Chapter 7 Relationships b/w Variables Flashcards
If a scatter plot is has columned data what method can be used to uncolumn data?
Jitter the Data
What methods can be used to correct for over saturated scatter plots?
Adjust Saturation
Use a Hex Bin Plot
What is a correlation?
A statistic intended to quantify the strength of a relationship between two variables
What two methods are used to correct for varying scales and distributions when computing correlations?
Standard Scores (Pearson product-moment correlation coeffcient)
Rank (Spearman rank correlation coefficient)
Mathmatical devinition of standard score?
z_i = (x_i - mean) / standard deviation
What is it better to use a percentile rank instead of a standard score for a coorelation tranformation?
If the data is skewed or has outliers it is more robust to use percentile rank
What is covariance?
The tendency of two variables to vary together
What is the mathmatical definition of covariance?
Mathmatical definition of Pearson’s Correlation?
Where Cov is the Covariance and S_x and S_y are the standard deviations of X and Y
What conclusion can be reached is Pearson’s correlation is near 0?
That the relationship is not a linear relationship
What are the three explanations for Correlation between two variables A and B?
- A causes B
- B causes A
- Some other set of factors causes both A and B
