’ is not
-This is why psychological instruments tend to take the form oof questionnaires rather than single questions.

Question 10

Reliability vs validity

Answer 10

Reliability: A measure is accurate and consistent
Validity: Measures what you think its measuring

Question 11

Types of reliability

Answer 11

-Test-retest
-Parallel form reliability
-Internal consistency

Question 12

Test retest reliability

Answer 12

-If I measure your extraversion today, will it be the same as when I measured it yesterday

Question 13

Parallel form reliability

Answer 13

Two alternative forms of the same standardised measure
Reduces risk of learning effects over time

Question 14

Internal consistency reliability

Answer 14

-Are all the items doing just as good a job as one another of measuring the psychological construct of interest.
-A sub scale should be unidimensional, so we look at whether every item on the sub scale measures the same thing
-Measured with cronbach’s alpha, between 0-1, anything above 0.7 indicates reliability.

Question 15

Test retest reliability pros and cons

Answer 15

PROS
-Demonstrates the measure s temporally stable

CONS
-Based on a total score (items could be completely unrelated)
-What about emotion or motivation
-How long between testing sessions

Question 16

Split-half technique

Answer 16

Quantifies internal consistency, split a set of items in half and check whether both halves correlate

Question 17

Internal consistency reliability pros and cons

Answer 17

PROS
-It’s essential poor internal consistency can only be due to items measuring different things
-Rubbish in, rubbish out

CONS
-If you increase the number of items, you increase Cronbach’s alpha- so exercise caution with very brief or very long measures
-Extremely high Cronbach’s alpha values might be ‘bloated’: too narrow a range of questions was asked.

Question 18

Inference validity

Answer 18

Internal: Can the causal relationship that we established be explained by other factors?
External: Can we generalise to other situations, populations

Question 19

Construct validity

Answer 19

-How well are we measuring what we want to measure
-Made up of translation and criterion validity

Question 20

Translation validity

Answer 20

Is the operationalisation a good reflection of the construct?
Face: Does the instrument APPEAR to measure the construct
Content: To what extent does the items actually represent the whole of the construct dimension we are trying to measure

Question 21

Criterion validity

Answer 21

How well does the measure agree with some external standard
Made up of
-Predictive validity
-Concurrent validity
-Convergent validity
-Discriminant validity

Question 22

Predictive validity

Answer 22

Does a score on the measure PREDICT the value of another variable in the future

Question 23

Concurrent validity

Answer 23

Does the measure NOW correlate with info from a related measure?

Question 24

Convergent validity

Answer 24

Does the measure correlate with another variable that it should theoretically be related to.

Question 25

Discriminant validity

Answer 25

Does the measure correlate with a conceptually unrelated construct (IQ and personality)