Find value changes across rows in R-CodePudding

I have a data frame which looks like the following:

dat <- data.frame(Target = c(rep("01", times = 8), rep("02", times = 4)),
                  targ1clicks = c(1, 1, 1, 1, 0, 0 ,0 , 1, 0, 0, 0, 1))

    Target targ1clicks
1      01           1
2      01           1
3      01           1
4      01           1
5      01           0
6      01           0
7      01           0
8      01           1
9      02           0
10     02           0
11     02           0
12     02           1

What I want to find is how many times the value in the targ1clicks column changes between 1 and 0. So for Target 01, it should be 2; for Target 02 it should be 1. I know there are some similar posts on this which helped me achieve the code I already have (below). The issue is my code is not generating the correct values and I can't figure out why.

Here is the code I have thus far:

clickRows <- which(dat$targ1clicks != dplyr::lag(dat$targ1clicks)) #find the row numbers 
 #where there is a change                            

dat2 <- dat[c(clickRows),] #filter by these row numbers and save as new df

dat2 <- dat2 %>% 
  group_by(Target) %>% 
  filter(targ1clicks == 1) #filter just on the '1' values

no.accClicks <- as.data.frame(table(dat2$Target)) #This should show the number 
 #of value changes for each target

The trouble is, this code doesn't give me the correct value for Target 01 - it gives me the following:

   Var1 Freq
1   01    1
2   02    1

The Freq column should indicate the number of value changes, so it should have 2 for the 01 target, and I can't figure out why this isn't working correctly! Any insight into why this isn't working or any alternatives on how to code this would be hugely appreciated.

CodePudding user response：

You were on the right track. Use lag to compare with previous values. This will give a vector of TRUE/FALSE (TRUE meaning value changed). Then take a sum of this vector for each group.

library(dplyr)
dat %>% 
    group_by(Target) %>% 
    summarise(changes = sum(targ1clicks != lag(targ1clicks), na.rm = TRUE))

CodePudding user response：

Try this

dat |> group_by(Target) |> 
summarise(Freq = sum(abs(diff(targ1clicks))))

# A tibble: 2 × 2
  Target  Freq
  <chr>  <dbl>
1 01         2
2 02         1