Download test adminitration notes and more Summaries Psychology in PDF only on Docsity!
TEST ADMINISTRATION
PSYC 3200 Winter 2026
WARM UP
- Grab a partner and take turns talking about a time when you took a test that you feel you did not perform as well as you could have but not due to not knowing the content?
FIRST…, VARIETIES OF TEST CONDITIONS
- People complete measures under a large range of conditions
- At home, in front of the TV
- Sitting in front of a computer also doing who knows what
- At a pre-specified location
- Laboratory
- Interview style
- One on one
- Group administration
WHAT DO YOU THINK HAS THE STRONGEST EFFECT ON TEST 1:1 PERFORMANCE?
- A. The relationship between the test administrator and test taker?
- B. The demographics of the test administrator??
- C. The demographics of the test taker?
- D. The test takers expectation about the test?
THE RELATIONSHIP BETWEEN EXAMINER AND TEST TAKER
- Examiners’ behavior as well as their relationship to the test taker can affect test scores
- Early study by Feldman and Sullivan (1960) contrasted two conditions, enhanced rapport vs. neutral rapport.
- For some age groups (not youngest) score discrepancy was dramatic
- Other studies have found that supportive comments by the administrator also had a positive impact on results
- Even subtle cues — ones the administrator may be unaware of — can be important
RACE OF THE TESTER (EXAMINER)
- Research is unclear on the extent to which the test taker and subject being of the same or different race affects outcomes
- Testing is supposed to be done according to standardized administration procedures
- Perhaps the simple presence of a test taker of a different race is enough to cause an effect
- Was earlier research showing an absence of impact affected by bias?
- There may be some impact of allowing the test taker’s latitude in how items are presented
STEREOTYPE THREAT: THE PROBLEM
- With equal preparation for college entrance exams African Americans (on average) underperform relative to White students.
- By puberty females start to underperform relative to males in mathematics achievement tests (and confidence about same)
- This all adds up to uneven access to resources and opportunities.
- E.g., advanced placement programs.
- Under-representation in certain fields, thus careers.
WHY DOES THIS OCCUR?
- No definitive answers but some working hypotheses
- Stereotype threat depletes cognitive resources such as working memory
- Self-handicapping can lead to reduced effort and performance
- Causes physiological arousal that disrupts performance (perhaps by affecting cognitive resources)
EXPECTANCY EFFECTS
- Rosenthal Effects – observations can be impacted by what experiment(er) expects to find.
- Princeton Engineering Anomalies Research – were results due in part to Rosenthal Effects?
- Imagine you are working as a psychometrist. An adolescent comes in for an assessment, beginning with administering the appropriate Wechsler Scale.
- You are told the young person is the son/daughter of accomplished parents (maybe neurosurgeons)
- You are told the young person is the son/daughter of poor parents who have unstable income due to a lack of education.
SUCH A PHENOMENON COULD MANIFEST IN OTHER TYPES OF MEASURES
- Imagine you are trying to obtain data about a sensitive topic, such as driving while intoxicated. Imagine some different questions.
- Have you ever driven while intoxicated. For instance, imbibing the equivalent of three or more drinks over a two-hour period?
- Though it is not appropriate, it is common for people to drive after having imbibed three or more drinks over a two-hour period. Have you ever done this?
MODE OF ADMINISTRATION & SUBJECT VARIABLES
- Mode of Administration
- There has been considerable research comparing traditional paper and pencil surveys versus computer administered.
- Subject Variables
- Transient conditions (headache, mild illness)
- Non-transient conditions such as test anxiety