Term
|
Definition
| the relationship between 2 variables. it tells us the magnitude and direction of the relationship through the correlation coefficient. |
|
|
Term
| most common correlation coefficient |
|
Definition
| pearson correlation coefficien |
|
|
Term
| equation of a straight line |
|
Definition
y = a + bX
a= y intercept and b = slope
|
|
|
Term
| 4 reasons why correlation may or may not imply causation |
|
Definition
1. X caused Y
2. Y caused X
3. The correlation between x and y was spurious
4. A third variable was responsible for the correlation between x and y |
|
|
Term
| three relevant questions about correlation |
|
Definition
1. is there a relationship between the variables.
2. if so, whats the strength of the relationship?
3. whats the nature of the relationship? |
|
|
Term
what numbers can the pearson correlation coefficient range from?
|
|
Definition
-1 to +1
0 means no correlation, -1 is perfect negative relationship, +1 is perfect positive relationship. Likelihood of finding a perfect or no relationship is very slim. |
|
|
Term
| how is magnitude determined? |
|
Definition
| how far the correlation coefficient is from 0 |
|
|
Term
|
Definition
| the prediction of one variable form one or more other variables |
|
|
Term
|
Definition
Y^= a + bX
a=slope of regression line
b= y intercept |
|
|
Term
|
Definition
| dependent variable, y, the variable thats being predicted |
|
|
Term
|
Definition
| independent variable, x, the variable from what predictions are made |
|
|
Term
| standard error of the estimate |
|
Definition
| amount of error when predicting y scores from x, also known as the residual |
|
|
Term
|
Definition
| two things that are related but not linear |
|
|
Term
|
Definition
| when you restrict a range of your sample data |
|
|
Term
|
Definition
| can cause a strong relationship to appear weak and vice versa |
|
|
Term
| standard error of the difference |
|
Definition
| The standard deviation of the sampling distribution of the difference between two independent means. It indicates how much sampling error will occur on average. |
|
|
Term
| standard error of the estimate |
|
Definition
| computing the standard error of the mean but taking into account degrees of freedom |
|
|
Term
|
Definition
| needs to be used in place of a z score when the standard error of the mean is unknown |
|
|
Term
| independent groups t-test |
|
Definition
used when
1. the dependent variable quantitative and measured on an interval level
2. the independent variable is between subjects in nature
3. the independent variable has 2 and only 2 levels |
|
|
Term
|
Definition
|
|
Term
|
Definition
| when you subtract the grand mean from all of your other scores |
|
|
Term
|
Definition
| amount of unexplained variability in the dependent variable, the variability that remains after the effects of the independent variable are removed |
|
|
Term
|
Definition
| the total variability in the dependent variable |
|
|
Term
|
Definition
| the amount of influence the independent variable had on the dependent variable |
|
|
Term
|
Definition
| indexes the strength of the relationship between the IV and DV, proportion of variability in the DV that is associated with the IV |
|
|
Term
| weak, strong, average eta squared scores |
|
Definition
.05= weak
.10 = average
.15 = strong |
|
|
Term
|
Definition
used when
1. the dependent variable is quantitative and is measured on an interval level
2. independent variable is WITHIN SUBJECTS in nature (why is why it differs from independent groups t-test)
3. the independent variable has 2 and only 2 levels |
|
|
Term
advantages of correlated groups t-test:
|
|
Definition
| controls for disturbance variables which makes it easier to detect a relationship, provides for a more sensitive test |
|
|
Term
| the most commong disturbance variable |
|
Definition
|
|
Term
| sampling distribution of the mean of difference scores |
|
Definition
| theoretical distrubution consisiting of mean difference scores across all individuals in a sample for all possible random samples |
|
|
Term
| assumptions of the correlated groups t-test |
|
Definition
| the sample is independently and randomly selected, the population of difference scores is normally distributed, the dependent variable is quantitative in nature and measured on an interval level |
|
|
Term
| eta squared in a correlated groups t-test |
|
Definition
| the proporation of variability in the DV that associated with the IV after variability from individual differences has been removed |
|
|
Term
| downside of correlated groups t-test |
|
Definition
| since it is within subjects design it could hold some carryover effects |
|
|
Term
|
Definition
| participants receive all levels of the IV |
|
|
Term
|
Definition
| values of the IV are split up between participants |
|
|
Term
| what happens when you calculate pooled variance |
|
Definition
| the variance should turn out to be the same for the groups |
|
|
Term
| why is nullifying important |
|
Definition
| you need to remove conditions in the independent groups t-test..in the correlated groups t-test it removes individual differences |
|
|
Term
|
Definition
| eliminates familiarity or intervening events when using within subjects design |
|
|
Term
|
Definition
| describe relationships between variables, make inferences about populations |
|
|
Term
|
Definition
| used for estimating variance after looking at several samples where the mean may vary but the variance is thought to be the same |
|
|
Term
| assumption of the independent groups t-test |
|
Definition
| the population variances are homogenous .. also called the homogeneity of variance "o1 2 = o22 = 02" |
|
|
Term
|
Definition
explains the proportion of variability in the dependent variable that can be explained by the independent variable
aka the coefficient of determination |
|
|
Term
|
Definition
| the coefficient of alienation |
|
|
