How do I add stronger Jitter to a dataset in R?

Question

red <- apply(red, 2, jitter)

The above is my current code. After applying it to my data frame, I noticed the strength of it to be quite small.

If I try the following, it doesn't work:

red <- apply(red, 2, jitter(red, factor = 5))

Answer 1

I don't think you need apply() for this at all, at least, not if I understand your intention. Here's an example using the amount parameter and some iris data.

From the docs, amount is described as:

numeric; if positive, used as amount (see below), otherwise, if = 0 the default is factor * z/50.

Default (NULL): factor * d/5 where d is about the smallest difference between x values.

First I use apply() as your example, then I increase the jitter without it.

data(iris) # data (for example)

red <- iris # keeping your data.frame name
red <- red[,!colnames(red)=="Species"] 
red <- apply(red, 2, jitter)
var(red)

             Sepal.Length Sepal.Width Petal.Length Petal.Width
Sepal.Length    11.528025   0.1123260    2.2771584    1.085271
Sepal.Width      0.112326   9.2040367    0.8921243    0.296827
Petal.Length     2.277158   0.8921243    9.9957082    1.680199
Petal.Width      1.085271   0.2968270    1.6801995    9.194739

red <- jitter(red, amount = 5)
var(red)

             Sepal.Length Sepal.Width Petal.Length Petal.Width
Sepal.Length   19.5928296   1.2155940    2.0093691   0.9365071
Sepal.Width     1.2155940  16.8435802   -0.6173289   0.4445300
Petal.Length    2.0093691  -0.6173289   16.6392921   1.7726687
Petal.Width     0.9365071   0.4445300    1.7726687  18.0234312

red <- jitter(red, amount = 500)
var(red)

             Sepal.Length Sepal.Width Petal.Length Petal.Width
Sepal.Length    79193.216    2570.665    6473.9229  -2702.2620
Sepal.Width      2570.665   74802.846   -5463.7883   2568.4585
Petal.Length     6473.923   -5463.788   87327.8161    865.1276
Petal.Width     -2702.262    2568.458     865.1276  87766.5185

Increasing factor also does work, but it's just potentially less apparent dependent on the distribution of the data.

Yet another approach is to take the difference of your data's values with and without jitter to get the components that's pure jitter and multiple that by any number you like and add it back to the original data.

head(red)

  Sepal.Length Sepal.Width Petal.Length Petal.Width
1          5.1         3.5          1.4         0.2
2          4.9         3.0          1.4         0.2
3          4.7         3.2          1.3         0.2
4          4.6         3.1          1.5         0.2
5          5.0         3.6          1.4         0.2
6          5.4         3.9          1.7         0.4

head((red + apply(red,2,jitter))*10)

  Sepal.Length Sepal.Width Petal.Length Petal.Width
1    101.91095    70.00209     28.03810    3.831979
2     97.93295    60.18538     27.88743    3.948451
3     93.88483    64.09243     26.13381    3.886840
4     92.19319    61.80549     29.87822    3.818302
5    100.00335    71.95687     28.09964    3.991321
6    108.03858    78.09225     34.06902    8.037211

Answer 2

You could also just apply jitter to the plots you make of the data, instead of the data itself, as below:

mtcars %>% 
    as_tibble() %>%
    select(cyl, vs) %>%
    ggplot(aes(x = vs, y = cyl, fill = cyl)) +
    geom_point() + 
    geom_jitter()

If you want the scatters to be reproducible, include set.seed(123) in your code.