Exploring relationships within groups

Relationships within groups

As demonstrated in Crabtree and Nelson (2017), group-level plots can help us see whether an expected relationship appears in particular cases. We’d use those patterns to guide further investigation, with the research design still determining what we can say about causes.

library(plotrr)
a <- runif(400, min = 0, max = 1)
b <- a + rnorm(400, mean = 0, sd = 1)
c <- rep(c(1:4), times = 100)
data <- data.frame(a, b, c)
bivarplots("a", "b", "c", data)

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

library(plotrr)
a <- runif(400, min = 0, max = 1)
b <- a + rnorm(400, mean = 0, sd = 1)
c <- rep(c(1:4), times = 100)
data <- data.frame(a, b, c)
dotplots("a", "b", "c", data, 20)

library(plotrr)
a <- runif(400, min = 0, max = 1)
b <- a + rnorm(400, mean = 0, sd = 1)
c <- rep(c(1:4), times = 100)
data <- data.frame(a, b, c)
violinplots("a", "b", "c", data)

Comparing measures

We can use these plots to examine competing measures within groups. Agreement between measures is one part of assessing construct validity (Trochim & Donnelly, 2008). A disagreement in a particular case can also suggest a useful question: what is each measure capturing there?

Other visualization functions

histplots() prints a histogram of x for each group. Set n to choose the number of bins; the y argument is unused.

library(plotrr)
a <- rnorm(400, mean = 2, sd = 100)
b <- a + rnorm(400, mean = 0, sd = 1)
c <- rep(c(1:4), times = 100)
data <- data.frame(a, b, c)
histplots("a", "b", "c", data, 5)

bivarrugplot() adds a mark along each axis for every observed value. The rugs help show where observations are concentrated.

library(plotrr)
a <- runif(1000, min = 0, max = 1)
b <- a + rnorm(1000, mean = 0, sd =1)
data <- data.frame(a, b)
bivarrugplot("a", "b", data)

## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
## ℹ The deprecated feature was likely used in the plotrr package.
##   Please report the issue at <https://github.com/lobsterbush/plotrr/issues>.
## This warning is displayed once per session.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.

## Warning: Use of `data[[x]]` is discouraged.
## ℹ Use `.data[[x]]` instead.

## Warning: Use of `data[[y]]` is discouraged.
## ℹ Use `.data[[y]]` instead.

## Warning: Use of `data[[x]]` is discouraged.
## ℹ Use `.data[[x]]` instead.

## Warning: Use of `data[[y]]` is discouraged.
## ℹ Use `.data[[y]]` instead.

Other functions

lengthunique() counts distinct non-missing values. makefacnum() reads a factor’s labels as numbers; non-numeric labels become missing values with a warning. clear() prints a form-feed character, which clears some console displays.