Term
|
Definition
| Tentative statement about how 2 variables are associated in larger populations. |
|
|
Term
|
Definition
| no comparison or relationship between variables. Any difference in sample is random error |
|
|
Term
|
Definition
| inferential statistic used to determine whether the mean for two groups of scores differs statistically and can be applied to entire population |
|
|
Term
|
Definition
IV: Categorical (2 variables)
DV: Quantitative |
|
|
Term
| T-Test steps of hypothesis |
|
Definition
1. formulate H1
2. Formulate H0
3. Select stat sig difference (p<0.05)
4. select correct test
5. Test stats for significance
6. make decision based on stats if stat sig and on HO & H1 |
|
|
Term
| Statistically Significant |
|
Definition
| Probability, the smaller the p, the more confident we are of legit difference |
|
|
Term
| Significance level is same as |
|
Definition
| Two tailed or P value must be below 0.05. Reject H1 if greater than 0.05. not statistically significant. How certain we are no relationship exists. |
|
|
Term
|
Definition
Degrees of freedom (df): N is considered
T: Farther from zero = stronger
P: lower than 0.05 |
|
|
Term
|
Definition
| Describes size of a difference of an association in STANDARDIZED unit. (magnitude of difference) COHNAN'S D. |
|
|
Term
| Effect size measures for T-Test: |
|
Definition
D = 0.20 (.15-.39) SMALL EFFECT
D = 0.50 (.4-0.74) MEDIUM EFFECT
D = 0.80 (.75-1.09) LARGE EFFECT |
|
|
Term
|
Definition
| Tells the magnitude of difference between two variables. If measures 0.05 very small or negligible effect. |
|
|
Term
| P (stat. sig) is tied to.. |
|
Definition
|
|
Term
|
Definition
| analysis used to measure strength of association between two quantitative variables. X scores tell us something about Y scores to help predictions |
|
|
Term
| Chi Squared (x2) variables |
|
Definition
ID: Categorical
DV: Categorical |
|
|
Term
| R (Pearson's Correlation) variables |
|
Definition
ID: Quantitative
DV: Quantitative |
|
|
Term
|
Definition
Graphical depiction of a correlation
Y axis: DV (Vertical)
X axis: ID (horizontal) |
|
|
Term
| 4 relationships of scatter plots |
|
Definition
1. No relation
2. Positive linear relation
3. Negative (Inverse) Relation
4. Curvilinear (non-linear) |
|
|
Term
|
Definition
| Knowing peoples level of variable X tells us NOTHING about variable y. Straight line. No shared Variance |
|
|
Term
| Positive linear relationship |
|
Definition
as X increases, so does y.
As X decreases, so does Y.
The steeper the slope line, stronger the relationship |
|
|
Term
| Negative (Inverse) relationship |
|
Definition
| Scores move opposite of eachother, as X increases Y decrease, as Y increases X decreases. |
|
|
Term
|
Definition
| relationship between X and Y is curved or it changes over time. Such as texting and driving (fear appeal works for certain amount of time until people ignore it) |
|
|
Term
|
Definition
| Sad dog commercial, worrying too much about grades makes 2nd guess and change. |
|
|
Term
|
Definition
| provides numerical estimate of direct and strength of the linear relationship between two variables |
|
|
Term
| Pearson correlation (r) must be what two things?? |
|
Definition
Presumes relationship is linear
Possible range (-1.00-+1.00): closer to 1 is a stronger relationship |
|
|
Term
|
Definition
| sign of the correlation (+/-) depicts the direction of the relationship between X and Y |
|
|
Term
| Strength/what related to. |
|
Definition
| size of correlation depicts the strength of relationship between X & Y AND related to effect size. |
|
|
Term
|
Definition
tells there is no relationship between X and Y
-complete overlap means will fall everytime! |
|
|
Term
| "Rule of thumb" for interpreting r |
|
Definition
(Strength of relationship)
r = 0.1 SMALL
r = 0.24 MEDIUM
r = 0.37 LARGE |
|
|
Term
Which is stronger?
r = 0.25 or r = -0.47 |
|
Definition
|
|
Term
| Coefficient of determination is called what in SPSS? |
|
Definition
|
|
Term
| Effect size for correlation (Coefficient of determination) |
|
Definition
| must square the r (r2) % of variance in variable X that is shared by Y value. r2=0.06 means 6% of X is shared by Y. |
|
|
Term
|
Definition
| exact percent of variance two scores share or the percent they move together |
|
|
Term
| CORRELATION DOESN'T EQUAL CAUSE. why? |
|
Definition
| just because there is a correlation between X and Y, by itself, doesn't show that X is the cause of Y |
|
|
Term
| 3 correlation doesnt equal cause reasons |
|
Definition
Co variation- are X and Y related? Is there a relationship?
Time order- Does X come before Y in time
Alternative explanations- Is anything besides X causing Y? |
|
|
Term
| Values needed in Correlation determination |
|
Definition
r: Strength of relationshpi
r2: percent related X to Y
N: sample size
Sig fig: probability |
|
|
Term
|
Definition
| inferential statistics used to address hypotheses nominal variables |
|
|
Term
|
Definition
IV: Nominal (2 categories)
DV: Nominal (2 categories) |
|
|
Term
| Statistical significance for chi square |
|
Definition
| compares what we observe/actually see vs what we expect by chance/average to see |
|
|
Term
| Degrees of freedom for chi test |
|
Definition
| Found on SPSS output (Rows-1)(Columns-1) |
|
|
Term
| Effect size measurement for chi test |
|
Definition
|
|
Term
| Effect size for all three tests |
|
Definition
Correlation: r2
Chi: Cramers
T-test: Cohens |
|
|
Term
|
Definition
takes on value from 0 to 1
0= one variable has no impact on other
1= one variable totally determines the other |
|
|
Term
| Rule values for Cramer's V |
|
Definition
0.10= SMALL EFFECT
0.24= MODERATE/MEDIUM EFFECT
0.37= LARGE EFFECT |
|
|
