Psychology : Psychometric Report

profileKyra2795
PSY4046andPSY3018Week15-NormalisingpersonalityscoresandtakingEQtestNEW.pdf

29/01/2019

1

PSY4046 and PSY3018

SESSION 5 Normalising and Understanding

Personality

Dr Jackie Meredith and Neelam Ghuman

1

The importance of Norms • Last week we covered some of the important issues regarding

psychometric tests – reliability and validity. • An important aspect of validity is understanding scores in CONTEXT. • Raw scores indicate the percentage of questions you answered positively

or negatively • They don’t indicate how you have scored in comparison to what would be

expected • For example in an intelligence test (IQ) your initial RAW score will be

compared to the body of people like yourself (age, gender, maybe other aspects) who have taken the test previously

• Norms are essential for understanding any distribution: • BMI, ability, university performance, GCSE and A Level performance, response

times, reaction times…

• The process of conversion is called STANDARDISATION or NORMALISATION of scores.

2

29/01/2019

2

Normalisation and Generalisation • Normalisation = process of determining what ‘normal’ is • Problem = how much is ‘normal’ relative?

• Generalisation = process of applying a principle or an observation from a sample to a wider group • At the ‘nth’ degree, this may become a rule

• Like BMI indications…

• Generalisability in psychology = the extent to which we can apply our findings to a wider population. • For example, when as psychologists we design a laboratory experiment

this may limit how much we can generalise the findings to all people in natural settings.

• How significant do you think the choice of NORM group is to the generalisation of a psychometric test?

3Think how you will explain this to the client….

How to calculate norms • Using formulae, a new score will be calculated that represents how

well the individual performed compared to what was expected of them • Because it has been observed in others like them.

• This NORMALISED score is your true score • There is a standard procedure for calculating norms • z scores

• But in order to make the score accessible to a non-statistician a further process is used • Known as the ‘t score’ formula • This is the accepted formula in education and business and is used to calculate IQ

norms, SAT norms, etc. from Z scores. • Ravid, R. (2005). Practical Statistics for Educators (3rd ed.). NJ: University Press of America.

• Not related to the ‘t-test’… 4

29/01/2019

3

PSY4046 : Think Method Section • The issue of normalisation is an important feature to

explain to your client. • How significant do you think the choice of NORM group

is to the generalisation of a psychometric test? • NB – this normalising process is often called

‘standardisation’ – this is not the same as standardised instructions, but the word reflects the fact that these are both ‘standard’ procedures. In your profiles you’re required to explain the standardising process – firstly in terms of standardised instructions and processes, and secondly in terms of normalised scores. These are separate things.

TE R

M 1

- PE

R SO

N AL

IT Y

W O

R KS

H O

PS

5

PSY3018 : Think Process Section • The issue of normalisation is an important feature which

supports the validity of psychometric testing • Make sure you explain the importance of these

processes in your section about the process of testing. • NB – this normalising process is often called

‘standardisation’ – this is not the same as standardised instructions, but the word reflects the fact that these are both ‘standard’ procedures. In your profiles you’re required to explain the standardising process – firstly in terms of standardised instructions and processes, and secondly in terms of normalised scores. These are separate things.

TE R

M 1

- PE

R SO

N AL

IT Y

W O

R KS

H O

PS

6

29/01/2019

4

Comparing a single score with Population Norms

• Calculating a single Z score Z = (x – μ)

σ • Calculating a T score (standardised score)

T = (Z score x 10) + 50

Where: x = any raw score μ = the mean of a comparable population of scores σ = the standard deviation of a comparable population

7

WHICH NORM? • Published measures will employ NORMS TABLES • There are no norms tables for the IPIP • We will need to normalise scores ourselves • by hand

• This should be a population equivalent in education and with a similar variation in participant qualities to our client

• We have chosen 2nd year psychology undergraduates • Comparatively well educated, interested in psychology, large sample of

scores available

• What features will this participant sample have?

8Let’s look at the descriptive statistics for the 349 undergraduates in our normative sample…

29/01/2019

5

Principles of the Normal Distribution

9

Extraversion

95.0 90.0

85.0 80.0

75.0 70.0

65.0 60.0

55.0 50.0

45.0 40.0

35.0 30.0

60

50

40

30

20

10

0

Std. Dev = 13.17 Mean = 66.9

N = 349.0 0

Agreeableness

100.0 95.0

90.0 85.0

80.0 75.0

70.0 65.0

60.0 55.0

50.0

80

60

40

20

0

Std. Dev = 9.52 Mean = 78.5

N = 349.0 0

Conscientiousness

100.0 95.0

90.0 85.0

80.0 75.0

70.0 65.0

60.0 55.0

50.0 45.0

40.0 35.0

30.0

70

60

50

40

30

20

10

0

Std. Dev = 12.96 Mean = 68.1

N = 349.0 0 NEUROT

100.0 95.0

90.0 85.0

80.0 75.0

70.0 65.0

60.0 55.0

50.0 45.0

40.0 35.0

30.0 25.0

20.0

50

40

30

20

10

0

Std. Dev = 15.34 Mean = 50.8

N = 350.00

Openness to Experience

100.0 95.0

90.0 85.0

80.0 75.0

70.0 65.0

60.0 55.0

50.0 45.0

40.0

70

60

50

40

30

20

10

0

Std. Dev = 10.31 Mean = 72.6

N = 349.0 0

Emotionality

Central tendencies & margins of error

• Because of the principle of Central tendencies, most scores in a normal distribution will fall within 1 standard deviation of the mean (about 68%) – this portion, in psychometrics, is considered the AVERAGE range.

• Personality tests, like all psychometric tests, are not an EXACT SCIENCE.

• For this reason, your exact score is not interpreted, but the RANGE your score falls into…

• In a test-retest situation a participant’s scores should fall into the same RANGE if the measure is of a concept that should remain broadly stable

• An introvert won’t become and extravert over the course of a couple of months… • Which is a weakness of TYPE tests like the MBTi…

TE R

M 1

- PE

R SO

N AL

IT Y

W O

R KS

H O

PS

10

29/01/2019

6

Comparing with the norm group - Ranges • Scores under 30 = VERY LOW

• Compared to psychology undergraduates, this is an extreme personality characteristic – e.g. VERY INTROVERTED

• Scores under 40 = LOW • Compared to psychology undergraduates, this is a distinct personality characteristic – e.g.

An INTROVERT • Scores from 40-44 = LOW AVERAGE

• This score is still within the average range of scores, but this personality feature is amongst the more notable of the this average range – e.g. amongst the more introverted.

• Scores from 45-54 = AVERAGE • The individual is very much like the majority of others in the norm group – a blend of

personality characteristics; may demonstrate a variety of characteristic depending on the environment – e.g. neither introverted nor extraverted, but can show some characteristics of each depending on the circumstances

• Scores from 55-59 = HIGH AVERAGE • This score is still within the average range of scores, but this personality feature is amongst

the more notable of the this average range – e.g. amongst the more extraverted. • Scores from 60 – 69 = HIGH

• Compared to psychology undergraduates, this is a distinct personality characteristic – e.g. An EXTRAVERT

• Scores of 70 and above = VERY HIGH • Compared to psychology undergraduates, this is an extreme personality characteristic –

e.g. VERY EXTRAVERTED

TE R

M 1

- PE

R SO

N AL

IT Y

W O

R KS

H O

PS

11

Normalisation and Generalisation • Normalisation = process of determining what ‘normal’ is • Problem = how much is ‘normal’ relative?

• Generalisation = process of applying a principle or an observation from a sample to a wider group • At the ‘nth’ degree, this may become a rule

• Like BMI indications…

• Generalisability in psychology = the extent to which we can apply our findings to a wider population. • For example, when as psychologists we design a laboratory experiment

this may limit how much we can generalise the findings to all people in natural settings.

• How significant do you think the choice of NORM group is to the generalisation of a psychometric test?

12Think how you will explain this to the client….

29/01/2019

7

A word about Averages...

13

0 25 50 75 100

l Now that scores are standardised, the mean is 50. l This is for ALL FACTORS as we have used the same formula for each. l Which RANGE each score falls in to is now important:

l For T scores, 45-54 is average for a British Undergraduate. Around 40-44 is low average, 39-30 is low, below 30 is very low.

l Around 55-59 is high average, 60-69 is high, and 70 and over is very high in comparison with your peers.

¢ Law of Averages = Most people will ALWAYS score WITHIN AN AVERAGE RANGE

A word about Averages...

14

0 25 50 75 100

l For example, you may have a raw score of 56 for Agreeableness. This sounds average. However, your T score is 26. In other words, compared to other undergraduates, your level of Agreeableness is actually very low.

l A low score will become more extreme as most undergraduates are scoring higher than you.

¢ Law of Averages = Most people will ALWAYS score WITHIN AN AVERAGE RANGE

29/01/2019

8

A word about Averages... ¢ Most people = AVERAGE ¢ AVERAGE = Normal!

15

0 25 50 75 100

• For example, you may have a raw score of 73 for Conscientiousness. This seems high. However, your T score is 54. In other words, compared to other undergraduates, your Conscientiousness is average.

• A high score value has been moderated because most undergraduates also score highly in this area – it is what is expected for someone studying for a degree...

Average can be an advantage in terms of personality, as it indicates a balance between two extremes.

THE FIVE FACTOR INTERPRETATIONS

16

¢ Extreme scores demonstrate strengths and challenges which go with strong characteristics.

¢ Average scores indicate a balanced profile – so aspects of both dimensions of each factor apply.

¢ Low average and High average scores indicate a balanced profile but a predominance of one aspect compared to others in the normed group

¢ Low Average = amongst the lowest in the expected (average) range

¢ High Average = amongst the highest in the expected (average) range

29/01/2019

9

CALCULATING NORMALISED SCORES

TE R

M 1

- PE

R SO

N AL

IT Y

W O

R KS

H O

PS

17

YOUR CALCULATIONS You need to apply the formulae to 5 items:

• IPIP scales • Extraversion…………….. Mean: 66.87 SD: 13.17 • Agreeableness………….. Mean: 78.51 SD: 9.52 • Conscientiousness……… Mean: 68.07 SD: 12.96 • Emotionality…...................Mean: 50.70 SD: 15.32 • Openness ........................Mean: 72.55 SD: 10.31

NOTE:

Means and SD’s are different for each factor

Which has the highest variation in scores?

29/01/2019

10

FOR EXAMPLE... • Your score for extraversion was 65. • This seems quite high. • The mean for the cohort is 66.87; SD 13.17 • Apply Z score formula:

Z = (x – μ) Z = (65 – 66.87) = -1.87 = -0.1419893** σ 13.17 13.17

• Apply T score formula: T = (Z score x 10) + 50 so... (-0.1419893 x 10) + 50

so... -1.419893 + 50 = 48.580107

FOR THE TABLE ROUND TO 2DP: Z = -0.14, T = 48.58 FOR THE CLIENT PROFILE ROUND TO 0DP: T = 49

Your standardised score is 49. Compared to the average British undergraduate, you are within the average range – neither totally extravert or totally introvert and no different to most undergraduates!

**NOTE - IT IS IMPORTANT TO RETAIN NEGATIVE SCORE IF YOU HAVE ONE AND TO KEEP THE DECIMAL PLACES INTACT WHILST YOU ARE CALCULATING

• Extraversion • Mean: 66.87 SD: 13.17

• Agreeableness • Mean: 78.51 SD: 9.52

• Conscientiousness • Mean: 68.07 SD: 12.96

• Emotionality • Mean: 50.70 SD: 15.32

• Openness to Experience • Mean: 72.55 SD: 10.31

**IPIP CALCULATIONS **

20

Calculating your Z score Z = (x – μ)

σ

Calculating your T score

T = (Z score x 10) + 50

Where x = single score

μ = mean

σ = standard deviation (SD)

29/01/2019

11

Mean score, Average range... 21

MEAN High Average

Low Average

Average Average!

USE THIS GRAPH TO PLOT YOUR SCORES AND IDENTIFY THE RANGE

IPIP 5 FACTOR SCORES

22

IPIP SCORES

Raw score

Sample Statistics (n=349)

Z score T score

Mean SD

Extraversion 66.87 13.17

Agreeableness

Conscientiousness

Emotionality

Openness

Now you can complete the table TO 2 DECIMAL PLACES

29/01/2019

12

Understanding your personality • Finalists – you can use your profile information in your work in

PSY3009 Individual Differences. • You will have your own table to complete relating to scores:

23

Factor Score Meaning Extraversion Average Gerard’s average score indicates he is neither and extravert nor an introvert.

His energy levels are much like everyone else, and he is confident in some situations but quiet in others. He gains energy both from time alone and from socialising with close friends ….. ….. …..

PSY4046: YOUR IPIP RESULTS • Introduce your table. • Complete the table • The table presents the process of standardising or normalising your client’s

scores. • Explanatory Comment • Explain to the client how they should understand the table – i.e. focus on

standardised (T) scores, new mean of 50* – so standardised scores show your client’s scores as they relate to the appropriate norms. • In this case, the norm for British psychology undergraduates.

• What is the range of scores your client is being compared to? • How has the client scored? Are they generally average or are there

extremes?

24

*Because scores have been ‘normalised’ by the process of conversion to T scores, the mean is now 50, and any score between 45 and 54 means you

are AVERAGE in that category – average for a British undergraduate, that is!

29/01/2019

13

PSY4046: IN ADDITION YOU WILL NEED TO MAKE PROFILE GRAPHS FOR YOUR CLIENT

25

Here you can see the red box indicated an average range, the green outer lines delineate low average or high average and anything outside this range is a low or high score in comparison to the normative sample. This is called a normative profile.

Instructions are in the Assessment folder on Moodle

THE FIVE FACTORS

• The five factor model emerged through extensive factor analysis and is the dominant model in psychology today in terms of reliable, universal personality traits. • It is based on the lexical approach – so words underlie every factor.

NB – There is no ‘right or wrong’ about personality characteristics. Your scores on the 5 domains indicate your PERSONAL PREFERENCE AND PERSONAL STYLE – each aspect has weaknesses and strengths associated with extreme scores.

29/01/2019

14

Big 5 traits - IPIP Personality Trait Facets

Extraversion Friendliness Gregariousness Assertiveness

Activity Level Excitement-Seeking Cheerfulness

Agreeableness Trust Morality Altruism

Cooperation Modesty Sympathy

Conscientiousness Self-Efficacy Orderliness Dutifulness

Achievement-Striving Self-Discipline Cautiousness

Neuroticism Anxiety Anger Depression

Self-Consciousness Immoderation Vulnerability

Openness to experience

Imagination Artistic Interests Emotionality

Adventurousness Intellect Liberalism 27

These are some of the trait

words associated with each of the Big

5 factors of personality. Use these words to describe the

characteristics of your client in

each of the areas – in

sentence form, of course.

FACETS OF THE UK NEO-PI

28

Neuroticism Anxiety Hostility Depression Self-Consciousness Impulsiveness Vulnerability to Stress

Extraversion Warmth Gregariousness Assertiveness Activity Excitement Seeking Positive Emotion

Openness to experience

Fantasy Aesthetics Feelings Actions Ideas Values

Agreeableness Trust Straightforwardness Altruism Compliance Modesty Tender-mindedness

Conscientiousness Competence Order Dutifulness Achievement Striving Self-Discipline Deliberation

6 traits or facets define each domain

The UK version of the NEO PI-R is published by Hogrefe(UK) Ltd

Trait words relate to characteristics displayed by scores – use these traits in your sentences

29/01/2019

15

TEST SESSION – TAKING THE EQ TEST

29

STANDARDISED INSTRUCTIONS FOR THE WECQ

• The WECQ comprises a list of statements numbered 1 to 70. • These statements relate to behaviours and attitudes likely to

arise in your everyday life; for example in the workplace. • Please read each statement and put a cross in the box under

the response that best suits the way you are now – for example, with the first statement, do you rely on your feelings when making decisions Hardly Ever, Seldom, Sometimes, Often or Almost Always?

• To help categorise responses, in addition to the questionnaire you are asked to please circle which of the following options applies to you:

• Male / Female 16-25 / 26-35 / 36-45 / 46-55 / 56-65

30

29/01/2019

16

Extra Instructions • There is no time limit to answering this questionnaire

but do not spend too long thinking about your responses.

• Please work in silence. • If you require help or do not understand one of the

statements, please raise your hand and someone will help you.

• Do you have any questions?

31You may start

THINKING AHEAD

• You need to SCORE THE WECQ using the EXCEL sheet available on MOODLE

• The procedure is the same as the hand-scoring shown for the IPIP

• You MUST bring your scores for the 5

DOMAINS of EIQ to the workshop in

Week 18 (after reading week).

• Detailed meanings relating to the 5 factors of personality will be available on MOODLE after this session.

• Next week – Interpreting Personality and writing to the client – Tests and Profiles 32