Chapter 9—Survey Methods Flashcards
(23 cards)
What can cause sampling issues in survey research? How does self-selection bias play in?
If the sample is not representative of population parameters, the sample is potentially biased.
Self selection bias: when the sample is composed of only those who voluntarily choose to respond, the results can be a biased sample
Describe response acquiescence
The tendency for participants to agree with statements regardless of the content
What is a double-barrelled question?
A question that asks for two responses at once, often because the survey writer includes too much info
What is a leading question?
A question structured so that it is likely to produce an answer desired by the asker.
What is nonresponse bias?
The effect that occurs when people who do return surveys tend to differ in some important way from those who do not, making the data less representative
What is sugging?
A term fro “selling under the guise of a survey”
Differ positive and negative correlation
Pos: increase associated with increase, decrease associated with decrease
Neg: increase associated with decrease, decrease associated with increase
What is Pearson’s r and what does it do?
The coefficient of correlation.
It’s size indicates the strength and direction of a correlation
What is the coefficient of determination r^2, and what does it describe?
The percent of variance in one variable that is explained by the other variable
Tells us how to interpret a correlation. The r^2 value represents the percentage of variability in one variable that can be associated with the other
What is regression analysis? What does it tell us?
Making predictions based on correlations.
If a statistically significant correlation exists between two variables, knowing a score on one variable enables you to predict a score on the other
What is the regression line?
The line of best fit; tells us the means for making predictions
In a regression equation, what are the two variables called? What’s their difference?
Criterion: the variable with an unknown that we want to find
Predictor: a given value that we use to predict the value of the criterion variable
What is the bivariate approach?
A statistical analysis investigating the relationship between 2 variables
What is the multivariate approach
Examining the relationships among more than 2 variables
What is multiple regression? What does issue does it solve, and what does it allow us to do?
MR: one criterion variable and a minimum of 2 predictor variables
Solves the problem of having more than one predictor of an outcome
Allows us to:
1. See how well all the predictors together explain or predict the outcome
2. Understand how much each predictor uniquely contributes after accounting for the others
What is the multiple correlation coefficient R, and how do we get it?
The correlation between the combined predictors and the criterion
We get to through multiple regression analysis
What is the multiple coefficient of determination R^2?
Tells us the proportion of variance in the criterion variable that is explained by the predictor variables.
What is the essential difference between non-experimental correlational research and experimental research?
In correlational research, it is impossible to hold all extraneous variables constant, thus lacking control and losing cause-effect.
What is the directionality problem in correlational research?
When the hypothesized causal relation found between variables occurs in both directions; A —> B and B —> A
What is a cross-lagged panel correlation, and what does it have to do with directionality?
A design that enables one to increase their confidence about correlational directionality; this procedure investigates correlations between variables at several points in time. A type of longitudinal design.
What’s the third variable problem?
There may be a third variable that underlies a correlation between A and B; it could be the case that C —> A and B
What is a partial correlation?
An attempt to control the third variable problem statistically; you correlate the variable with A and B individually, and incorporate all 3 correlations to measure the remaining relationship between A and B with C controlled
What is a mediating variable? how is it different from a moderating variable?
Mediating variable: explains how or why an IV affects the DV
—> Like a middle step; A —> C —> B
Moderating variable: affects the strength or direction of the relationship between the IV and DV
—> It answers “when” or “for whom” the effect is stronger
