Chapter 5 Flashcards
A statistical association between variables
correlation
A statistical association between variables
correlation
Examining potential associations between variables. This is the research into statistical relations and the relations might be a coincidence.
correlational research
What is the difference between correlational research and causational research?
- measure variable x in correlational and manipulate in causational research
- Must eliminate confounding variables in causational research
What is the same between correlational research and causational research?
Attempting to reduce confounding variables
Higher scores of one variable increase with the other variable
positive corrolation
Higher scores of one variable increase as the other variable decreases
negative corrolation
A statistical measure that measures the direction and strength of the linear relation between two variables that have been measured on an interval or ratio scale
pearson’s r
How to understand pearson’s r
the closer it is to -1.00 or 1.00, the stronger the linear relationship is
A statistic used to measure the relation between two quantitative variables when variables are measured on an ordinal scale
spearman’s rho
two things that might effect spearman’s rho
- the way higher or lower are numerically coded can change the results being a negative number to a positive number
- the wording used can change the results
A graph in which data point portray the intersection of X and Y
scatter plot
Why use a scatter plot
- They can show non linear relations
- provide a visual representation of the strength of corrolations
Pearson’s r of .10 to .29
small association
Pearson’s r of .30 to .49
moderate association
Pearson’s r of . 50 to 1.00
strong association
What happens when pearson’s r is squared
shows the variation in the results
Three key criteria used in drawing causal inferences
- Covariation of X and Y. As X changes, Y changes
- Temporal order. Changes in X occur before changes in Y
- Absence of plausible alternative explanations
Why can a correlational study not draw conclusions?
Because X variable is not manipulated, temporal order cannot be established
The problem of ambiguity about whether X did cause Y
Bidirectionality problem
The problem is that there might be another variable between X and Y
variable problem
A correlation between variable X and variable Y is computed while statistically controlling for their individual correlations with a third variable Z
Partial Correlation
Each person participates on one occasion, and all variables are measured at that time
cross-sectional research design
Data are gathered on the same individuals or groups in two or more occasions
longitudinal research design
a type of longitudinal design where variable X is measured at an earlier point then variable Y
prosepective design
Three steps of cross-lagged panel design
- Measure X and Y
- Meausre X and Y again
- examine the pattern of correlations of all variables
Three steps of cross-lagged panel design
- Measure X and Y
- Meausre X and Y again
- examine the pattern of correlations of all variables
What are the main issues with correlational research that does not allow for causational conclusions.
The lack of control over confounding variables and not manipulating the independant variable
A predictor that explores the quantitative, linear relation between two variables. It is often used to predict scores of one variable based on another.
Regression analysis
The varible that we are trying to eliminate or predict
criterion variable
The variable whose scores are used to estimate the criterion variable
predictor variable
A line that is the visual representation of the average on a scatter plot
regression line
Predicting the linear relations between multiple variables
multiple regression
key concept of multiple regression
each new predictor variable must enhanse our ability to predict the criterion variable
Uses for correlational research
- Standardized tests
- Measures used for mental disorders
- Test validation
- Experiments that cannot manipulate the independent variable
- Hypothesis testing to find new theories
Uses for correlational research
- Standardized tests
- Measures used for mental disorders
- Test validation
- Experiments that cannot manipulate the independent variable
- Hypothesis testing to find new theories
Occurs when the range of scores obtained have been limited
range restriction
When scores cluster around the maximum
ceiling effect
What statistical measurement to use in associations involving categorical variables
pearson’s r or spearman’s rho
When can you not use pearson’s r
with non linear relationships
