Chapter 8: Bivariate Correlational Research Flashcards
Bivariate Correlation
An association that involves two variables. Also called bivariate association.
At minimum, how many variable are there in an association claim?
Two
What characteristics of a study make it correlational?
All variables are measured.
What do the following three scatterplots look like?
- one showing a positive correlation
- one showing a negative correlation
- one showing a zero correlation
- Positive: A scatterplot with a positive correlation is a graph showing that all data points are in a pattern trending upwards from left to right. The scatterplot shows that, as x increases, y increases as well, which means the data points have a positive association or relationship.
- Negative: A scatterplot with a negative correlation is a graph showing that all data points are in a pattern trending down from left to right. The scatterplot shows that, as x increases, y decreases, which means the data points have a negative association or relationship.
- Zero: When all the points on a scatterplot lie on a straight line, you have a perfect correlation between the two variables (see below). A scatterplot in which the points do not have a linear trend (either positive or negative) is called a zero correlation or a near-zero correlation.
What do the two bar graphs look like?
- one that shows a correlation between two variables
- one that shows no correlation
A bar graph that shows a correlation should have bars at different heights; a bar graph with a zero correlation would show two bars of the same height.
When do researchers typically use a bar graph instead of a scatterplot to display correlational data?
In a bar graph, we would examine the difference between the group averages to see whether there is an association. It shows the mean. The difference in means indicates an association between variables.
Mean
An arithmetic average; a measure of central tendency computed from the sum of all the scores in a set of data, divided by the total number of scores.
What validities are used to interrogate association claims?
- Construct, statistical, and external.
- Although internal validity is relevant for causal claims, not association claims, you need to explain why correlation studies do not establish internal validity.
Effect Size
The magnitude, or strength, of a relationship between two or more variables.
- Tiny effect sizes can become more important when aggregated over many situations and/or people
Statistically Significant
In NHST, the conclusion is assigned when p < .05, that is, when it is unlikely the result came from the null hypothesis population.
Restriction of Range
In a bivariate correlation, the absence of a full range of possible scores on one of the variables, so the relationship from the sample underestimates the true correlation.
- the limitation—via sampling, measurement procedures, or other aspects of experimental design—of the full range of total possible scores that may be obtained to only a narrow portion of that total.
Curvilinear Association
An association between two variables which is not a straight line; instead, as one variable increases, the level of the other variable increases and decreases (or vice versa)
To establish causation, a study must satisfy three criteria:
- covariance of cause and effect. The results must show a correlation, or association, between the cause variable and the effect variable.
- Temporal precedence. The method must ensure that the cause variable precedes the effect variable; it must come first in time.
- Internal validity; there must be no plausible alternative explanation for the relationship between the two variables.
Directionality Problem
In a correlational study, the occupance of both variables is measured around the same time, making it unclear which variable in the association came first.
Third-variable problem
In a correlational study, the existence of a plausible alternative explanation fo the association between two variables.
Spurious Association
A bivariate association that is attributable only to systematic mean differences in subgroups within the sample; the original association is not present within the subgroups.
Spurious Association: Internal Validity
A third variable is potentially behind the relationship.
Spurious Association: Non-internal Validity
This means that a third variable is not behind the relationship. Looking at height and weight, we see a positive correlation. Even when we separate the participants by gender, we still see the same relationship.
Moderator
A variable that, depending on its level, changes the relationship between two other variables.
In one or two brief sentences, explain how you would interrogate the construct validity of a bivariate correlation.
Questions about the researchers’ operationalization’s
-Does the measure have good reliability?
-Is it measuring what is it intended to measure?
-Face, concurrent, discriminates, convergent validity.
What are six questions you can ask about the statistical validity of bivariate correlation? Do all the statistical validity questions apply the same way when bivariate correlations are represented as bar graphs?
What is the effect size?
Is the correlation statistically significant?
Could outliers be affecting the association?
Is there a restriction on the range?
Is the association curvilinear?
Questions about outliers and curvilinear associations may not be relevant for correlations represented as bar graphs.
Which of the three rules is almost always met by bivariate correlations? Which two rules might not be met by a correlational study?
If there is a significant association between two variables, covariance will be established.
- other two that might not be met is temporal precedence, and internal validity
Give examples of some questions you can ask to evaluate the external validity of a correlational study.
You ask whether the association can be generalized to other people, places, and times.
- who the participants were and how they were selected
- the way the sample was selected from the population of interest (I.e., were selected using random sampling?)
If we found that gender moderates the relationship between deep talk and well-being, what might that mean?
It would mean the relationship between deep talk and well-being is different for men than for women. For example, the relationship might be stronger for women that it is for men.
