nhstplot is a fairly simple package to use. This vignette is intended to explain the basics (plotting using the defaults), before showing how to use the options.

The basics

After installing the library with install.packages("nhstplot") you need to load the library:

library(nhstplot)

nhstplot is composed of 4 functions, one for each major NHST test “family” :

Let’s see how to use each one without changing the graphical options.

\(\chi^2\) tests

The plotchisqtest function only requires 2 arguments : The first one is the \(\chi^2\) value (parameter : chisq), and the second one is the degrees of freedom (parameter df).

Here’s an example with respectively 8 and 4.

plotchisqtest(chisq = 8, df = 4)

Note that the same is achieved with plotchisqtest(8,4).

You can also use the function by passing an object created by chisq.test()…

test <- chisq.test(c(A = 37, B = 18, C = 25))
plotchisqtest(test)

…or a model comparison from the anova() function.

set.seed(1)
y <- rbinom(10, 1, .4) ; x <- 2*y + rnorm(10)
fit1 <- glm(y ~ 1, family = binomial)
fit2 <- glm(y ~ x, family = binomial)
comp <- anova(fit1, fit2, test = "Chisq")
plotchisqtest(comp)

\(F\) tests

The plotftest function only requires 3 arguments : The first one is the \(F\) value (parameter : f), and the second and third ones are respectively the degrees of freedom of the numerator (parameter dfnum) and the denominator (parameter dfdenom).

Here’s an example with respectively 4, 3 and 5.

plotftest(f = 4, dfnum = 3, dfdenom = 5)

Note that the same is achieved with plotftest(4,3,5).

You can also use the function by passing an object created by lm()…

x <- rnorm(10) ; y <- x + rnorm(10)
fit <- lm(y ~ x)
plotftest(fit)

…or by passing an F-change test computed with the anova() function:

set.seed(1)
x <- rnorm(10) ; y <- x + rnorm(10)
fit1 <- lm(y ~ x)
fit2 <- lm(y ~ poly(x, 2))
comp <- anova(fit1, fit2)
plotftest(comp)

\(t\) tests

The plotttest function only requires 2 arguments : The first one is the \(t\) value (parameter : t), and the second one is the degrees of freedom of the numerator (argument df).

Here’s an example with respectively 2 and 10.

plotttest(t = 2, df = 10)

Note that the same is achieved with plotttest(2,10).

By default, the plotttest function plots a two-tailed test. However, a one-tailed test can be plotted by adding the argument tails = "one":

plotttest(2, 10, tails = "one")

The left or right tail is automatically selected using the sign of provided \(t\):

plotttest(-2, 10, tails = "one")

You can also use the function by passing an object created by t.test()…

test <- t.test(rnorm(10), rnorm(10))
plotttest(test)

…or cor.test().

test <- cor.test(rnorm(10), rnorm(10))
plotttest(test)

\(z\) tests

The plotztest function only requires 1 argument : The \(z\) value (parameter z).

Here’s an example with a \(z\) value of 2.

plotztest(z = 2)