Assessment & Research Methods Flashcards
(33 cards)
Test-retest reliability
Degree to which a test of a stable trait generates same results when it’s given more than once to the same person
Alternate-form reliability
Two versions of test are designed
Scores from each are correlated to each other
Internal consistency
Split-half reliability
Coefficient alpha
Degree of reliability within a test
Measure of internal consistency (compared responses on odd numbers to even numbers)
Average of intercorrelations of all items in a test
Face validity
Items in test resemble characteristics associated with concept being tested
Content validity
Test includes a representative sample of all behavs thought to be related to the construct it’s designed to measure
(Reflects overall construct)
Criterion validity
Arises bcuz some qualities are easier to recognize than to define completely
(Trait is assessed properly)
Construct validity
Refers to validity of a test within a specific theoretical framework
(More useful for abstract traits, like self-esteem)
Brain imaging techniques
Computed tomography (CT)
Narrow band of xrays projected thru head onto crystals; Creates 2D image or cross-section of brain
Magnetic resonance imaging (MRI)
Magnetic field produced around head, causing electrons to move… When it stops, the electrons go back to OG place and emit radio waves; Shows metabolic changes in brain
Positron emission tomography (PET)
Radiation is generated by injected or inhaled radioisotopes; Shows distribution of neurotransmitters
Bender Visual-Motor Gestalt Test
Oldest neuropsychological screening test
Copy designs onto another card then draw from memory
Can produce false negatives (ppl with impairment can complete the test with a few errors only)
Clinical interviews
Unstructured: Open-ended; allows interviewer to pursue specific line of questioning or follow patient’s lead
- Creates trust between clinician and patient
Structured: Has specific wording of questions
- Increases reliability but can reduce trust
Semi-structured: Optional guided questions, depending on what of the patient’s functions needs to be covered
- Typically uses mental status examination
- Designed to look for specific problems
Rating scales: Helps organize info and encourage reliability and objectivity
Sir Francis Galton
First scientific study of intellectual functioning
Tested if intelligence has hereditary aspect
Could be measured by physiological cues (sensory intelligence)
Alfred Binet
First widely accepted and successful test of intelligence
Predicted academic performance (IQ)
David Wechsler
Developed most popular IQ test (Wechsler Adult Intelligence Scale)
For adults
What is the correlation between IQ scores and academic performance?
0.5-0.7
Other than IQ, what else impacts academic performance?
Self-discipline
Education (school attendance)
Projective tests
Personality assessment
Person presented w/ ambiguous stimulus will project their unconscious motives onto it
Rorschach Inkblot Test
- Exner system developed yo increase reliability and validity by standardizing scoring of responses
Thematic Apperception Test (TAT)
- Ppl construct stories of the images on the cards
Personality inventories:
Minnesota Multiphasic Personality Inventory (MMPI)
Millon Clinical Multiaxial Inventory (MCMI)
Personality Assessment Inventory (API)
Minnesota Multiphasic Personality Inventory (MMPI)
- Assesses multiple aspects of personality
- Used contrasted-groups method to get validity (items only chosen if ppl known to have characteristic the scale is intended to measure responded differently to ppl who don’t)
Millon Clinical Multiaxial Inventory (MCMI)
- Helps clinicians make diagnostic judgments about personality disorders
Personality Assessment Inventory (API)
- Provides info relevant for clinical diagnosis, treatment planning, and screening for psychopathology (and severity)
- Uses 4-poing Likert scale
- Ppl tend to say yes (response sets) bcuz of test taking attitude
- Ppl tend to answer in a way the tester would desire (demand characteristics)
- Ppl answer in a way to look good (social desirability)
What is person by situation interaction?
By Walter Mischel
Predicting person’s behav requires knowledge of both the person’s typical behav patterns and the characteristics of the setting
Behavioural and cognitive assessments:
Observational
In vivo
Analogue observational setting
Observational techniques
- Uses rating scales to assess various behavs
- Looks at antecedents (what happens before behav) and consequences (what happens after behav)
In vivo observation: Behav in everyday enviro
- Can be impractical bcuz of time constraints and unpredictability of modern family life
Analogue observational setting: Artificial setting structured to elicit specific classes of behavs in ppl
Limitations:
- Reactivity - Ppl may change how they behav in front of observer
- Observer drift - Observer may deteriorate in accuracy as they get tired or mistakenly shift criteria over time
Cognitive-based assessments
Uses questionnaires to assess thoughts
Self-monitoring: Patients note their own behavs in real time
How do you deal with the noise in experimental methods?
Pretest: Participants assessed on many measures prior to treatment
Posttest: Participants assessed on dependent variables (rating scales)
Placebo effect: Ppl try to report better than experienced
- Placebo or double-blind procedure can help
Internal vs external validity
Internal validity: Degree to which changes in dependent variables are a result of manipulation of independent variable
External validity: Generalizability of findings in other settings or individuals
Quasi-experimental study
Confound
Cause-effect interferences
Participants in experimental group not randomly assigned
Confound: Occurs when 2+ variables influence at same time, making it hard to find causal role of any variable
Limitation:
- Cause-effect inferences: Can’t match participants on all factors except for subject of research
Correlational method
What kind of study is best for this?
Measures degree of relationship between 2 variables
Longitudinal studies
- Helps reduce ambiguity
- Longterm examination of early factors before onset of disorder
- Uses high-risk method (usually only takes sample of ppl more likely to exp desired outcome)