I am running into some problems with plotting a histogram.
The dataset is comprised of 4 Sampling days. During these four days prependicular distances to individuals were counted. But now i need to plot the frequency of the perpendicular distance of every individual per day into a histogram.
My dataset has all four samplings(days) in one excel sheet under each other. but now i need to seperate the data for each day so i can plot seperate histograms.
I was able to plot the histogram of the four days together with hist() and ggplot()
hist(WPTD2$Distance,
xlab = 'Perpendicular Distance',
ylab = 'Number of Observations',
main = 'Sampling 1-4 Frequency of Observation per Perpendicular Distance (WP)',
col = 'skyblue',
breaks = 20)
and
ggplot(WPTD2, aes(x=Distance)) +
geom_histogram(binwidth = 5,boundary=0)
And this was no problem to do so but i can't seem to figure out how to plot only the data related to the specific Sampling day.
I have tried seperating the days into a factor with four levels with
WPTD2$Sampling <- as.factor(WPTD2$Date)
which resulted in: Result Date -> Sampling
I also tried altering the dataset with:
WPTD2$Sampling <- ave(WPTD2$Date, WPTD2$`Transect nr.`, FUN = seq_along)
but this wielded no usable result.
using filter()
function you can subset day and pass the filtered data to ggplot()
:
library(tidyverse)
WPTD2 %>% filter(Date==44915) %>% ggplot(aes(x=Distance)) +
geom_histogram(binwidth = 5,boundary=0)
If the Date is factor, then try to place the Date in the quotes inside filter()
function:
WPTD2 %>% filter(Date=="44915") %>% ggplot(aes(x=Distance)) +
geom_histogram(binwidth = 5,boundary=0)