Embedded Files
Correlation- what is measures and what it does not
Correlation- what is measures and what it does not
You have probably heard this a lot, but it might be one of the most important lessons you can learn- a relationship between to variables (i.e., a correlation) does not mean one thing causes the other one. The same is true with regression (i.e., regressions cannot show causation). One thing can predict the other, but that does not mean one things causes the other. Here are some resources to demonstract this effect.
You have probably heard this a lot, but it might be one of the most important lessons you can learn- a relationship between to variables (i.e., a correlation) does not mean one thing causes the other one. The same is true with regression (i.e., regressions cannot show causation). One thing can predict the other, but that does not mean one things causes the other. Here are some resources to demonstract this effect.
CrashCourse: Correlation Doesn’t Equal Causation (12:17)
CrashCourse: Correlation Doesn’t Equal Causation (12:17)
A common analytical error hinders biomedical research and misleads the public.
A common analytical error hinders biomedical research and misleads the public.
Ionica Smeets: The danger of mixing up causality and correlation (5:56)
Ionica Smeets: The danger of mixing up causality and correlation (5:56)
Research By Design: Correlation and Causality – Don’t Confuse Them (11:20)
Research By Design: Correlation and Causality – Don’t Confuse Them (11:20)
Pearson Correlation Basics
Pearson Correlation Basics
Below are resources that will teach you the basics of the bi-variate covariance (i.e., two variables vary together) and the correlation between the two variables. Covariance and correlations are essentially the same thing. Covariance is unstandardized (i.e., not always on the same metric) and is often difficult to interpret, while correlation is standardized (i.e., is on the same metric) having a range from -1 to +1 which makes it easier to interpret.
Below are resources that will teach you the basics of the bi-variate covariance (i.e., two variables vary together) and the correlation between the two variables. Covariance and correlations are essentially the same thing. Covariance is unstandardized (i.e., not always on the same metric) and is often difficult to interpret, while correlation is standardized (i.e., is on the same metric) having a range from -1 to +1 which makes it easier to interpret.
Correlation coefficients measure the strength of association between two variables. The most common correlation coefficient, called the Pearson product-moment correlation coefficient, measures the strength of the linear association between variables measured on an interval or ratio scale.
Correlation coefficients measure the strength of association between two variables. The most common correlation coefficient, called the Pearson product-moment correlation coefficient, measures the strength of the linear association between variables measured on an interval or ratio scale.
Introductory Statistics: Open Textbook Library
Introductory Statistics: Open Textbook Library
Research By Design: Four Uses of Correlation in Statistics (4:32)
Research By Design: Four Uses of Correlation in Statistics (4:32)
Research By Design: Correlation Scatterplots (11:51)
Research By Design: Correlation Scatterplots (11:51)
Non-Parametric Correlations
Non-Parametric Correlations
Often times we cannot met the assumption of normality with our data (or we do not have continuous data). All is not lost, there are a series of statistical tests we can use (called non-parametric or distribution free) when our data volatile the assumptions of normal or continuous data.
Often times we cannot met the assumption of normality with our data (or we do not have continuous data). All is not lost, there are a series of statistical tests we can use (called non-parametric or distribution free) when our data volatile the assumptions of normal or continuous data.
Shawn Janzen: Non-parametric: Spearman's rho and Kendall's tau (16:55)
Shawn Janzen: Non-parametric: Spearman's rho and Kendall's tau (16:55)
Tetrachoric, polychoric, point-biserial, biserial, and polyserial correlations
Tetrachoric, polychoric, point-biserial, biserial, and polyserial correlations
Tetrachoric correlation is used to measure rater agreement for binary data; Binary data is data with two possible answers—usually right or wrong. The tetrachoric correlation estimates what the correlation would be if measured on a continuous scale
Polychoric correlation measures agreement between multiple raters for ordinal variables (sometimes called “ordered-category” data). Ordinal variables can be placed in order, but can’t be divided or multiplied.
The point-biserial correlation coefficient is a special case of Pearson’s correlation coefficient. It measures the relationship between two variables when one continuous variable (must be ratio scale or interval scale) and one is a naturally binary variable.
Biserial correlation is almost the same as point biserial correlation, but one of the variables is dichotomous ordinal data and has an underlying continuity.
Polyserial correlation is between a quantitative variable and an ordinal variables, based on the assumption that the joint distribution of the quantitative variable and a latent continuous variable underlying the ordinal variable is bivariate normal.
Report abuse