correlation formatting doesn't apply digits to the CI formatting

[DominiqueMakowski]

Received by email:

However, I noticed that when one changes the number-of-digits display using the digits argument, that the display for the Pearson’s r value and for the associated t statistic are changed, but the confidence interval limits stay at 2-digits status regardless of the argument value.

CI stays with 2 digits (but it works in parameters)

correlation::correlation(iris, digits=1)
#> # Correlation Matrix (pearson-method)
#> 
#> Parameter1   |   Parameter2 |    r |         95% CI | t(148) |         p
#> ------------------------------------------------------------------------
#> Sepal.Length |  Sepal.Width | -0.1 | [-0.27,  0.04] |   -1.4 | 0.152    
#> Sepal.Length | Petal.Length |  0.9 | [ 0.83,  0.91] |   21.6 | < .001***
#> Sepal.Length |  Petal.Width |  0.8 | [ 0.76,  0.86] |   17.3 | < .001***
#> Sepal.Width  | Petal.Length | -0.4 | [-0.55, -0.29] |   -5.8 | < .001***
#> Sepal.Width  |  Petal.Width | -0.4 | [-0.50, -0.22] |   -4.8 | < .001***
#> Petal.Length |  Petal.Width |  1.0 | [ 0.95,  0.97] |   43.4 | < .001***
#> 
#> p-value adjustment method: Holm (1979)
#> Observations: 150

parameters::parameters(lm(Sepal.Length ~ Sepal.Width, data=iris), digits=1)
#> Parameter   | Coefficient |  SE |      95% CI | t(148) |      p
#> ---------------------------------------------------------------
#> (Intercept) |         6.5 | 0.5 | [ 5.6, 7.5] |   13.6 | < .001
#> Sepal Width |        -0.2 | 0.2 | [-0.5, 0.1] |   -1.4 | 0.152
#> 
#> Uncertainty intervals (equal-tailed) and p-values (two-tailed) computed
#>   using a Wald t-distribution approximation.

^{Created on 2025-01-19 with reprex v2.1.1}

[strengejacke]

correlation::correlation(iris, digits=1, ci_digits = 5)
#> # Correlation Matrix (pearson-method)
#> 
#> Parameter1   |   Parameter2 |    r |               95% CI | t(148) |         p
#> ------------------------------------------------------------------------------
#> Sepal.Length |  Sepal.Width | -0.1 | [-0.27269,  0.04351] |   -1.4 | 0.152    
#> Sepal.Length | Petal.Length |  0.9 | [ 0.82704,  0.90551] |   21.6 | < .001***
#> Sepal.Length |  Petal.Width |  0.8 | [ 0.75690,  0.86484] |   17.3 | < .001***
#> Sepal.Width  | Petal.Length | -0.4 | [-0.55088, -0.28795] |   -5.8 | < .001***
#> Sepal.Width  |  Petal.Width | -0.4 | [-0.49721, -0.21870] |   -4.8 | < .001***
#> Petal.Length |  Petal.Width |  1.0 | [ 0.94905,  0.97299] |   43.4 | < .001***
#> 
#> p-value adjustment method: Holm (1979)
#> Observations: 150

^{Created on 2025-01-19 with reprex v2.1.1}

[DominiqueMakowski]

what about the inconsistency with parameters() in which it gets changed by ci?

[DominiqueMakowski]

I would still have by default ci_digits=digits, and only have them different if both are specified by the user

[strengejacke]

Ok, we just need to add it to the print and/or format methods, I guess?

[DominiqueMakowski]

I think so probably here

correlation/R/methods_format.R

Lines 4 to 9 in 94f3d0a

	format.easycorrelation <- function(x,
	digits = NULL,
	p_digits = NULL,
	stars = NULL,
	format = NULL,
	...) {