Psychology 309 Exam II: Validity

• Exam II for Psychology 309 is worth 150 points and will consist of 75 questions worth 2 points each.
• The questions will be multiple choice or true/false.
• The exam emphasizes application of principles.

Validity

• Validity refers to the evidence for inferences made about a test score.
• The three main types of validity evidence are content, construct, and criterion.
• Reliability is required for validity.

Face Validity

• Face validity isn't actually empirical validity.

Content Validity

• Content validity assesses whether a measure covers the construct fully, only the construct, and nothing but the construct.
• Construct underrepresentation and construct-irrelevant variance impact content validity.
• Construct underrepresentation occurs when the full construct is not measured.
• Construct-irrelevant variance happens when something outside of the construct is measured.

Content Validity Ratio

• The content validity ratio (CVR) is calculated using ratings from expert raters.
• Raters rate items as "essential", "important but not essential", and "not essential".
• The CVR formula is CRV=(Ne-N/2) / (n/2).

Criterion-Related Validity

• A criterion is an external measure used to judge the validity of a measure.
• Criterion-related validity is about how well a measure predicts an external criterion.

Criterion-Related Validity Subtypes

• Predictive validity assesses how well a measure predicts future performance such as how well the SAT predicts college performance.
• Concurrent validity assesses how well the measure matches the "daytime impairment questions"
• Postdictive validity compares test results to a previously achieved criterion.

Validity Coefficient

• The validity coefficient shows the relationship between the predictor variable and the criterion variable.
• The squared validity coefficient signifies the variance in the criterion variable predicted by the predictor variable.

Incremental Validity

• Incremental validity refers to the improvement in prediction beyond an existing measure.

Construct Validity

• Construct validity relates to how much the music of the song you pop to exist correlates.

Evidence in Construct Validity

• Evidence in construct validity comes in two types: convergent and divergent.
• The test should measure what it is supposed to measure or not measure what it doesn't report to measure.

Reliability vs Validity

• Reliability comes first before validity.

Logical vs Statistical Validity

• Face validity and content validity are logical and not statistical.

Construct Validity

• Construct validity has been referred to as "the mother of all validities" or "the big daddy."

Examiner & Test Taker

• The relationship between examiner and test taker is important because there are testing effects and expectancy effects.

Examiner Race & Intelligence Scores

• There is no strong evidence to suggest there is a relationship between test examiner race and intelligence scores.
• The effects increase, as the test becomes less structured.

Examiner race effects

• Examiner race effects are smaller on IQ tests than on other psychological tests because IQ tests are more standardized and structured.

Test Takers Language

• The standard for test takers who are fluent in two languages, is to give tests in a language they are fluent in. Fluency is important.

Expectancy Effects

• Expectancy effects are introduced by expectations of performance helped by test giver.
• Expectancy effects are associated with Rosenthal.

Expectancy Effects Study

• A review of studies showed that expectancy effects do exist in situations.

Testing Procedures

• Examiners might need to deviate from standardized testing procedures in certain situations.

Computer-Administered Tests Advantages vs Disadvantages

• Advantages: less error, test takers like it more, honesty, control, timing, standardization, and ability to give different questions at different times.
• Disadvantages: test is nonhuman, with possible interpretive gaps.

Subject Variables

• Subject variables like feeling watched, stress, expectancy, sickness, tiredness, sleepiness, and distraction can impact testing.

Major Problems in Behavioral Observation Studies

• Reactivity occurs when raters react to being checked.
• Expectancy occurs when raters expect subject behavior to happen.
• Drift occurs when raters stray from strict training and adopt idiosyncratic definitions of behavior.
• Contrast effect is caused by positioning effects.

Reactivity

• Reactivity is when performance increases when checked or observed.

Drift

• Drift leads to idiosyncratic definitions of behaviors and contrast effects, which refers to the tendency to evaluate the same behavior differently based on position context. Drift can be addressed with stricter training.

Experimenter Expectancies

• Experimenter expectancies refer to what the experimenter expects performance to be, subtly expressed via behaviors and body language.

Detecting Deception/Lies

• People are not good at detecting deception or lies.

Halo Effect

• The Halo Effect involves ascribing attributes to someone based on something other than the trait.

Good Interviewer

• A good interviewer builds alliance, encourages people to talk (social facilitation), creates a safe environment, and stays in control.

Interviews vs Tests

• Interviews are similar to tests, they asses gathering data validity and reliability.

Interpersonal Variables

• Interpersonal attraction attracts influence.

Interview Statements

• Should be avoided: confrontation, judgment, evaluation, probing (why), hostility, or false reassurance.

Interview Goals

• The main goal of interviewing is for the interviewee to give open and honest information while keeping the interaction flowing.

Transitional Phrases

• Examples include "uh huh," "ok," "yeah," and "wow," followed by open-ended questions to get more information and continue the theme.

Direct Questions

• Direct questions should be used in a structured interview when data cannot be obtained any other way, when there is little time, or when there is no cooperation.

Structured Clinical Interviews

• Structured clinical interviews are standardized but may yield less meaty responses and require cooperation.

Mental Status Examination

• A mental status examination evaluates psychosis and brain trauma.
• Covers Orientation.

General Standoutishness

• General standoutishness can bias judgment.

Interview Reliability

• Interview reliability is much higher (2X) for structured interviews.

Structured Interviews

• Criticism: the data is NOT broad nor flexible

Social Facilitation

• Social facilitation involves keeping the conversation rolling, building alliance, and getting data.

Interviews Error

• Judgment is the largest source of error in interviews.

Taylor's Research Traditions

• Three independent research traditions identified by Taylor, to study human intelligence are: cognition, psychometrics, and information processing.

Binet's Intelligence

• Binet believed intelligence was expressed through judgment, attention, and reasoning.
• Binet's two major concepts were judgement attention and reasoning. Mental age could be defined by a single number g

Age Differentiation

• Tasks that a 4-year-old can do are more tasks than a 2-year-old can do which indicates a positive manifold where intelligence tests tend to correlate with one another.

Binet's Task Completion

• Binet searched for tasks that could be completed by 66-75% of children in a particular age group.

Spearman's Concept

• Spearman introduced general intelligence (g).

Spearman's Statistical Method

• Spearman developed factor analysis to support his notion of g.

gf-gc Theory

• The two basic types of intelligence according to the gf-gc theory; fluid and crystalized.
• Fluid intelligence is like aptitude.
• Crystalized intelligence is learning.

IQ Calculation

• IQ = (mental age / actual age) x 100.

Deviation IQ

• The deviation IQ was used in the Stanford-Binet scale.

Basal vs Ceiling

• Basal is getting the floor for the intelligence test, usually getting three right in a row.
• Ceiling is the point where no more correct answers can be achieved.

Wechsler's Focus

• Wechsler focused on processing speed in addition to verbal spatial reasoning and working memory.

Binet Scale Criticisms

• Those before him had not.
• Adults aren't kids.
• There are non-intelligence effects

Wechsler Scales

• WAIS: ages 16-90
• WPPSI: ages 2.5-7.5
• WISC: ages 6-16

Point Scale

• Inclustion significance improvement is to compare between subscales and compare to mental age.

WAIS-IV Functions

• WAIS-IV measures verbal comprehension, perceptual reasoning, working memory.

Deviation scores

• Scaled scores: Mean is 10, standard deviation is 3
• Index scores: have a mean of 100 and a standard deviation of 15
• Standard scores: Mean of 100, standard deviation of 15

IQ Scores

• Calculated.
• Evalauated by the normed sample.

Index score

• Purpose from wach index.
• Verbal comprehension index.
• Working memory index.
• Perceptual reasoning index and processing speed index

Pattern Analysis

• Attempting to link patterns of scores with mental disorders
• Doesnt take into account variability.

Hold Subtest

• Requires to hold info.

Cerebral Dysfunction.

• Sub-tesis sensitive to it.
• Links will be.

Crystallized intelligence vs fluid

• Measure learning vs measure cyistalize.

Traditional Intelligence Tests

• Abilities TAIL.

Alternative Intelligence Test

• Bad psychometrics, not standardized.

Advantage Alternatives Intelligence

• Good nonverbal etc.

Infall development tests

• They dont get at it.

Surveillance vs screening

• Screeninf advise.
• Surveillance documented history and ridgit history.

Brazelton

• Baily brain damage and gives good 27.

Sensitivity

• Specificity negative, not normal.

Accepatable sensitivity

• Screening 75%, 70-80 range.

Learning disability

• It sucks ITS THE WOODCOCK JOHNSON.

Woddcock

• Achievement no achievement or ability.

Achievemnet

• No the score is really

Validity Types

• 3 types of validty content cirtereon and constant.
• Spearman g how it helps factpr analysis.
• Divergent and convergent validity provide.
• Examples.

Interview Qualities

• Accordint to class is good.

Effects that can be cause

• Examples what it matters

Questions Mitigate

• Discovered.

Binet

• Discrepancy sensitive.

4-6 Paragraphs

• Specitivity

• Exaples

• What is it

• What we do where is should be

• What is the main thing should be said.

• What is address?