What are the social functions of tests?
Entering new social spheres, access control, allocation (class/program/security level), identifying strengths/problems, learning/research/curiosity, bureaucracy (“ticking boxes”).
How can tests involve power dynamics?
They may control access to opportunities, act as a “rite of passage,” determine allocation, and sometimes may not be helpful to the person being tested.
Why is it important to consider consequences of testing?
Tests benefit some parties more than others; it’s crucial to ask whose interests are served.
Why are tests fallible?
They can be unfair, misused, or misinterpreted; remembering limitations and proper uses is essential.
What is a psychological test?
->obtains a sample of behavior
->measures psychological variables (thoughts, feelings, attitudes, behaviors, abilities)
->specifies a temporal frame
->generates data (qualitative or quantitative).
Difference between state-like and trait-like variables?
State-like: temporary, momentary (e.g., mood).
Trait-like: enduring, general tendencies (e.g., personality traits).
What is a scale in testing?
A test generating a number on a continuum, often norm-referenced, with possible predefined cut scores for interpretation.
What is an inventory?
- a collection of scales or tests
-usually sold as a set and designed to be administered together
-bundled to answer questions about an area of functioning (e.g. a person’s intelligence or personality)
What is a test battery?
A collection of tests assembled to answer broader questions, may be standard (same for all clients) or ad-hoc (customized).
How is testing different from assessment?
Testing: variable-centered, measures specific traits.
Assessment: referral-centered (EX. “what is this persons risk level?”), integrates multiple data sources (tests, interviews, observations, collateral info) to answer a specific question.
->involves expertise, interpretation, and judgment
What are some caveats of testing? Assumptions and limitations?
->since tests give us only a sampling of behavior, we must extrapolate from this
A:
- Behavior they are exhibiting is more of an extreme (Hawthorne effect)
- Assuming behavior is consistent across time and somewhat consistent across context
Limits:
- Behavior isn’t always constant across time/unpredictable/only predictable to a point
- Validity limits; reliability limits, item interpretation, normative sampling limits; interpretation limits; administration error
What is psychometrics?
defined as the science of psychological measurement.
->looks at psychometric soundness (whether it has reliability and validity)
Who are psychometricians?
Professionals who use, analyze, and interpret psychological test data.
Types of interviews in assessment?
Fully structured
Semi-structured
Unstructured
Describe a fully structured interview
- Every item’s wording, and response options, are pre-set (like a survey)
- Not commonly used in clinical psyc - could take too long to complete and get bored, doesn’t allow client to elaborate, not client centered
- Used for like phone survey’s
- Pros: Consistence across patients, standardized data, almost eliminates tester bias, easy to administer
Describe a semi-structured interview
- Item content and order are pre-determined, but leeway is given for interviewer to deviate (e.g., the SCID-5)
- Most often used in clinical work
- Can ask to clarify or open ended remarks
- Might be lying
- Looking for details and being picky to see if they meet the criteria for a certain diagnosis
Describe a unstructured interview
- No predetermined content, except for (maybe) a set of areas to cover or prompts (e.g., a clinical intake interview)
- Just trying to deepen the conversation
- Client lead
- Just offers prompts every once and a while
The quality of information obtained in an interview often depends on…
…the skills of the interviewer
(e.g., rapport, familiarity with psychopathology-know what questions to ask to discern between intrusive thoughts or obsessive thoughts between 2 different disorders)
Other tools of assessment?
-Case history data,
-behavioral observation,
-physiological/biometric measures,
-computers for administration/scoring,
-biofeedback,
-medical equipment (thermometers, blood pressure),
-specialized tools (e.g., penile plethysmograph).
Computers as tools
-Scoring may be done on-site (local processing) or at a central location (central processing).
Computer-assisted testing (CAPA) vs Computer-adaptive testing (CAT)?
CAPA:
->built-in scoring/interpretation = greater accessibility, don’t have to hand score it, don’t have to have person in the room with you
CAT:
->adjusts difficulty based on responses to estimate level accurately
Advantages and disadvantages of online testing?
- Greater access to potential test users.
- Scoring and interpretation tend to be quicker.
- Costs tend to be lower.
- Facilitates the testing of otherwise isolated populations and people with disabilities.
Disadvantages of Internet testing?
- Test security (e.g., copying test materials)
- Testing integrity (e.g., openness to cheating, timing differences)
- Emergency situations (e.g., client has a psychiatric emergency - call someone they know and trust, hospital security, go to emergency room, If online it’s harder )
- Behavioral observations (e.g., fidgeting, in their pajama’s, observe their environment/state of room etc.)
- Physical test materials (may need a trained facilitator)
Who are the key parties in testing?
Test developer, test user, test taker, society, organizations/agencies, academicians.
