# PSYCH 501 Module 4

The flashcards below were created by user ralissa on FreezingBlue Flashcards.

1. predictor variable
a variable which is believed to predict another variable
2. covariance
describes the linear association between two quantitative variables
3. scatterplot
graphical representation of the association between two quantitative variables
4. Pearson Correlation
• a measure of the association between two variables which is defined on a scale from -1 to 1, with 1 being a perfect positive relationship and a -1 being a perfect negative relationship
•
5. coefficient of determination
measure of the association between two variables, equal to the squared Pearson correlation. Describes the proportion of variance in one variable that is linearly associated with the other variable
6. linear regression model
can be used to descrive the relation between x and y in a random sample of participants
7. random-x model
linear regression model in which each person in a random sample is assigned a pair of x and y scores, where the x values observed in the sample will not be known in advance.
8. fixed-x model
linear regression model in which the values of x are predetermined by the researcher
9. slope
describes the change in the predicted y score associated with any 1-point increase in x; 1
10. least squares estimate
the unique values that minimize the sum of all squared residual scores ()
11. centering predictor variable
subtracting  from each x score so that the predictor variable's mean will equal 0 and
12. Fisher transformation
• used to transform a Pearson correlation coefficient to obtain a confidence interval
13. dummy coded predictor variable
setting xi=1 for every participant in the first group and xi=0 for every participant in the second group; in this model,
14. point-biserial correlation
describes the strength of the relation between a quantitative response variable and a dichotomous predictor variable
15. reliability
the squared Pearson correlation between the true scorfes and the measured scores
16. measurement error
the difference between the true score and the observed measurement of the true score
17. alternative form reliability
the Pearson correlation between two measurements of a randomly selected participant's particular attribute
18. test-retest relibaility
Pearson correlation between two occasions of the same form for a randomly selected participant
19. inter-rater reliability
the Pearson correlation between two measurements of a randomly selected participant given by two different raters
20. validity
the degree to which a measurement assesses the attribute it claims to measure
21. criterion validity
obtained by assessing the magnitude of a correlation (Pearson or point-biserial) between the measurement in question and some objective behavioral criterion or performance on a specific task
22. construct validity
obtained by assessing the magnitude of a correlation between the measurement in question (y) and the measurements of several other psychological measurements
23. linearity assumption
the assumption that the relation between y and x is linear
24. errror normality assumption
the assumption that the prediction errors have an approximate normal distribution in the study population
25. bivariate normality assumption
in random-x models, the assumption that x and y have an approximate bivariate normaldistribution
26. outlier
unusually small or large x or y scores
27. data transformations
A logarithmic, square root, or reciprocal transformation of an x or y score to normalize the scores
 Author: ralissa ID: 187347 Card Set: PSYCH 501 Module 4 Updated: 2012-12-12 16:32:17 Tags: Module Folders: Description: Definitions for module 4 Show Answers: