9: Survey Methods and Correlational Research Flashcards
What are some sampling issues in survey research
if the sample is not representative of the population then it is potentially biased
self selection bias: who chooses to partake in a survey (magazines/newspapers)
The difference between surveys and psychological assessment
in psychological assessment, the tests undergo rigorous tests of reliability and validity to accurately reflect a construct
What should be considered when creating an effective survey?
Create items that effectively answer the empirical question(s) at hand and be very careful of the structure and wording of these items
Types of survey questions or statements
Types of survey questions or statements
Open ended questions
closed ended
what is response acquiescence and how is it combated
A tendency to agree with statements
□ In surveys a way to avoid this is to have a balance of favorable and unfavourable statements
Describe the correlation coefficient
Pearson’s r
strength and direction of the correlation ranging from -1 to +1
Describe the coefficient of determination
r^2
the percent of variance in one variable that is explained by the other variable
shared variance
In Y=a+bX, what does each letter mean?
Y is the criterion variable, what is being predicted
X is the predictor variable, what is being used to make the prediction
a is the Y int
b is the slope
what is the regression line
a straight line that best summarizes a correlation
what do correlations help us do
make predictions (when the correlation is significantly greater/less than 0)
what are the two approaches in correlations
Bivariate and multivariate
What does the multivariate approach use
multiple regression:
* One criterion variable (Y)
* Several predictor variables (the Xs)
* The bs are know as “regression weights” - Reflect the relative importance of each predictor
what is the advantage of the multivariate approach
when the influences of several predictor variables are combined, prediction can improve far beyond the single regression cases.
Problems that we can experience with correlation
not causal
directionality problem
third variable problem
What is a partial correlation and how is it implemented
Attempts to control for third variables statistically, used when there is reason to suspect a third variable
Involves incorporating all correlations
- What results is a partial correlation that measures the remaining relationship between the two initial variables with the potential third partialed out or controlled
