correlational research Flashcards
whats the difference between a predictor variable and a criterion variable
predictor variable is generally the independent variable, and the criterion variable is general the dependent variable
what are the three characteristics that a correlational coefficient describes
the direction, form, and strength of a relationship
whats the difference between quantitative and categorical variables
quantitative have meaningful, numerical value such as scores on rating scales
categorical are distinct categories based on qualitative differences
which correlational coefficient is used for two variables at the ratio or interval level
pearsons r
which correlational coefficiant is used for two variables at the ordinal level
spearman rank-order r
which correlational coefficient is used when there is one dichotomous (2 categories) variable
point-biserial r
what test do you use if both variables are non-numerical
chi square test
The likelihood that the sample’s
association came from a population
in which there is no association
p value
The magnitude, or strength, of a relationship between
two or more variables
effect size
what is the coefficient of determination
the squared value of a correlation
what is the coefficient of determination used for
it measures the percentage of variability in one variable, determined or predicted by its relationship to the other variable
what are the strengths of correlational research (3)
- describes relationships
- nonintrusive
- high external validity
what are the weaknesses of correlational research (4)
- cannot assess causality
- third variable problem
- directionality problem (dont know which effects the other)
- low internal validity
what are outliers
an extreme or highly unusual score. highly deviant from the rest of the data
what are the two types of outliers
online outliers and offline outliers
what do online outliers do to the correlation coefficient
they inflate the correlation coefficient
true or false: offline outliers inflate the correlational coefficient
false, they weaken the relationship
how do you detect outliers
scores that are + or - 3 standard deviations from the mean, or median absolute deviation
how do you handle outliers
either remove from the dataset before inferential stat analysis, or keep it in but recode with values equal to 3 SDs from the mean
*must be reported
what is restriction of range
a sample under study does not include full range of variables
a statistical procedure for studying multivariate relationships
multivariate regression
Statistical analysis that computes the relationship between the predictor variable and the criterion variable, controlling for other predictor variables.
multiple regression/multivariate regression
Holding a potential third variable constant, statistically or
experimentally, when investigating the relationship between
two other variables
controlling for third variable
what does it mean to control for a third variable
finding the unique overlap between the predictor and criterion of interest; removing proportion of variability that can be explained by third variable
