Assign value of 1 per each group in dataset

In the dataset provided below, we can see that I have ID repeat three times for three possible alternatives meaning that a respondent has to choose 1 of 3 alternatives. The last column chosen contains a dummy variable which has 0 in each row as a default value.

Problem: I do not understand how can I randomly assign a value of 1 (indicating that first, second or third alternative was chosen) for each group. For example, when ID == 1, value 1 has to be randomly assigned either to first, second or third row and so on for the rast of the data.

Here's what I've tried:

for (i in seq(1, nrow(sim_data), 3)) {  # loop through each group of three rows
  chosen_index <- sample(i:(i+2), 1)  # generate a random index within the group
  sim_data$Chosen[chosen_index] <- 1  # assign 1 to the chosen index
}

Since my data has 300 rows and the cycle goes up to 298, it didn't work out.

Dataset:

attributes <- expand.grid(
  Company = c("Metalac", "NikolaTeslaAirport", "Jedinstvo", "Energoprojekt"),
  Return_rate = c(0, 0.05, 0.10, 0.15),
  Dividend = c(0, 1.5, 3.0, 4.5, 6),
  Trend = c("Trend1", "Trend2", "Trend3")
)

# Generate simulated data for 100 respondents
set.seed(123) # for reproducibility
sim_data <- data.frame(
  ID = rep(1:100, each = 3), # three alternatives per respondent
  alternative = rep(1:3, times = 100), # alternative number
  attributes[sample(nrow(attributes), size = 100 * 3, replace = TRUE), ],
  chosen = 0
)

I'm sure there are more elegant ways, but one dplyr solution would be to use slice_sample() to randomly sample 1 row per group, assign it a value of 1, then join the sampled data frame back with the full data frame (and then I do some clean up sorting back to the original and dropping temp variables).

sim_data %>% 
  slice_sample(n = 1, by = ID) %>%
  mutate(temp = TRUE) %>%
  full_join(sim_data) %>% 
  mutate(chosen = case_when(temp ~ 1, TRUE ~ chosen)) %>%
  arrange(ID, alternative) %>% select(-temp)

# or with older dplyr versions

sim_data %>% 
  group_by(ID) %>%
  slice_sample(n = 1) %>%
  mutate(temp = TRUE) %>%
  full_join(sim_data) %>% 
  mutate(chosen = case_when(temp ~ 1, TRUE ~ chosen)) %>%
  arrange(ID, alternative) %>% select(-temp)

Output

  ID alternative            Company Return_rate Dividend  Trend chosen
1  1           1          Jedinstvo        0.15      6.0 Trend2      1
2  1           2          Jedinstvo        0.15      3.0 Trend3      0
3  1           3          Jedinstvo        0.00      1.5 Trend3      0
4  2           1 NikolaTeslaAirport        0.15      0.0 Trend1      0
5  2           2          Jedinstvo        0.00      3.0 Trend3      1
6  2           3 NikolaTeslaAirport        0.10      0.0 Trend3      0
7  3           1 NikolaTeslaAirport        0.00      4.5 Trend1      1
8  3           2 NikolaTeslaAirport        0.05      3.0 Trend2      0
9  3           3          Jedinstvo        0.10      3.0 Trend1      0
# .....