Function using conditions - r

Question

I am trying to create a function that considers a determined number of samples (rivers) each one with a determined number of observations. Given 10 samples each one with 12 observations in a lognormal distribution with mean=4 and sd=1.4, I would like to obtain the number of times a particular number (6 - it refers to a standard number for water quality measurement) is counted.
The following is the code for one experiment, considering "limit" as the maximum number of observations allowed to ovverpass 6.
set.seed(1001)
nobs<-12
limit<-round(0.10 * nobs, digits = 0)
h2o <- as.data.frame(matrix(rnorm(10*12, mean = 4, sd = 1.4), ncol = 12))
paste(rep("Riv", nrow(h2o)), c(1:nrow(h2o)), sep = "")
rownames(h2o) <- paste(rep("Riv", nrow(h2o)), c(1:nrow(h2o)), sep = "")
colnames(h2o) <- paste(rep("Obs", ncol(h2o)), c(1:ncol(h2o)), sep = "")
#Number of rivers declared impared based in the assumptiom that the number of observations per river are 2 or more?
ifelse(h2o >=6,1,0)
h2o$Test<-rowSums(ifelse(h2o >=6,1,0))
length(h2o$Test[h2o$Test>1])
The function should resume the previous data and works for different observations with different samples.
Thanks

Answer 1

Here's a function using dplyr and tidyr from the tidyverse.
library(tidyverse)
test_h2o <- function(data, threshold_quality = 6, limit = 1) {
table <- data %>%
rownames_to_column("river") %>%
gather(observation, value, -river) %>%
mutate(over_lim = value > threshold_quality)
table_wide <- table %>%
select(river, observation, over_lim) %>%
mutate(over_lim = over_lim %>% as.integer()) %>%
spread(observation, over_lim)
summary <- table %>%
group_by(river) %>%
summarize(over_lim_count = sum(over_lim))
result <- summary %>%
summarize(num_impaired = sum(over_lim_count > limit))
list(table_wide, summary, result)
}
Here's the output, meant to show the steps in your example:
> test_h2o(h2o)
[[1]]
river Obs1 Obs10 Obs11 Obs12 Obs2 Obs3 Obs4 Obs5 Obs6 Obs7 Obs8 Obs9
1 Riv1 0 0 0 0 0 0 0 0 0 0 0 0
2 Riv10 0 0 0 0 0 0 0 0 0 0 0 0
3 Riv2 0 0 1 0 0 0 0 0 0 0 0 0
4 Riv3 1 0 0 0 0 0 0 0 0 0 0 0
5 Riv4 0 0 0 0 0 0 0 0 0 0 0 1
6 Riv5 0 1 0 0 0 0 0 0 0 0 0 0
7 Riv6 0 1 1 0 0 1 0 0 0 0 0 0
8 Riv7 0 0 0 0 0 0 0 0 0 0 0 0
9 Riv8 0 0 0 0 0 0 0 1 0 1 0 0
10 Riv9 1 0 0 0 0 0 0 0 0 0 0 0
[[2]]
# A tibble: 10 x 2
river over_lim_count
<chr> <int>
1 Riv1 0
2 Riv10 0
3 Riv2 1
4 Riv3 1
5 Riv4 1
6 Riv5 1
7 Riv6 3
8 Riv7 0
9 Riv8 2
10 Riv9 1
[[3]]
# A tibble: 1 x 1
num_impaired
<int>
1 2