Bivariate Correlational Res (Week 8) Flashcards
What is an important distinguishing feature of association claims?
They include at least two variables that are measured
Define bivariate?
two variables
What is a correlation?
What influences the researchers choice in terms of IV and DV
The direction and strength of two variables
Theory and interpretation
What is a correlation coefficient?
What are the different standardization?
Determines the strength of a correlation - also known as Pearson’s r
Standardized between -1 and 1
Positive c: r > 0
Negative c: r < 0
Neutral: r = 0 (or fairly close)
What components determine statistical validity?
Hint: CORES
Effect size Statistical significance Outliers Restriction of range Curvilinear relations
Define effect size, what is the question?
What is an absolute value?
What is a big correlation?
Tells us the strength of a correlation (or magnitude) - the closer the absolute value of the correlation is to 1, the larger the effect size
Absolute value: taking away the negative sign (ex: absolute of -0.65 is 0.65)
A large effect size depends on context, give an example of this?
0.3 correlation for heart failure and pollution is more significant than 0.6 between meeting online and marital satisfaction
What is a p-value?
If we assume there is no effect/association, p-value is the probability of finding this effect/association or a larger one, given the data
How do you determine statistical significance?
When the p-value is less than 0.5 the result is very unlikely to have come from 0 association population
What is an outlier?
An extreme score that stands out from the pack
Depending on where it is, can make a medium-sized correlation look stronger or a strong one appear weaker than it really is
Define restriction of range? Give an example
If there is not a full range of scores on one variable in an association it can make correlation look smaller than it is
Ex: If you only look at college grades and SAT scores in a highly selective college, the association seems smaller than it actually is
What is a curvilinear association?
The relationship between two variables is not a straight line, positive up point and then becomes negative
What is a spurious association? what kind of validity is it associated with?
the bivariate correlation is there, but only because of some third variable - a reason to dig deeper and ask more questions
Internal validity
What is a moderating variable?
When the relationship between two variables changes depending on the level of another variable that other variable is an M
What is the best visual method for determining correlations?
Scatterplot
