Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

2801 Research Methods > Week 8 : Measurement > Flashcards

Week 8 : Measurement Flashcards

1
Q

What is measurement?

A
  • measurement is the assignment of scores to individuals so that the scores represent some characteristic of the individuals
  • psychological measurement is often referred to as psychometrics
  • measurement does not require any particular instruments or procedures, but… it does require systematic procedure for assigning scores to individuals or objects so that those scores represent the characteristic of interest
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are we measuring?

A
  • variables and constructs
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What are we measuring?

Variables…

A
  • things that vary
  • anything with a range of values (e.g. age, gender, shoe size, extroversion, aggression, depression, intelligence, etc.)
  • some are straightforward but some are not (aggression)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What are we measuring?

Constructs…

A
  • variables that cannot be observed directly (traits, emotion, attitudes, abilities)
  • cannot be directly observed cuz they represent tendencies to think, feel or act in certain ways and they often involve internal processes
  • we have ideas about what these things are but they are difficult to define & measure
  • The conceptual definition of a psychological construct describes the behaviours + internal processes
  • predictor variable…
  • dependent variable…
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Defining your variables…

conceptually vs operationally

A
  • conceptually define your construct… what do you think this construct IS?
  • operationally define your construct… how will this construct be measured? (self report, behavioural, physiological)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Types of measurement…

A
  1. self report
  2. behavioural
  3. physiological
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Types of measurement…

Self-report

A
  • interviews or questionaires
  • ppl report their beliefs, behaviour, history, etc.
  • pros: cheap, easy, straightforward
  • cons: social desireability bias e.g.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Types of measurement…

Behavioural measures

A
  • observations of behaviour
  • could be naturally occuring (e.g. flirting in a bar) or lab-induced (e.g. having ppl wear a heart monitor & record their heart rate)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Types of measurement…

Physiological measures

A
  • assessment of body states
  • e.g. brain imaging, heart rate, etc.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

How do you choose your measurement instrument?

A
  1. previous research
  2. methodological advances
  3. feasibility (resource limitations [e.g. time, money] may constrain your choice)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

4 levels of measurement?

A
  • these emphasize the genrality of the concept of measuremnt and they can serve as rouch guidelines to the stats procedures that can be used with the data
    1. nominal level
    2. ordinal level
    3. interval level
    4. ratio level
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

4 levels of measurement?

Nominal level…

A
  • used for categorial variables & involves assigning scores that are category labels
  • category labels communicate whether any 2 individuals are the same or different in terms of the variable being measured
  • they do not imply any ordering among the responses
  • nominal scales embody the lowest level of measuremnt
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

4 levels of measurement?

Ordinal level…

A
  • involves assigning scores so that they represent the rank order of the individuals
  • ranks communicate whether any two individuals are the same or different in terms of the varaible being measured and whether one individual is higher or lower on that variable
  • ordinal scales allow comparisons of the degree to which individuals rate the variable
  • ordinal scales fail to capture important info that will be present in the other levels of measurement we examine
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

4 levels of measurement?

Interval level

A
  • involves assigning scores using numerical scales in which intervals have the same interpretarion throughout (e.g. farenheit or celsius scales)
  • interval scales do not have a true zero point even if one of the scaled values happens to carry the name ‘zero’
  • since an interval scale has no true 0 point, it does not make sense to compute ratios of temperature
  • in psych the IQ is considered to be measured at the interval level
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

4 levels of measurement?

ratio level

A
  • involves assigning scores in a way that there is a true 0 point that represents the complete absence of the quality (e.g. height and weight)
  • think of a ratio scale as the three earlier scales rolled up in one… nominal: provides name or category for each object, ordinal: objects are ordered, interval: the same difference at two places on the scale has the same meaning
  • however, the same ratio at two places on the scale also carries the same meaning
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

4 levels of measurement?

Summary…

A
  • NOMINAL : category labels
  • ORDINAL : category labels & rank order
  • INTERVAL : category labels, rank order & equal intervals
  • RATIO : category labels, rank order, equal intervals & true zero
17
Q

Accuracy

A
  • no measure will be perfectly accurate (in psych)
  • true score = ‘real’ score on the variable
  • obtained score = the score the measure gives
  • measurement error = difference bewteen true score & obtained score
  • we want to minimize measurement error
18
Q

Reliability

A
  • does your instrument consistently measure the same thing? or does your measure give consistent results under the same conditions?
  • a scale should report the same weight each time, like questionnaire results should not change if taken twice
  • the results should not be identical but they should be similar
19
Q

How do you test the reliability of a self-report measure?

A
  1. test-retest reliability
  2. parallel-forms reliability (if applicable)
  3. internal consistency
20
Q

How do you test the reliability of a self-report measure?

Test-retest reliability

A
  • same test is given twice with some time in between
  • using the measure on a group of ppl at one time, using it again on the same group at a later time and looking at test-retest correlation between the 2 sets of scores
  • good for stable quslities like personality
  • not so good for temporary states (e.g. mood)
21
Q

How do you test the reliability of a self-report measure?

Parallel-forms reliability (if applicable)

A
  • Many behavioural measures involve significant judgment on the part of an observer or a rater
  • Inter-rater reliability is the extent to which different observers are consistent in their judgments
  • different forms of the same test used
  • if you take 2 different version of the test you will get a similar score on both
22
Q

How do you test the reliability of a self-report measure?

Internal consistency

A
  • internal consistency is the consistency of people’s responses across the items on a multiple-item measure.
  • test it with split-half correlation (top half of questionnaire is compared to the bottom half)… split-half correlation of +.80 or greater is generally considered good internal consistency
  • test it with Chronbach’s alpha… which tests how all the iterms are intercorrelated
  • Cronbach’s alpha is a score between 0-1 (0.90+ excellent, lower than 0.60 poor)
  • this is perfect for pinpointing items that do not belingin our scale (questions) and allow us to remove them to make it more reliable & accurate measure
  • do all the items in the scale measure the same thing
23
Q

How do you test the reliability of observational measures?

A
  • interrater reliability… multiple raters = reliability is testable
24
Q

Validity

A
  • does your instrument consistently measure the right thing?
  • are you measuring what you think you are measuring
25
Q

How do you test the validity of a measure?

A
  1. face validity
  2. content validity
  3. criterion validity
  4. discriminant validity
26
Q

How do you test the validity of a measure?

Face validity

A
  • does the measure look like it measures the thing its meant to measure?
  • if it is not reliable, it is not used
  • many established measures in psychology work quite well despite lacking face validity
27
Q

How do you test the validity of a measure?

Content validity

A
  • does the measure capture all the important facets of the construct?
  • assessed by carefully checking the measurement method against the conceptual definition of the construct
  • the whole range (might have to add other category)
28
Q

How do you test the validity of a measure?

Criterion validity

A
  • Criterion validity is the extent to which ppls scores on a measure are correlated with other variables (known as criteria) that one would expect them to be correlated with
  • convergent validity… does it correlate with similar variables
  • predictive validity… does it predict expected outcomes
29
Q

How do you test the validity of a measure?`

discriminant validity

A
  • Discriminant validity is the extent to which scores on a measure are not correlated with measures of variables that are conceptually distinct
  • your measure should NOT correlate with theoretically different variables
  • also referred to as ‘divergent validity’
  • if you do not achieve discriminant validity, your measure it likely too general or broad
30
Q

There are 4 steps in the measurement process…

A
  1. conceptually defining the construct
  2. Operationally defining the constructs
  3. Implementing the measure
  4. Evaluating the measure
31
Q

There are 4 steps in the measurement process…

conceptually defining the construct

A
  • having a clear and complete conceptual definition of a construct is a prerequisite for good measurement
  • it allows you to make sound decisions about exactly how to measure the construct
32
Q

There are 4 steps in the measurement process…

Deciding on an Operational Definition

A
  • using an existing measure
  • creating your own measure
33
Q

Deciding on an Operational Definition:

Implementing the measure

A
  • implement in a way that maximizes its reliability & validity
  • there might be reactivity, socially desireable responding, demand characteristics
  • to avoid this you have to… clear & brief, guarantee anonymity, standardize all interactions between researchers & participants

2801 Research Methods (11 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions