问题
I have a big data set which could be represented something like this:
plot 1 2 3 3 3 4 4 5 5 5 5 6 7
fate S M S S M S S S M S S M M
where plot is a location, and fate is either "survivorship" or "mortality" ( a plant lives or dies.) The plot number of a plant corresponds to the fate under it. Thus in plot 5 there are 4 plants. 3 of them survive, 1 dies.
I want to figure out a way to make R calculate the fraction of individuals that survive in each plot for all of these. It is proving very challenging.
Example: Plot 5 would return a survivorship value of 3/4 or 75% Plot 3 would return a survivorship value of 2/3 or 66%
Any help would be much appreciated. Thank you
Data
structure(list(plot = c(1, 2, 3, 3, 3, 4, 4, 5, 5, 5, 5, 6, 7
), fate = structure(c(2L, 1L, 2L, 2L, 1L, 2L, 2L, 2L, 1L, 2L,
2L, 1L, 1L), .Label = c("M", "S"), class = "factor")), .Names = c("plot",
"fate"), row.names = c(NA, -13L), class = "data.frame")
回答1:
Here is one solution with dplyr; I've created valu column with 1 if survived and 0 if not. After that it is only a matter of sum the 1's and divide them by the total number of elements of plot.
library(dplyr)
df %>% group_by(plot) %>%
mutate(valu = ifelse(fate == "S", 1, 0)) %>%
mutate(perce = (sum(valu)/n() )*100 )
Source: local data frame [13 x 4]
Groups: plot
plot fate valu perce
1 1 S 1 100.00000
2 2 M 0 0.00000
3 3 S 1 66.66667
4 3 S 1 66.66667
5 3 M 0 66.66667
6 4 S 1 100.00000
7 4 S 1 100.00000
8 5 S 1 75.00000
9 5 M 0 75.00000
10 5 S 1 75.00000
11 5 S 1 75.00000
12 6 M 0 0.00000
13 7 M 0 0.00000
回答2:
There are many possibilities. Here's one:
plot <- c(1,2,3,3,3,4,4,5,5,5,5,6,7)
fate <- as.factor(c("S","M","S","S","M","S","S","S","M","S","S","M","M"))
sapply(lapply(split(fate, plot), function(x) round(prop.table(table(x))*100, 2 )), "[", "S")
# 1.S 2.S 3.S 4.S 5.S 6.S 7.S
# 100.00 0.00 66.67 100.00 75.00 0.00 0.00
来源：https://stackoverflow.com/questions/31439306/how-to-make-an-average-of-a-variable-assigned-to-individuals-within-a-category