Week 2 Flashcards
Representation model
Before you develop a testing instrument, a scale must have proven interval properties before it can be used
What must a scale look like when developing a test?
The scale must be ostensive (visualizable) like a yardstick or clocks
Basic problems in measurement theory
No single approach to measurement of any construct is universally accepted
-no conversion in measurement scales
Error of measurement
There will always be measurement error with humans because humans are testing other humans due to biases, limited samples
What did Mischel (1968) say
Proclaimed that behaviors and traits are situationally dependent and are not intrinsic, the way we behave is completely dependent on the environment ex: very active people who go on airplanes tend to have a difficult time, there behavior changes due to the environment
Why is what Mischel (1968) said a problem?
This is an issue for psychologists because we want to have a pattern of behavior for a diagnosis. This makes clinical psych no longer able to be used (obsolete)
State vs. Trait
Interviews-Anxiety
Loved one passing away-Depression
-if there is a trait, you would see that trait among multiple situations, so your correlation should be high, but its situations that cause this behavior not the trait alone
Accurate predictions of behavior comes from
Multiple items average over various situation and multiple occasions
The key to understanding traits is
Developing good measurement strategies
When cross-situational and multiple item measures are used correlations for generalized traits are
Raised to .60 or above, this explains that intrinsic traits do exist
Marshmallow test (Mischel)
This study was done to see if there were intrinsic traits, and explained that there was, the traits persist in the persons lives
Psychological assessment vs. psychological testing
Testing: just looking at test scores, benefit the needs of the client & Assessment: taking test scores and put it in within larger context, have a relationship between patient and clinician
Main focuses of testing
On the deficit
Main focuses on psychological assessment
To assess challenge areas and areas of potential
Matarrazo (1990)
Explains that clinicians do not understand assessment
Matarrazo explains two major areas of assessment that are used too infrequently
1) work records, school records, grades
2) Do not have a firm understanding of psychometric properties of instruments
2 areas that clinicians need to be more psychometrically aware are as follows
1) standard error of measurement
2) individuals given the same exam multiple time leads to various testing scores that are often not similar
Standard error of measurement
how repeated measures of a person on the same instrument tend to be distributed around his or her true score
Two theories on stability of traits
Classical trait perspective & radical contextual perspective
Classical trait perspective
Traits are biologically based temperaments that do not change over time
Radical contextual perspective
Emphasizes the role of environment and experience and the personality is fluid and is changeable
Personality traits and cognitive traits
Are relatively stable throughout the lifetime, increases more from age three and on
Cumulative continuity principle
Genetic factors contribute to the increase of personality traits over the lifespan (combination of the two perspectives)
Set point
Certain amount we can change but we have a set point that can be set to on levels, we all regress back to our set point
Unidimensional
Examining ordinal ranking on one attribute (spelling test) or having a child read aloud to hear that they can read
Just noticeable difference (Fechner)
Slight differences in our perception of sensations
Response centered approaches
Each respondent is asked to rank order his or her preference for a set of stimuli or to rank order a set of statements in terms of their proximity to his or her own personal beliefs
Fechners methodology
Is what we use today in modern psychological testing, indirect methods: need to measure objects to get at attributes and relies on the subject to discriminate
Scaling items in psychology
1) judgments vs. sentiments
2) absolute vs. comparative responses
Judgments
Where there is a correct response
Ex: how much is 2 plus 2?
Sentiments
Where’s the answer is based on preference
Ex: how much do you like soup?
Absolute vs. Comparative responses
Examines a particular stimulus whereas comparative responses relate to two or more stimuli (do you like peas or do you like peas more than corn?)
Standard deviation
Average of raw scores from the mean, raw scores are the deviations from mean
WAIS
100 is mean and standard deviation is 15
Dudychas study of personality
Examined the trait of punctuality in college students in various events from classes, meals, activities, correlation was significant, however it was seen that intrinsic traits do exist