Week 11
No description yet. This curated deck is ready for daily study.
Card preview (19)
When do we use correlation?
• Exploring and describing possible relationships • Measuring validity and reliability • Equipment calibration
What symbol is used as the correlation coefficient
We use and report with the symbol r
Define Pearson Correlation Coefficient
is a measure of linear correlation between two sets of data
Define p-value
A value that tells us if our results/ findings r statistically significant
What is the p-value threshold for a significant result
<0.05
What does <0.05 denote
5% probability that the results happened by chance
Classifying correlations - ,r - 0, - ,r = 0.1 - 0.3 = - ,r 0.4 -0.6 = - ,r 0.7 - 0.9 = - ,r = 1 =
• ,r - 0, No effect • ,r = 0.1 - 0.3 = Weak • ,r 0.4 -0.6 = Moderate • ,r 0.7 - 0.9 = Strong • ,r = 1 = Perfect
Correlation tells yo about the relationship but does not tell you …..
whether one variable causes the other
Define regression
A measure of the relation between the mean value of one variable and corresponding values of other variables
Define bivariate regression
is a linear equation describing the relationship between an explanatory variable and an outcome variable, specifically with the assumption that the explanatory variable influences the outcome variable, and not vice-versa
What is done to a variable during bivariate regressio
Squared
Regression equation
Y =bX + C
What does each letter of the regression equation represent
• Y is the dependent variable (Need to find) • X is the independent variable (Given X) • B is the slope or regression coefficient • C is the intercept of the Y axis (Need to find)
Bivariate regression can be used for what two things
• Assess the shared variance between two variables 2. Predicted a value on one variable, using the value of another variable
Define multiple regression
Extension on linear regression with more variables
Three types of multiple regression
Forced entry stepwise hierarchical
Explain each type of multiple regression
• Forced Entry • Produce one R(squared) value, e.g. .65, or 65% 2. Stepwise • Produce one or more R(squared) values for variables that explain variance • Height .35 or 35% • Vigorous PA .15 or 15% 3. Hierarchical • The best method • Experimenter can decide what order variables are entered Produce R(squared) values at each step
Multiple regression equation
• Y = bX1 + bX2 + bX3…… + c
Multiple regression can be used for what three things
• Assess the shared variane between more than two variables • Difference approaches to assess shared variance • Predicted a value on one variable, using the value of other variables