Psychological - Revision PSY3041 Catherine

There are Seven Assumptions of Psychological Testing

Psychological Traits & States exist & can be quantified & measured

Test-related behaviour predicts non-test-related behaviour

Tests & Measures have strengths & weaknesses

Error is part of the measurement process

Testing & Assessment can be fair and unbiased and Benefits society

• Psychological traits are stable & distinctive patterns of behaviour that characterise an individual & his/her reactions to the environment

• Psychological states are also distinguishable, though less enduring

• Assumption 1 applies with administration & test construction

Assumption 2 is rarely tested

Focus instead in on the development of test instruments

Attention is focused on emotional states, psychological traits of intelligence & personality

There are multiple theories of intelligence & of personality

Psychological Traits & States exist & can be quantified & measured

Test-related behaviour predicts non-test-related behaviour

Tests & Measures have strengths & weaknesses

Error is part of the measurement process

Testing & Assessment can be fair and unbiased and Benefits society

Psychological Traits are Stable, Distinctive Patterns of behaviour that characterise an individual & his/her reaction to their environment

Psychological States are also distinguishable, though less enduring

Assumption 1 is made by both Test Administrators & Test Developers

Assumption 2 is rarely tested - Michell (1997) complains about this stating that psychologists side-step the scientific task of demonstrating an attribute is quantititative, skipping straight to the instrumental task of creating procedures for assigning magnitude to these attributes.

Focus instead in on the development of test instruments

which examine emotion states, & the psychological traits of intelligence and personality

The 3 Main Theories of Intelligence are:

The Factor Analytic/Psychometric Approach

The Information Processing Approach

Gardener’s Multiple Intelligence Approach

studies Behaviour to understand intelligence (e.g. responses to numerical, spatial, verbal reasoning)

uses Complex statistical techniques to identify complex patterns of individual differences across tests

The approach evolved via the work of: Galton, Binet, Spearman

focus is on HOW information is processed rather than WHAT is processed

information can be processed sequentially or simultaneously (Luria) (step by step vs all at once)

Sternberg’s Triachic Theory of Intelligence proposes low-level abilities affect high-order cognition:

Knowledge Aquistion components

Performance componenets

Meta-Components

Gardener’s theory of Multiple intelligence proposes at least 8 distinct types

some of which function independently others are interdependent:

linguistic,

logico-maths,

spatial,

musical,

kinaestetic,

interpersonal,

intrapersonal,

naturalistic

also proposed spiritual & existential intelligence

No g - general intelligence

The 4 Main Theories of Personality are:

The Psychoanalytic Approach

The Social-Cognitive Approach

The Type Approach

The Trait Approach

This approach emphasises:

Childhood experience & the unconscious in motivating human actions

Uses Projective Methods or Techniques

Through disclosing their conscious/unconscious needs, desires, impulses,

The individual supplies the structure to the unstructured stimuli

The Social-Cognitive Theory is:

Grounded in both behaviourism & cognition

Reciprocal Determinism is core concept and is concerned with

the interaction between behaviour, person and environment

The Type Approach came from the Psychodynamic Approach

Personality is basically a classification system

e.g.

Myers-Briggs Type Inventory;

Type A & B;

Holland’s Self Directed Search;

Hippocrates (melancholic, phlegmatic, choleric & sanguine)

Individuals differ from one another in terms of:

distinguishable, stable, consistently expressed characteristics

e.g.

The Five Factor OCEAN Model -Measures: NEO-PI-R;

Cattel’s 16 Factor;

Eysenck Personality Inventory

A tests objective is to provide an indication of a test-taker’s behaviour;

i. e. ‘predictive validity’

There are 2 types of performance tests:

Maximum-Performance tests &

Typical-Performance Tests

A tests objective is to provide an indication of a test-taker’s behaviour;

e.g.

SAT’s, &

Big-Word / little-word test

\oth have ‘predictive validity’ which is a type of ‘criterion validity’

Maximum-Performance Tests intend to capture the test-taker’s Best Work

e.g.s:

Achievement Tests,

Aptitude Tests

Psychoeducational Tests

these are used to make predictions for future abilities (entry into Univeristy, Graduate school, Employment)

Typical-Performance Tests however,

focus on what a test-taker can actually do rather than their potential

e. g.

* personality tests (no correct/incorrect answers)

Psychological tests are imperfect measures of psychological attributes

Test Developers aim to maximise the strengths & minimise the weaknesses of psychological testing

Two key characteristics of tests that must be considered when constructing a test are:

reliability &

validity

Test Conceptualisation

Test Construction

Test Try-out

Item Analysis

Test Revision

Repeat steps 3, 4, 5 as required

Measurement rules: involves assigning no.s to the attributes of individuals according to nominal, ordinal, interval or ratio rules

there are 4 kinds of scales of measurement:

Likert

binary

forced choice

semantic difference scales

-When writing items start with an item pool & refine; using either selected response or constructed response items

-A Selected Response item could be

a matching choice,

multiple choice or

true/false test item

test-taker chooses the most appropriate option

Whereas with a Constructed Response e,.g.,

completion item,

short answer

essay question

the test-taker creates their response.

Good items are:

Valid,

Reliable, &

Discriminate between test-takers

e.g. items answered correctly by high scoring and incorrectly by low scoring test takers

NB: items which do the opposite are “Bad” items

Item Difficulty index

Item-Discrimination index

Item-Validity index

Item-Reliability index

NB: Item-Difficulty & Item-Discrimination are of particular interest to test-constructors

-The Item-difficulty index is calculated as

the proportion of test-takers who answered the item correctly (p)

p can range from 0 (no-one answered correctly) to 1 (everyone answered correctly)

Each items on a test has a corresponding p-value

(p1 is item-difficulty index for item 1 etc)

The index of item difficulty for a test is the average of all the p-values for the test items

The Optimal average item difficulty is 0.5 (items should range from 0.3 - 0.8)

An Item-Difficulty of 0 or 1 does not discriminate between test-takers

-Measures the extent an item differentiates correctly on the behaviour the test is designed to measure

"Good" item (high +scores max of 1): most high scorers answer it correctly & most low scorers don't

"Bad" item is the opposite (- d = big problem)

symbolised by d -Compares top 27% with bottom 27% of test-takers

U = No. of High Scorers

L = No. of Low Scorers

n = The total No. of Scorers

d = discrimination index score

d = U - L /(Divided by) n

Measurement error refers to the idea that variables other than those measured by the test influence test performance

To understand the concept of error, it is necessary to understand reliability & validity

Test-taker scores contain the True Score & Error (this is Classical Test Theory)

Test Construction Item or content sampling

Test Administration environment

Test-taker & examiner related variables

Test Scoring & Interpretation Hand scoring or subjective judgements

Test-retest Reliability:

Correlating pairs of scores on 2 different administrations of the same test

Internal Consistency Reliability:

split-half testing

Cronbach (nondichotomous items)

Kuder & Richardson (dichotomous items)

Inter-scorer Reliability

The degree of agreement between scorers

Content Validity:

Items are reflective of what they are testing requires little knowledge of psychological theory

Construct Validity:

how well do inferences drawn from a test relate to theories about psychological constructs?

Use Factor Analysis & calculation of Convergent & Divergent validity

Criterion-related Reliability:

The extent the test correlates with & thus predicts characteristics of a person assessed independently of the test Assessed via concurrent & predictive validity

Test Publishers strive to develop tests which are fair by ensuring they accord with key ethical principles and ethical codes

There are 3 Theoretical perspectives on Ethics:

Deontological (Duty Based)

Consequentialist (Outcome Based)

Personality Based (Virtue Ethics)

Focuses on the concepts of Duty & Moral Obligation

A commitment to act in accordance with absolute Ethical Standards

Exemplified by Kant's 'categorical imperative' i.e. humans are an 'end' in themselves & act so the principle of your action could become law

Focuses on the consequences or outcomes of actions rather than the underlying intention

The 'good' or 'right' action is that which promotes the greatest good for the greatest number of people i.e. utilitarianism

The moral worth of an action is dependent on its outcome;

cannot know if it is a 'right' act until all consequences have played out

AKA: the pleasure principle (Bentham) or the greatest happiness principle (John Stuart Mills)

-Exemplified by Aristotle's virtue ethics More concerned with a virtuous individual than 'right' & 'wrong' acts

Virtuous characteristic traits:

Wisdom

creativity

curiosity

fairness

forgiveness

appreciation of beauty & excellence

prudence

integrity

respectfulness

benevolence

These are difficult to teach & apply & uphold specific ethical principles

DEPHOGS

Dignity

Equitability

Prudence

Honesty

Openness

Goodwill

Suffering Prevention

General Principle: A - Respect for the Rights and Dignity of people & peoples B - Propriety C - Integrity

The APS Code of Ethics is most reflective of the Deontological and Consequentialist views, however, there are shades of Virtue ethics contained within

Psychologists regard people as intrinsically valuable & respect their rights, including the right to autonomy & justice. Psychologists engage in conduct which promotes equity & the protection of people's human rights, legal rights, and moral rights. They respect the dignity of all people and peoples. (APS, 2007, pg 11)

Psychologists ensure that they are competent to deliver the psychological service they provide.

They provide psychological services to benefit, & not to harm.

Psychologists seek to protect the interests of the people and peoples with whom they work.

The welfare of the clients and public, & the standing of the profession, take precedence over a psychologists's self-interest.

(APS, 2007, pg 18)

Psychologists recognise that their knowledge of the discipline of psychology. their professional standing, & the information they gather place them in a position of power & trust.

They exercise this power appropriately & honour this position of trust.

Psychologists act with probity & honesty in their conduct.

(APS, 2007, pg 26)

The 4 Ethical Traps are:

The Commonsense Objectivity Trap

The Values Trap

The Circumstantiality Trap

The Who Will Benefit? Trap

The belief that "commonsense, objective solutions" to dilemmas are easy since psychologists are ethical people using this approach every time they face a dilemma:

often the law is guiding our decisions (privacy & confidentiality);

objectivity is difficult to come by

we bring & impose our own belief system on situations, impacting objectivity

The Prominence of personal values over professional values & standards:

our religious, moral or personal values may conflict with professional code.

e.g. bias against divorce may negatively influence how we work with a client in a Family Violence situation

The Belief that there are no 'right' or 'wrong' answers to ethical dilemmas, that it is all down to circumstance.

Whereas, behaviour is often right or wrong according to professional ethical standards regardless of the circumstances under which it occurred.

Results from confusion about Who will Benefit from a specific ethical dilemma

Resolving an ethical dilemma often means taking sides among 2 or more conflicting interests

The resolution of the dilemma may not always be in the clients best interest

Simply that psychological testing & assessment benefits society.

Although testing and assessment may seem undesirable at times, do the benefits of testing outweigh the costs?

An historical example of the SOCIAL COSTS of testing

H. H. Goddard and the menace of the feeble minded 15

Popularised the Frenchman Albert Binet’s intelligence scale

Source of the descriptor “moron” – below average IQ “high grade defectives”

Wanted to prevent the entry of the feeble minded into the US

Application of this test led Goddard to believe that the majority of refugees arriving to America from Eastern and Southern Europe were “morons”/feeble- minded

An historical example of the Social Benefits of testing:

Ability and the emergence of meritocracy

Awarding jobs and rewarding people on the basis of ability is historically novel

e.g.,Samuel Pepys, 1st person in the 1660s,was a middle-class Briton who was among 27 the first officials to make his way up in government by virtue of the fact that he was smart.