Home > Tools > Data Relationships > Data Relationships Column Descriptions

Data Relationships Column Descriptions

The Data Relationships table displays a number of different measures for the different types of calculations. A description of the statistics available is found below:

All calculations

Option	Description
Y (numerical/categorical)	The name of the Y column concerned.
X (numerical/categorical)	The name of the X column concerned.
p-value	The calculated p-value, representing the degree to which the first column predicts values in the second column. A low p-value indicates a probable strong connection between two columns.
n	The number of valid pairs.

Linear regression

Option	Description
FStat	The F-statistic calculated according to [Ref. Arnold].
RSq	The squared correlation value.
R	The correlation value.
Df	The degrees of freedom = the number of non-empty rows in the column pair - 2.

Spearman R

Option	Description
FStat	The F-statistic calculated according to [Ref. Lehmann].
Rank R sqared	The square of rank R.
Rank R	The correlation of the ranked values of the X and Y columns.
Df	The degrees of freedom = the number of non-empty rows in the column - 2.

Anova

Option	Description
FStat	The F-statistic. See Anova algorithm for more information.
S2Btwn	The sum of squares between groups.
S2Wthn	The sum of squares within groups.
dfBtwn	The degree of freedom between groups.
dfWthn	The degree of freedom within groups.

Kruskal-Wallis

Option	Description
H-stat	The H-statistic. See Kruskal-Wallis algorithm for more information.
Df	The degrees of freedom = k-1, where k is the number of categories.

Chi-square

Option	Description
Chi2-stat	The Chi2-statistic, which is a direct relationship between the observed and the expected values.
Df	The degrees of freedom = (I-1)(J-1) where I is the number of unique values in the first column and J is the number of unique values in the second column.