PS280 Lecture Notes - Inter-Rater Reliability, Criterion Validity, Xerostomia
Assessment, Classification, and Diagnosis
Basics of Clinical Assessment
● Goal: to find out what is wrong, what may have caused it, what can be helpful
● Multimethod approach: different tests/techniques, starting broad and trying to narrow
down understanding of problem
● Validity: ensure that test measures what it was designed to measure
○ Content validity: make sure test is getting at every aspect it should to cover what
it is meant to measure
○ Criterion validity: how test relates to other measures
■ Concurrent: does the measure relate to other things in the way it should
at the same point in time
■ Predictive: does measure predict some other variable in the future as it
should
○ Construct validity: whether a test measures what we intend it to measure
● Reliability: consistency of measurement
○ Interrater: extent to which 2 judges agree (psychologist A and B agree OCD)
○ Test-retest: consistency in measurement across time (jan 17, feb 17 anxiety test)
○ Alternate-form: 2 diff forms of test (jan 17 anxiety test form A, feb 17 anxiety test
form B)
○ Internal consistency: should be related if supposed to measure the same thing
(item 1 - dry mouth, item 2 - sweaty palms)
● Standardization: normative values → tests administered to diff groups of people, results
in normative values (eg for men/women, with anxiety disorder/without) → compare
results of client to normative values to see where they fit
Assessment Techniques
● Psychological assessment
○ Clinical interviewing: used in many areas
■ Presenting problem - is why the client came to therapy
■ Detailed history - understand person, context of their life
■ Interviews can be unstructured, semi structured, or structured (structured
questions are carefully worded and researched)
● Disadvantages of structure: less spontaneous/flexible, patient may
be less likely to volunteer important info
● Main advantage of structure: reliability → diff clinicians would come
to same diagnosis
■ SCID - structured clinical interview for DSM
● Used to diagnose DSM disorders
● Includes open ended qs, skip structure (eg if no symptoms move
on)
find more resources at oneclass.com
find more resources at oneclass.com
● Symptoms are rated
○ ? = inadequate info
○ 1 = absent or false
○ 2 = subthreshold
○ 3 = threshold or true
● Typically takes 2 hours
○ Psychological testing
■ Test is a standardized procedure designed to measure performance on a
task or personality
■ Personality testing
● Projective
○ Psychoanalysis
○ Use of ambiguous stimuli - presented to a client who must
describe what they see → theory is that people project their
own personalities onto the ambiguous stimuli and your
unconscious thoughts are revealed
○ Validity and reliability? → controversy
○ Used by many clinicians
○ Rorschach Inkblot Test
■ Hermann Rorschach 1921, 10 bilaterally
symmetrical inkblot cards, problems = can be
administered and interpreted differently
■ Exner’s Comprehensive System: focus is on form
rather than on content, response and inquiry
phases, scoring =
location/determinants/content/popular response?
○ Thematic Apperception Test (TAT)
■ Murray 1938 - tell a story about the picture → often
project their feelings
■ Scoring = quantitatively or qualitatively (qualitative
looks at themes across the stories)
■ Useful?
● Objective
○ Minnesota Multiphasic Personality Inventory (MMPI)
■ Hathaway and McKinely 1943
■ 567 qs in T/F format → psychological, physical
symptoms, items that distinguished clinical groups
(eg those with depression are less likely to say they
tease animals)
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
Goal: to find out what is wrong, what may have caused it, what can be helpful. Multimethod approach: different tests/techniques, starting broad and trying to narrow down understanding of problem. Validity: ensure that test measures what it was designed to measure. Content validity: make sure test is getting at every aspect it should to cover what it is meant to measure. Criterion validity: how test relates to other measures. Concurrent: does the measure relate to other things in the way it should at the same point in time. Predictive: does measure predict some other variable in the future as it should. Construct validity: whether a test measures what we intend it to measure. Interrater: extent to which 2 judges agree (psychologist a and b agree ocd) Test-retest: consistency in measurement across time (jan 17, feb 17 anxiety test) Alternate-form: 2 diff forms of test (jan 17 anxiety test form a, feb 17 anxiety test form b)
