1/11
includes big things we delved in during class, as well as textbook things such as assessment professionals and everyday psychometrics
Name | Mastery | Learn | Test | Matching | Spaced | Call with Kai |
|---|
No analytics yet
Send a link to your students to track their progress
Dr. Delphine Courvoisier
conducts quality-of-life research on people with Rheumatoid arthritis
her focus is on health related QoL, degree of functional disability, and disease activity prevention. Achieves this with a 36 Q short form health survey (SF36)
also uses Health assessment questionnaire (HAQ) and disease activity score(DAS)
all data is employed to evaluate effectiveness of treatment regiments and adjust when necessary
Looks at utility of methods used, with example being responding to series of telephone calls
important to see how compliance vs non compliance affects results
compliant people (who chose to answer) might be more conscientious, so data collected would be causally related to conscientiousness
this is why testing (and rejecting) hypothesis that only patients high on conscientiousness answer the phone
conducted a study with a personality test and EMA in a series of phone interviews, found that compliance is not attributed to personality factors
Dr. Wanda Carlos on Flecha Esmerelda
wanted to change status quo of giving anyone with a valid drivers license and no criminal record the chance to do a 3 month probationary period. Checked with supervisors over course and if good performance, they got job.
Designed the FERT:
employee screening test
utility of test is analyzing whether or not it is better than just the 3 month probation period
cost analysis
false negative: not enough employees b/c FERT cut off too harsh
false positive: bad employees, less proficient drivers on rode
Face validity:
weed out bad people, better advertisement??
buy in: less likely to sue for discrimination, applicants take it more seriously
Content validity:
sample from all aspects of driving, maybe sample workers
time, conditions, situations to be considered
goal: establish criterion validity
predictive validity:
hire drivers with method
but also give them FERT
supervisor ratings for 1st 3 months
compare how well FERT predicts supervisor ratings
restriction: nobody who did not work out during 3 month period will be assessed
Selection types:
Top down:
best performer gets a job, go down the line until all slots filled
risk of disparate impact: unintentionally harming protected groups
if 20th and 21st had basically same score, why pick one over other
Half in and half out:
the status quo, 30 in and 30 out (so 30 bad drivers on road)
No false negatives:
positive: do not fail to hire anyone who would have worked
negative: only difference of 3 from status quo (27 bad drivers doing 3 month period compared to 30)
No false negatives:
positive: No bad drivers doing probation period
negative: only sure about 6 drivers, not worth it, also 24 competent drivers not able to do the probation period
Highest hit rate:
best method (31 people accepted, only 5 bad drivers on road, only denied 4 good drivers, and accurately not accept 20 bad drivers)
Use of Body Cameras
ariel et al found that using body cams resulted in less violence and less sueing for violence. but it is hard to figure out the why
Asexuality test
Utility Big picture 1
Assumption 7 from Ch4: testing and assessment benefit society
utility: how does this help people make decisions,
often quantitative
Utility Big picture 2
the ways of considering utility:
cost analysis
time savings
cost-benefit ratio - some costs are worth it
clinical utility:
helps us to make correct diagnoses
selection: can we select right people? (in and out)
circumstances to have sorting tasks in tests?
overcomes social desirability
having kids put peers in classes into different “buckets” think tier lists
limiting amount you can put in each bucket counters putting everyone into same bucket
show how a floor affect can ocur in prisoners
measuring integrity in prisoners would not provide any utility since they would all probably not be very integral
ceiling effect example
extraversion in grovers, everyone would score high (Context is important!)
item difficulty index
aiming for max discrimination, (so with guessing should yield .5)
item reliability index
1 of many items on a test,
looks at how each item specifically helps or hurts
Cultural Group tool Adaptation
Looked at how things such as health can be misinterpreted between Western and Non-Western groups.
Australians view health holistically, encompassing mental, physical, and the spiritual
ill health often perceived as a misalignment of these domains
Americans would perceive Australians as having mild health complaints
Australians are very reserved and do minimal eye contact, something that Westerners could view as a sign of psychopathology
However, Australian patients will “come alive” if a family member is present during the interview with a Caucasian clinician
We need a yarning approach with Australian members
interview strategy characterized by creation of an atmosphere that promotes the interviewee to conversationally tell their own story in their own way. Stark contrast to this is interrogation, or one question after another
Clinicians created the Here and Now Aboriginal (Australian) Assessment - HANAA
involves traditional storytelling style involving family and social yarning
can figure out if patient has anhedonia (inability to be happy) by asking “have you lost interest in things you used to like doing”
this engages in culturally appropriate activities with story telling by probing with things such as fishing
can also probe about what a “weak spirit” is, and how frequently they feel that way