この節の作者: Ravi Selker, Jonathon Love, Damian Dropmann

Partial Correlation (`corrPart`)¶

Description¶

Partial correlation matrices are a way to examine linear relationships between two or more continuous variables while controlling for other variables

Usage¶

corrPart(
  data,
  vars,
  controls,
  pearson = TRUE,
  spearman = FALSE,
  kendall = FALSE,
  type = "part",
  sig = TRUE,
  flag = FALSE,
  n = FALSE,
  hypothesis = "corr"
)

Arguments¶

`data`	the data as a data frame
`vars`	a vector of strings naming the variables to correlate in `data`
`controls`	a vector of strings naming the control variables in `data`
`pearson`	`TRUE` (default) or `FALSE`, provide Pearson's R
`spearman`	`TRUE` or `FALSE` (default), provide Spearman's rho
`kendall`	`TRUE` or `FALSE` (default), provide Kendall's tau-b
`type`	one of `'part'` (default) or `'semi'` specifying the type of partial correlation to calculate; partial or semipartial correlation.
`sig`	`TRUE` (default) or `FALSE`, provide significance levels
`flag`	`TRUE` or `FALSE` (default), flag significant correlations
`n`	`TRUE` or `FALSE` (default), provide the number of cases
`hypothesis`	one of `'corr'` (default), `'pos'`, `'neg'` specifying the alernative hypothesis; correlated, correlated positively, correlated negatively respectively.

Details¶

For each pair of variables, a Pearson's r value indicates the strength and direction of the relationship between those two variables. A positive value indicates a positive relationship (higher values of one variable predict higher values of the other variable). A negative Pearson's r indicates a negative relationship (higher values of one variable predict lower values of the other variable, and vice-versa). A value of zero indicates no relationship (whether a variable is high or low, does not tell us anything about the value of the other variable).

More formally, it is possible to test the null hypothesis that the correlation is zero and calculate a p-value. If the p-value is low, it suggests the correlation co-efficient is not zero, and there is a linear (or more complex) relationship between the two variables.

Output¶

A results object containing:

results$matrix a (semi)partial correlation matrix table

Tables can be converted to data frames with asDF or as.data.frame(). For example:

results$matrix$asDF

as.data.frame(results$matrix)

Examples¶

data('mtcars')

corrPart(mtcars, vars = vars(mpg, cyl, disp), controls = vars(hp))

#
#  PARTIAL CORRELATION
#
#  Partial Correlation
#  ----------------------------------------------------
#                           mpg       cyl       disp
#  ----------------------------------------------------
#    mpg     Pearson's r         —
#            p-value             —
#
#    cyl     Pearson's r    -0.590         —
#            p-value        < .001         —
#
#    disp    Pearson's r    -0.606     0.719        —
#            p-value        < .001    < .001        —
#  ----------------------------------------------------
#    Note. controlling for 'hp'
#

Partial Correlation (corrPart)¶