Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
62 Cards in this Set
- Front
- Back
Internal validity
|
questions cause-effect relationships
|
|
External validity
|
not assuming results can be generalized to other contexts
|
|
Construct validity
|
shows if it measures an underlying psychological construct
|
|
Content validity
|
shows if it measures everything about a specific topic
|
|
Criterion validity
|
shows if it measures a set of abilities needed for something
|
|
reliability
|
consistency
|
|
test-retest reliability
|
shows if a test is reliable over time
|
|
parallel forms reliability
|
shows if different forms of a test are consistent
|
|
internal consistency reliability
|
shows if the items on a test/survey all measure one thing
|
|
interrater reliability
|
shows if raters are consistent
|
|
Test for test-retest reliability
|
pearson r (pretest and posttest)
|
|
test for parallel forms reliability
|
pearson r (form A and form B)
|
|
test for internal consistency reliability
|
chronbach's alpha on all item responses
|
|
test for interrater reliability
|
divide number of agreements by total number of observations
|
|
independent variable
|
the cause (what is manipulated, example: SSRI, CBT)
|
|
dependent variable
|
the outcome or effect (example: mood and depression)
|
|
nominal data
|
categorical, mutually exclusive categories (ex: gender, democrat/republican/indepdendent, happy/sad)
|
|
ordinal data
|
rank ordered, ranks show which is greater or less than another but not by how much (ex: 1st is better than 2nd place but we don't know by how much they won based on place) (ORDinal for ORDer)
|
|
interval data
|
regular number scale (ex: test scores)
|
|
reporting nominal data
|
using n and %
|
|
reporting ordinal data
|
using frequencies, mean rank, interval data, M and SD and may also want N
|
|
descriptive statistics
|
used to describe or summarize data (M and SD)
|
|
inferential statistics
|
statistical comparisons that determine if a significant difference or reltaionship exists between populations or samples
|
|
range
|
highest number minus lowest number
|
|
variance
|
square of standard deviation
|
|
standard deviation
|
the average deviation from the mean
|
|
normal distribution
|
mode, mean, and median are equal, bell shaped, symmetrical
|
|
confidence interval
|
estimates the true population mean (in a range) based on a sample mean
|
|
95% confidence interval
|
you can be 95% confident that he true population mean is in the interval
|
|
coefficient of determination
|
how much of the variance in one variable can be accounted for by the variabnce in another variable (square the r-value to find this)
|
|
research hypothesis
|
your educated guess about what hte data will show, states there is a statistical significance between groups
|
|
type 1 error
|
rejecting null hypothesis when it is actually true, 5% chance of this when alpha is .05
|
|
type 2 error
|
accepting null hypothesis when it is not true, decreases when sample size increases
|
|
t test
|
compares two means, results in a t value
|
|
one sample t test
|
compares sample mean to another value to determine if they are the same or different. The other value can be a population mean, probability of an event occuring, or other known value
|
|
independent samples t test
|
compares the means of 2 groups on one variable to determine if they are the same or different. The 2 groups must be independent of each other
|
|
paired samples t test
|
compares 2 means to determine if same or different. The 2 means can come from one group measured twice or 2 groups matched by at least one variable.
|
|
t value positive or negative?
|
absolute value!
|
|
chronbach's alpha
|
measures internal consistency
|
|
chronbach's alpha varies between what numbers
|
0 and 1 (but can get higher)
|
|
reliability scores that are acceptable
|
.7 or higher
|
|
effect size
|
d, a simple way to quantify difference between two groups, measure the magnitude of treatment
|
|
correlation coeffecient very strong relationship
|
.8 to 1
|
|
correlation coeffecient strong relationship
|
.6 to .8
|
|
correlation coeffecient moderate relationship
|
.4 to .6
|
|
correlation coeffecient weak relationship
|
.2 to .4
|
|
correlation coeffecient no relationship
|
.0 to .2
|
|
validity
|
the property of an assessment tool that indicates that the tool does what it says it does
|
|
critical value
|
the value you would expect the test statistic to yield if the null hypothesis is true
|
|
obtained value
|
the result of a specific statistical test
|
|
why ANOVA over t tests
|
ANOVA can compare more than 2 means, t tests only 2, less type 1 error
|
|
one-way ANOVA
|
1 IV, 1 DV
|
|
between-subjects ANOVA
|
2 or more IV, 1 DV
|
|
within-subjects ANOVA
|
no IV, 2 or more DV
|
|
mixed-model ANOVA
|
1 or more IV, 2 or more DV
|
|
Tukey
|
a post-hoc (after the fact) test
|
|
HSD
|
honestly...significant difference.
|
|
ANCOVA
|
an ANOVA with a covariate
|
|
linear regression
|
measures relationships (like pearson r but WITH prediction), uses one variable to explain or predict another
|
|
multiple regression
|
Shows how the combination of predictor variables is related to the dependent variable
|
|
chi square
|
used to test nominal (categorical) data
|
|
nonparametric test
|
doesn’t make assumptions about the population parameters
|