Correlational Research Strategy
Strengths an Weaknesses
Strengths
foundation for further directional research
Nonintrusive
High external validity
Weaknesses
No Information on causality
low internal validity
Third Variable Problem
Directionality Problem
Goals
Differences to
describe relationship bewteen two are more variables
no attempt to
manipulate
control
interfere
Data presentation
Experimental research
Differential research
focuses on difference between two groups
demonstrates cause-and-effect relationship
Data in Scatter Plot
Relationship
relationship characteristics
Direction
positive relationship
negative relationship
r < 1
r > 1
Form
Consistency(Strength)
linear
monotonic
straight line in scatter plot
Pearson Correlation
Spearman correlation
one-directional trend
amount of increase need not to be constantly at the same size
variable Y indreases in a consistently predictable amount compared to X
Data in list
Non-Numerical Relationships
one numerical & one boolean value
sign of correlation become meaningless
Use the non numerical variable to organize the scores into separate groups: the data would consist of a group of scores for one boolean value, and a group of scores for the second boolean value
Both Non-Numerical Scores from Nominal Scales
Chi Square Test
Coefficient Determination
r²
measures percentage of variability in one variable that is determined/predicted
Statistical Significance
relationship found in sample is unlikely random
statistical significant =/= correlation is large
with a large sample it is possible for a correlation of r=0.1 to be statistically significant
Usage of Correlational Research
Prediction
criterion variable
correlational research allows researchers to use knowledge in variable to explan (predict)the second variable
predictor variable
variable being explained
Regression
Goal
find equation that produces most accurate predictions of Y (criterion variable) for each value of X (predictor variable)
simple & well defined
complex and unknown
variable predicting the second one
Reliability
Validity
evaluates the consitency/stability of two measurements
Test-Retest-Reliability
Relationship between an original set of measurements and a follow up set of measurements
Concurrent Validity
The valditiy of the test can be established by demonstrating that the scores from one test are strongly related to the scores from an established test
evaluates the extent to which the measurement actually measures what it claims to measure
Multiple Regression
for more than 2 variables
Note:
Predictor variables only predict, not explain a relationship
used to examine relationship between two specific variables while controlling the influence of other potentially confounding variables
adding predictor variables one at a time shows their individual influence