Foundations Of Testing Flashcards
What is a test
Standardised procedure
Meaningfully described outcomes eg - categories and scales used to make inferences
Norms and standards (reference point to make results meaningful and objective)
Made up of items (stimuli or questions)
Ability tests
- Intelligence tests - ability and global area
- aptitude tests - potential on specific task
- achievement tests - previous learning or accomplishment
- Creativity test - novel or original thinking and unique solutions
Personality and Behavioural tests
- Personality test - traits or behaviour, features of individuality
- interest inventory - preferences, determine job or social choices
- behavioural procedures - describe or count behaviour
- neuropsychological tests - cognitive, perceptual, sensory or motor control
Why use psychological test
- Classification
- diagnosis and treatment planning
- self knowledge
- Program evaluation
- Research
Responsibilities with psych tests
- Test developers and publishers (test construction and standards)
- test administrators (who should be using test?)
- test takers (impact on individual)
- Society (individual differences acknowledged by systems organise this complexity eg who is bipolar)
- others (government, compares, sponsers)
Factors affecting testing
- Test characteristics
- Standardisation
- psychometric properties
- Test taker characteristics
- familiarisation
- rapport
- test anxiety
- motivation
- reason for test
- test administrator characteristics
Early history -China
Han dynasty (206 BC - 200)
Developed test batteries for issues
Most for jobs in public office
Early history - Britain
Early 19th century
Two methods developed in opposition to inheritance of intelligence being linked to inheritance of social position
1: Experimental - scientific method to qualify psychological phenomena
2: observational- Darwins hypothesis applied to human behaviour by Galton
Early history - France
Late 19th century
Tests to categories people as worth determined by ability and merit (meritocracy)
Binet - first major intelligence test 1905- standardised and representative sample 1908- Mental age 1911- revision 1916- MA/AA x 100 = IQ
Early history - WW1
Needed efficient way to test lots of people = group test
Woodworth first self report personality test
Army alpha- reading ability
Army beta- illiterate adults
Early history - WW2
Group intelligence tests reaffirmed
Gave rise to clinical psychologists as tester - shift for psychotherapy
Post WW2
More Binet revisions
Wechsler first version with nonverbal scales, multiple facets of IQ, pattern and combination of abilities
Personality testing measures behaviour (traits) not ability
1: structured- psychometrically sound, factor analysis (MMPI)
2: projective - TAT and Rorschach
Bias and misuse
Eugenics - improvement of human species through selective parenthood - Galton
Goodard- translated stanford-binet to English = culturally bias to screen immigrants
Feebleminded should be sterilised- idea adopted by Nazi
Jensen-1960/70 - genetic basis of IQ - race based
Psych tests are powerful
Cultural differences
Biggest issue is language
Impact of culture - language, assumed knowledge, interpretation of performance/behaviour
Cultural fair tests have no distortion from cultural background
Should test be used? How is test score interpreted?
Objective - testing VS assessment
Testing -gauge ability from score and categorisation eg- measure behaviour
Assessment- answer referral question, solve problem, multiple tools of evaluation eg- ascertain diagnosis through interviews, observations, history ect.
Process testing VS Assessment
Testing- administer and score according to specific rules (manual)
Assessment- consider processes beyond score eg- select tests considering individual factors
Evaluator role testing VS assessment
Testing - no influence - standardised
Assessment - key to process, selection of tools and formulation of conclusions
Outcome testing VS assessment
Testing - final score
Assessment- answer referral question
Assessment process
1: obtain referral into
2: Conduct clinical interview
3: psychometric testing
4: collect collateral info if relevant
5: formulation of presenting issues
6: formal diagnosis if appropriate
7: treatment recommendations and plan
8: feedback to client
Assumptions
1: psychological traits and states exist
2: psychological traits and states can be quantified and measured
3: test behaviour predicts nontest behaviour
4: tests and measurements have strengths and weaknesses
5: various sources of error are part of measurement
6: testing can be fair and unbiased
7: testing/assessment benefits society
Scales of measurement
Nominal - named groups
Ordinal - named groups, in order
Interval - named groups, in order, equal intervals
Ratio - named groups, in order, equal intervals, absolute zero
Basic stats
Percentile stats- % scores fall below particular scare
Quartiles - equal fourths
Deciles- equal tenths
Mean
Standard deviation - average deviation around mean
Z score- mean = 0 SD = 1
T score - mean = 50 SD = 10
Norms
Give info about population based on observations of standardised sample
Z scores, mean, quartile are norms
Norm referenced test
Age related norms- NAPLAN and tracking babies weight
Cultural related norms- culture free intelligence tests
Criterion referenced tests
Mastery of specific skill
Driving test
Dance exam
Norm issues
Initial sample must be large enough to be representative
Updated regularly to reflect population
Criteran-referenced tests sometimes based on arbitrary cut off
