Stat Final

Front 1

ANOVA

Back 1

Hypothesis testing procedure for studies with three or more groups

Front 2

One-way ANOVA

Back 2

One categorical independent variable
One continuous dependent variable

Front 3

Within-groups estimate of the population variance

Back 3

Average each sample’s estimate of the population variance into one single pooled estimate
- combined variation of the groups studied

Front 4

Between-groups estimate of the population variance

Back 4

Estimate of the variance of a population of individuals based on the variation among the means of the groups being studied

Front 5

F ratio

Back 5

Between-groups population variance estimate to the within-groups population variance estimate
Between-groups:Within-groups

Front 6

how to compute within-groups estimate of the population variance (S^2 within or MS within)

Back 6

find S^2 within or MS within which = S^2 one + S^2 two + S^2 three +… S^2 last / (divided by) N groups

Front 7

How to Compute between-groups estimate of the population variance

Back 7

find S^2M which = Σ (M – GM)2 / dfbetween =
then to find S^2 between multiply S^2M (n) =

Front 8

dfbetween =

Back 8

Ngroups -1

Front 9

S^2M

Back 9

estimated variance of the distribution of means

Front 10

GM =

Back 10

Grand Mean
the overall mean of your scores, or mean of your means
GM = (ΣM)/Ngroups

Front 11

F Ratio

Back 11

F = S^2 between / S^within

Front 12

Between-groups degrees of freedom
_______________________________________
Within-groups degrees of freedom

Back 12

df between(numerator)
_________________________
df (denominator)

Front 13

Hypothesis testing w/ analysis of variance

Back 13

1) restate the Null, and research hypothesis
2) Determine the characteristics of the comparison distribution (an F distribution for analysis of variance)
3)Determine the cut off at which the null can be rejected
4) Determine your sample's score on the comparison distribution (determine the F ratio for the sample)
5) decided weather to reject the null (if the F ratio is larger than the cutoff we can reject the null

Front 14

Planned contrasts

Back 14

Particular means to be compared are decided in advance

Front 15

Post-hoc comparison

Back 15

Exploratory analysis after an ANOVA
Most popular: Scheffé Test

Front 16

Factorial ANOVA

Back 16

Way of organizing a study in which the effects of two or more variables are studied at once by making groupings of every combination of the variables
ex: a 3 x 2 factorial design, relationship status: married, dating, single (x) gender: male, female

Front 17

Main effect

Back 17

Difference between groups in one variable in a factorial design in analysis of variance

Front 18

Interaction effects

Back 18

A combination of variables have an effects that could not be predicted from the effects of the variables individually

Front 19

Correlation

Back 19

Association between scores on two variables
Equal-interval numeric variables

Front 20

Scatterplot

Back 20

One variable is represented by the x-axis (horizontal)
One variable is represented by the y-axis (vertical)
Allows us to see the relationship between the 2 variables of interest

Front 21

Linear correlation

Back 21

Scatter diagram roughly approximates a straight line

Front 22

Curvilinear correlation

Back 22

Scatter diagram does not approximate a straight line but instead follows a systematic pattern (or a complex curve)

Front 23

Positive Correlation

Back 23

variables go in the same direction; highs go with highs and lows go with lows. Increase or decrease together (go in same direction)

Front 24

Negative Correlation

Back 24

low scores on one axis are paired with high scores on the other, middle scores with middle scores, and high scores paired with low scores.One increases other decreases (go in opposite directions)

Front 25

Correlation Coefficient

Back 25

Range from +1 to -1

Front 26

Cross-product of Z scores

Back 26

Multiplying a person’s Z score on one variable by the person’s Z score on another variable

Front 27

Correlation Coefficient (r)

Back 27

r = sum of ZxZy / N
measure of degree of linear correlation between two variables raning from -1 (perfect negitive correlation) through 0 (no correlation) to +1 (perfect positive correlation).

Front 28

How to computing statistical significance of a correlation coefficient

Back 28

t = ________r_______
√(1 −𝑟2)/(𝑁−2)

Front 29

Correlation Matrix

Back 29

Report of correlation coefficients among several variables

Front 30

Correlation does NOT imply causation

Back 30

there may be a 3rd variable, which causes both X and Y
however, If X preceded Y, then you can be confident that Y cannot cause X

Front 31

Correlation =

Back 31

Relationship

Front 32

Regression =

Back 32

Prediction

Front 33

How we make predictions for an individual on one variable

Back 33

based on knowledge of their score on a second variable
If we are using X to predict Y:
X = predictor variable
Y = criterion variable

Front 34

Prediction using Raw-Scores

Back 34

To predict a raw score for the criterion variable (Y):
Multiply the predictor variable (X) by the raw-score regression coefficient (b) and then add a number called the regression constant (a)
Y = Xb + a

Front 35

Prediction using Z scores

Back 35

To predict a Z score for the criterion variable (Y):
Multiply the Z score for the predictor variable (X) by the standardized regression coefficient (β)
Y = Zxβ

Front 36

Slope

Back 36

Amount that the line moves up for every unit it moves across

Front 37

Intercept

Back 37

The point where the regression line crosses the vertical axis (y axis)
Regression constant

Front 38

Multiple Regression

Back 38

Use when there are multiple predictor variables
Suppose we believe that SAT scores, IQ, and high school GPA will all predict cumulative college GPA:
Y = College GPA
X1 = SAT score
X2 = IQ score
X3 = High school GPA

Front 39

Bivariate vs. Multiple Regression

Back 39

In bivariate regression, standardized regression coefficient (β) is equal to the correlation between the predictor and the criterion
In multiple regression, (β) refers to the amount of unique contribution of the given predictor to the criterion variable
There is likely to be some overlap among the predictors

Front 40

What kind of relationship does this show

Back 40

Linear Correlational Relationship

Front 41

What relationship does this graph show?

Back 41

Curvilinear Correlational Relationship

Front 42

What type of line does this graph represent

Back 42

Linear Regression Line

Front 43

What are the three different correlations displayed

Back 43

A) Perfect Positive Correlation (+1.00)
B) No Relation (0.00)
C) Perfect negative correlation (-1.00)

Front 44

Predictor variable (usually X)

Back 44

Variable used to predict scores of a individual on another variable

Front 45

Criterion Variable

Back 45

Variable being predicted

Front 46

Regression Constant (a)

Back 46

in a linear prediction rule, a = particular fixed number added to the prediction

Front 47

Regression Coefficient (b)

Back 47

Number multiplied by a person's score on a predictor variable as a part of linear prediction rule

Front 48

Linear Prediction Rule (or Model)

Back 48

Formula for making predictions; that is, formula for predicting a person's score on a criterion variable based on the person's on predictior variable(s).

Front 49

Standardized regression coefficient

Back 49

regression coefficient in standard deviation units it shows the predicted amount of change in SD units of the criterion variable if the value of the particular variable increased by 1 SD