Remember, no causal language with a correlational design
This is harder than you think...
What NOT to Write
"...we expected that liberalism would lead to less support for the death penalty..."
"...the effect of liberal political views on death penalty attitudes..."
"...the influence of liberalism on..."
What You SHOULD Write
"...we expected that liberalism would be associated with more opposition to..."
"...the relationship between liberal political views and death penalty attitudes..."
"...the correlation of liberalism with death penalty support..."
Why Include Means?
Means and s.d. are included for descriptive purposes only
They do not tell us anything about the relationship between variables
The fact that 5.0 > 3.8 has nothing to do with the correlation
Correlations Again
Describe the relationship between two variables
They do not compare means on variables
They do analzye the entire range of both variables
Study #1
We measured two continuous variables
We did not classify/categorize people as either pro- vs. anti-DP
We did not classify people as liberal vs. conservative
Study #1 (contíd)
Therefore, we can say the more liberal a person was, the less DP support she showed
We cannot say that liberals were more supportive of the DP
We cannot say that people with liberal attitudes were more likely
to oppose the DP
Back to Overheads Main Page
Back to 381-002 Main Page