• Shuffle
    Toggle On
    Toggle Off
  • Alphabetize
    Toggle On
    Toggle Off
  • Front First
    Toggle On
    Toggle Off
  • Both Sides
    Toggle On
    Toggle Off
  • Read
    Toggle On
    Toggle Off
Reading...
Front

How to study your flashcards.

Right/Left arrow keys: Navigate between flashcards.right arrow keyleft arrow key

Up/Down arrow keys: Flip the card between the front and back.down keyup key

H key: Show hint (3rd side).h key

A key: Read text to speech.a key

image

Play button

image

Play button

image

Progress

1/63

Click to flip

63 Cards in this Set

  • Front
  • Back
Variance
SS/(N-1)
Standard Deviation
square root of SS/(N-1)
Standard Error of the Mean
standard deviation of the population/square root of N

it is the standard deviation of the sampling distribution of the mean
Power
1-beta

where beta is the type II error (the odds of retaining the null hypothesis when it is really false; or the odds of saying there is no result/difference when there really is one)
single-sample chi-square
1 variable, nominal data
multiple-sample chi-square
2 or more variables, nominal data
Mann-Whitney U test
2 independent groups, ordinal data
Wilcoxon matched-pairs test
2 correlated groups, ordinal data
Kruskal-Wallis test
2 or more independent groups, ordinal data
t-test for single sample
sample vs. population for interval/ratio data
t-test for correlated samples
2 correlated groups, interval/ratio data
t-test for independent samples
2 indepedent groups, interval/ratio data
one-way ANOVA
1 IV, 2 or more independent groups, interval/ratio data
factorial ANOVA
2 or more IVs, interval/ratio data
repeated measures ANOVA
2 or more correlated groups, interval/ratio data
Variance
SS/(N-1)
Standard Deviation
square root of SS/(N-1)
Standard Error of the Mean
standard deviation of the population/square root of N

it is the standard deviation of the sampling distribution of the mean
Power
1-beta

where beta is the type II error (the odds of retaining the null hypothesis when it is really false; or the odds of saying there is no result/difference when there really is one)
single-sample chi-square
1 variable, nominal data
multiple-sample chi-square
2 or more variables, nominal data
Mann-Whitney U test
2 independent groups, ordinal data
Wilcoxon matched-pairs test
2 correlated groups, ordinal data
Kruskal-Wallis test
2 or more independent groups, ordinal data
t-test for single sample
sample vs. population for interval/ratio data
t-test for correlated samples
2 correlated groups, interval/ratio data
t-test for independent samples
2 indepedent groups, interval/ratio data
one-way ANOVA
1 IV, 2 or more independent groups, interval/ratio data
factorial ANOVA
2 or more IVs, interval/ratio data
repeated measures ANOVA
2 or more correlated groups, interval/ratio data
mixed ANOVA
independent and correlated groups, interval/ratio data
ANCOVA
removes extraneous variables, interval/ratio data
randomized block ANOVA
extraneous variables, interval/ratio data
trend analysis
quantitative IV, interval/ratio data
MANOVA
2 or more dependent variables, interval/ratio data
SST (sum of squares total)
SST = SSB + SSW

sum of squares between plus sum of squares within
MST (mean squares total)
MST = SST/df
F
F=MSB/MSW

F=(treatment+error)/error
Pearson Product Moment (r)
variable 1 = interval or ratio

variable 2 = interval or ratio
Spearman Rank-Order (rho)
variable 1 = rank-ordered

variable 2 = rank-ordered
Phi
variable 1 = true dichotomy

variable 2 = true dichotomy
Tetrachoric
variable 1 = artificial dichotomoy

variable 2 = artificial dichotomy
Contingency
variable 1 = nominal

variable 2 = nominal
Point Biserial
variable 1 = true dichotomoy

variable 2 = interval or ratio
Biserial
variable 1 = artificial dichotomy

variable 2 = interval or ratio
Eta
Used to assess non-linear relationships

variable 1 = interval or ratio

variable 2 = interval or ratio
LISREL
linear structural relations analysis

used when a causal model involves recursive (one-way) and non-recursive (two-way) paths

examines the relationship between observed variables and takes into account latent traits the variables are believed to measure and the effects of measurement error
Path analysis
translating a theory about the causal relationships into a path diagram
p (item difficulty)
total number of examinees passing the exam/total number of examinees

p=0.5 optimal unless true false test...then p=0.75 optimal
Item Characteristic Curve (ICC)
constructed for each item

information on the relationship between an examinee's level on the ability or trait and the probablity that they will respond to the item correctly
Item Response Theory (IRT)
the latent trait model

test scores are reported in terms of an examinee's level on the trait being measured rather than in terms of a total test score

makes it possible to equate scores from different sets of items and from different tests
Spearman-Brown prophecy formula
provides an estimate of what the reliability coefficient would have been had it been based on the full length of the test (or used to say what it would be if the tests were lengthened or shortened)
Cronbach's coefficient alpha
average reliability that would be obtained from all possible splits of the test
Kuder-Richardson Formula 20 (KR-20)
coefficient alpha (split half measure of internal consistency and reliability) used when items are scored dichotomously
kappa statistic
or coefficient concordance

inter-rater reliability
Standard Error of the Measurement
=standard deviation of the test scores multiplied by the square root of 1 - the reliability coefficient

used to calculate the confidence interval
multitrait-multimethod matrix
to test convergent and discriminant validity

correlations with different methods of the same trait and the same methods of different traits

methods for assessing construct validity
orthogonal
factors are uncorrelated
oblique
factors are correlated
base rate
(true positive + false negatives)/total number of people
positive hit rate
true positive/total positive
validity and reliability relationships
validity is less than or equal to the reliability
z
=(score-mean)/SD