Importance of Psychological Testing and Psychometrics Flashcards
Importance Of Psychological Testing
-our lives are shaped by psychological tests; we experience the different attributes we intend to measure on a daily basis
-society receives information and recommendations based on test results
-most people have been impacted by psychological tests
-important for practitioners working in the field
The Use of Tests
- decision-making
- (psychological) research
- self-understanding & personal development
- Decision-Making
- tests are used in a range of different fields and impact many personnel
-example: clinical/counselling
->prescribing the correct treatment should benefit patients whereas prescribing the incorrect treatment would be harmful for patients
- Research
- Psychological tests are fundamental to research; research findings need to be robust and the researchers need confidence in the tools they are using
-researchers will explore differences relating to people
-important to know if these are genuine differences (& not measurement error)
-if research uses poor tests = inaccurate and untrustworthy results; finding the best tests and measures is a critical part of research
- Self-Development
- knowledge of tests help us become inquisitive members of society
-tests help us learn more about themselves and reflect on items like feelings and behaviours and visiting a GP - tests help us learn more about others
Understanding Tests
- it is important that tests are of high quality; reliable and valid
-test providers have a responsibility to critically evaluate information about the meaning and the quality of the test
-without the knowledge of basic principles of psychological measurement, tests users risk: misinterpreting information from tests, making ill-informed decisions may harm patients, making ineffective decisions that cost money - important to be able to identify the strongest possible tools to: understand how these tests are applied and administrated, understand how they are scored, understand the results and practical meaning
Ethical Importance
-test providers have an ethical responsibility to individuals & groups they provide tests to or may be indirectly impacted
- it is important that tests are not unnecessary or harmful (they’re not biased)
- cannot claim ignorance over how to use a test properly
The BPS had 4 primary ethical principles
- respect: for the dignity of people regardless of background
- competence: able to provide specialist standards of practice
- responsibility: ensure others’ trust is not abused and high duty of care is offered
- integrity: honest, truthful, accurate, consistent in decisions, methods and outcomes
Observable Attributes
- observable attributes: there are many instruments used to measure the physical world (objects) -> e.g., height of a person = tape measure or width of a piece of wood = use of a ruler
- these objects are observable
-psychology may use methods to measure observable aspects e.g., facial expressions
Unobservable Attributes
- instruments can be used to measure aspects not directly observable
-psychological attributes are invisible
-no direct measurement e.g. depression, confidence - referred to as latent constructs
-to measure latent constructs, we observe human behaviour (or test scores) to infer the existence of unobservable psychological attributes
‘Unobservable’ Attributes
- infer an invisible attribute from an observable behaviour/test response
- observed behaviour/response should be theoretically linked to the invisible attribute; psychological measurement requires knowledge of theory explaining a construct
- we must assume the theoretical concept actually exists
-they will use operational definitions to understand constructs e.g. self esteem; indicators designed to help measure hypothetical constructs
What is a Psychological Test?
- “a psychological test is a systematic procedure for comparing the behaviour of two or more people (Cronbach, 1960)”
3 Important Components of a Psychological Test
- the test involves behavioural samples
- behavioural samples must be collected in a systematic way
- purpose of the test is to compare differences between people
-could also mean differences in the same person at two different timepoints
Strength Of This Definition
- applicable to a variety of psychological tests
-> standardised scales
-> laboratory experiments
-applicable to various types of information
-psychological tests must be capable of comparing scores
-> between different people or groups (inter-individual differences)
-> change in the same people over time/ different circumstances (intra-individual differences)
Additional Aspects
A psychological test is:
❖ Systematic procedure
❖ For obtaining samples of behaviour
❖ Relevant to cognitive, affective, or interpersonal functioning
❖ For scoring and evaluating those samples according to standards
➢ Samples should have cognitive, emotional, and interpersonal
❖ i.e., cannot use a psychological test on objects (e.g., a flower)
➢ Psychological tests are often described as standardised
❖ Ensure uniformity in administration, scoring, and interpretation of tests
o As best as possible, same conditions & scoring procedure for everyone
❖ Many tests have a standards for evaluating scores
o Data from a representative group (e.g., normal averages & variability)
o To compare against individuals’ scores
How Tests Differ
- tests can differ in various dimensions
1. content
-the concept being assessed (intelligence, motivation, anxiety, stress, personality)
-the number of questions
2. response required - open-ended (unstructured): people can say anything they like
- close-ended (structured): specific questions scored by predetermined options (1-5 scale)
3. Methods of Administration - some tests are given to individuals one at a time
-other tests can be given to a group all at once
4. Purpose: different tests are used & interpreted in different ways
Criterion-Referenced Test
-these tests have specific criteria to determine an individual test score
-does an individual exceed or fall below a cut-off score to establish the level of an attribute
-performance is independent of other people’s scores
Norm-Referenced Tests
- these tests are compared against a representative sample
- an individual will be above or below the average person (e.g., in an age group)
- in relation to others (i.e., not everyone can be above average)
How Test Types Differ
- Timing
-Tests can vary in the time expected of respondents
❖ Speeded tests (timed)
-Individuals are not expected to answer all questions
-Scores are allocated on how many questions are answered
❖ Power tests (not timed)
-Individuals are expected to answer all questions - Meaning of “indicators”
❖ Reflective (effect) indicators:
-Test scores are caused by the individual’s psychological attribute
-> i.e., intelligence scores caused by actual intelligence level
❖ Formative (causal) indicators:
-Test scores define the individual rather than be caused by an attribute
-> e.g., scores for gender/ socio-economic status explained the individual - Meaning of indicators
-reflective (effect) indicators: test scores are caused by the individual’s psychological attribute
-formative (causal) indicators: test scores define the individual rather than be caused by an attribute
What is Psychometrics?
- Psychological tests are designed to measure psychological attributes of people
- Psychometrics is the science of evaluating the (theoretical) attributes of tests
Three Psychometric Properties of Particular Interest
- Type of information generated by the use of psychological tests
- Reliability of data from psychological tests
- Issues concerning validity of data obtained from psychological tests
Why do we need psychometrics?
-we can never be sure that a measure is perfect
-need to clarify the degree of reliability and validity
- need to be confident that these differences are genuine and meaningful
Challenges to Measurement
-complexity of psychological concepts
-participant reactivity
-observer expectancy/bias
-composite scores: tests typically involve a range of items that are averaged together; important to consider if specific questions are problematic
-score/scale sensitivity: some scales offer more sensitivity than others
-need to ensure a measure is capable of identifying meaningful differences in an attribute
-lack of knowledge of psychometrics
