Psych 303 Exam 1
Terms
undefined, object
copy deck
- operational definition
- translate something into concrete definition. i.e. aggression= number of times you hit bobo
- participant (subject, individual differences variable)
- intelligence, age, gender, aggressiveness
- Nominal Measurement Scale
- No numerical/ quanitiative properties. Assigning names and categories. Described normally as relative frequency ( how many blondes do this?)
- ordinal measurement scale
- you can rank things (4 stars for movies) different between 3 and 4 star may differ between 2 and 3 star
- ratio measurement scale
- quanitative wut akk numerical properties (has zero reference point) weight/reaction time
- interval measurement scale
- difference between numbers are equal. quantitative but not meaningful zero reference. ex:thermometer, calendar
- counterbalancing
- all possible orders of presentation of conditions are presented complete comunterbalancing is not always practical. 3!
- Latin Square
- Limited set of orders but each condition appears at each ordinal position. Each condition proceeds and follows each condition one time.
- cross sectional methods
- measured different groups at one point in time
- longitudinal method
- same group observed at different times as they get older
- cohort affecr
- people born in different times vary on more than age. family size, cultural values, child rearing practices
- repeated measures design advatages
- fewer participants extreme sensitivity to statistical differences = more likely to detect effect ( act as their own controls)
- repeated measures design-disadvantages
- sometimes you can do within subjects design (males and females) learn something in one condition that makes it impossible to do in another. too difficult for one person to do all conditions
- repeated measures design
- same participants in all groups. (shallow vs. deep processing. have same participants do both)
- use of time intervals
- can be used as a rest period to counteract fatigue, allow drug to wear or or emotional state return to normal, attending to unrelated task in bwetween conditions can reduce contrast effects
- naturalistic observation
- field work or field observations. used to understand how people live work and experience the setting. no experiments or manipulation. post hoc hypothesis
- negative case analysis
- looking for one case to disprove hypothesis (may need to use with naturalistic observation-easy to have bias)
- systematic observation
- Systematic observation refers to the careful observation of one or more specific behaviors in a particular setting.
- reactivity
- possibility that the presence of an observer changes people’s behavior. Reactivity is a potential problem in all studies (both observational and experimental). Fails to provide a measurement of the behavior under natural circumstances (really bad in naturalistic studies since you really want to maximize this). one way mirrors, hidden cameras
- inter rater reliability
- Correlation between the observations of raters (should be 80 percent)
- case study
- A case study provides a description of an individual.Case studies are done when an individual possesses a particularly rare, unusual, or noteworthy condition.
- psychobiography
- A psychobiography is a type of case study in a researcher applies psychological theory to explain the life of an individual.
- advantages/disadvantages of case studies
- -ich source for developing ideas for later experimental studies Opportunity for clinical innovation (SMT example) Method to study rare (and perhaps non reproducible) events Brain damage or someone with rare unusual abilities. Possible way to challenge theoretical assumptions DIS Can’t draw cause and effect conclusions. Can’t control extraneous variables. Susan did SMT really cause her to get better? Placebo effects, no control group. Unusual abilities like “Sâ€: genetics, or experiences, training, etc., or what was about his experiences? External validity: can you generalize findings to other people?
- archival research
- Archival research involves using previously compiled information to answer research questions. Researcher is not collecting the data. Three basic types of archival research data: 1. Statistical records 2. Survey archives 3. Written and mass communication records
- test-retest reliability
- Test the same person on two or more occasions. Scores should be highly correlated. (but ppl could get better with practice)
- cronbachs alpha
- researcher calculates the correlation of each item with every other item. alpha is the average of all the correlation coefficients. (items can be thrown out- inter raters)
- how do you increase inter rater reliability
- Maximizing Interater reliability Provide clear definitions of behaviors and events to be recorded (good coding system). Provide training for observers. Provide feedback on discrepancies. Raters talk about events that they coded differently. Usually during training phase or pilot study.
- parallel/alternate forms
- Two alternative forms of measure/test given to same person successively (on same day). Correlation is measure of reliability. No memory recall problem. No time for actual change to take place. (hard to make two tests the same)
- split half reliability
- A measure of the reliability of a test based on the correlation between scores on two arbitrarily formed halves of the test, often the odd-numbered and even-numbered test items.
- construct validity
- how appropriate are the operational definitions? does it test what it should? 3rd variables?
- concurrent validity
- whether two or more groups of people differ on the measure in expected ways
- predictive validity
- the extent to which the measure allows you to predict behaviors it should predict (in the future)
- criterion orientated validity
- examines the relationship between scores on the measure and some criterion. does measure predict real world outcomes? data on actual behavior- does it correspond to scores on measure
- convergent validity
- the extent to which scores on the measure in question are related to scores on other measure of the same or similar constructs (one measure of aggression correlates to another)
- discriminant/divergent validity
- does the measure differentiate the behavior of interest from other behaviors
- face validity
- do the measure appear to measure what it should-subjective
- internal validity
- what you did in the study casued the results of the study- replication with different populations and settings. are confounding variables controlled. x causes y?
- external validity
- ability to generalize the results to other populations. replicated with other populations?
- ecological validity
- must approximate the situation being questioned to the real situation
- field experiments
- not observational studies. experiments done in actual environment (classroom) - better ecological validity
- basic vs applied research
- curiosity vs. how to improve something
- between subject
- only used once
- within subjects
- used more than once
- matched pairs design
- ensures groups are equivalent on the matching variable prior to IV
- independent groups design
- between subjects design. only assigned to one group