Research Design, Statistics, Tests, and measurements Flashcards
Correlational
IV not manipulated
Naturalistic observation
Researcher does not intervene; measures behavior as it naturally occurs
Quasi-experimental
IV manipulated; subjects not randomly assigned to groups
True experiment
IV manipulated; subjects randomly assigned to groups
Stratified random sampling
Each subgroup of the population is randomly sampled in proportion to its size
Between subjects
2 groups. No 1:1 pairing
Matched groups
2 groups. Yes 1:1 pairing
Within groups
1 group that serves as their own control- experience multiple levels of the IV
Demand characteristics
Cues in research situation that suggest to the subject what is expected
Descriptive statistics
Organizing, describing, quantifying, and summarizing a collection of actual observations
Inferential statistics
Concerned w making an inference from the sample involved in the research to the population of interest, and providing an estimate of popular characteristics
Frequency distribution
A graphic representation of how often each value occurs.
Z scores
How many SD’s your score is above or below the mean. The mean distribution of z scores will always be 0 and the SD will always be 1
Normal distribution
About 68%of scores fall within 1 SD of the mean; about 96% of the scores fall within 2 SD of the mean. Mean, median, & mode are identical
T-scores
Has a mean of 50 & SD of 10. Ie a t-score of 60 is 1 SD above the mean.
Correlation coefficient
Range from -1 to +1. How well we can predict one variable from another variable. 0 indicates no relationship
Factor analysis
Seeing how many factors (variables that are measuring the same thing) there are by looking at a correlation table. “Armchair” factor analysis
Significance testing process
- Formulate alternative and null hypotheses
- Decide on alpha level or criterion of significance
- Perform sig test- number above alpha confirm null, numbers owner than alpha reject null
Type I error
Null is true & reject null. Prob same as ur alpha level
Type II error
Null is false. Accept null. More conservative. Beta
T-test
Compares means of two groups
Anova
More than 2 groups
Factorial design- each level of a given independent variable occurs with each level of the other independent variables
Interaction- when the effects of one IV are not consistent for all levels of the other IVs
Chi-square test
Use when individual observations are names or categories
Meta-analysis
Stat procedure that can be used to make conclusions on the basis of data from different studies
Norm- referenced testing
Assessing an individual’s performance in terms of how that individual performs in comparison to others.
Domain (criterion) referenced testing
Concerned w what the test taker knows about a specified content domain. I.e drivers test- not comparing u to ur peers
Reliability
Consistency of a test. Individual scores same when taking test multiple times. SEM is how much we expect an individuals score to vary
Test-retest
The same test is administered to the same group of people twice
Alternative method form
Examiners are given two different forms of a test that are taken at two different times
Split- half reliability
Only take 1 test, but test is divided Into equal halves & look @ correlation btw scores on each half. High + correlation (.8) indicates a high level of reliability
Content validity
Does the test measure various facets of what it’s supposed to measure? Adequate coverage?
Face validity
Does the test look like it measures what it’s supposed to measure?
Criterion/ concurrent
How well test can predict an individual’s performance on an established test of the same skill
Predictive
Does test predict future success in that area?
Construct validity
Convergent- is it correlated w things it should be correlated w?
Discriminant- is it not correlated w things it shouldn’t be correlated w?
Reliability & validity
A test must be reliable to be valid, but it does not need to be valid to be reliable
Ratio IQ
Aptitute. Binet- Simon. Mental age/ chronological age x 100
Deviation IQ
Stanford Binet. Indicates how well a person performed on an IQ tar relative to her/his same age peers
Wechsler tests
Have all items of a given type grouped into subtests. Not organized by age level. Verbal & performance scale
MMPI & CPI
Self- rating devices. Test taker is given a limited # of ways to respond to the statements (yes, no, uncertain)
Rorschach inkblot test, TAT, Rotter incomplete sentences
Relatively ambiguous stimuli are presented to the test taker; the test taker is asked to interpret the stimuli
Interest inventories
Used to assess interest in different lines of work. RIASEC system
