My data looks like this:
| hh_id | indl | ind_salary | hh_income |
|---|---|---|---|
| 1 | 1 | 200 | |
| 1 | 2 | 450 | |
| 1 | 3 | 00 | |
| 2 | 4 | 1232 | |
| 2 | 5 | 423 |
Individuals with the same hh_id lives in the same household so they will have the same household income. And for that the variable hh_income equal the sum of the salary of all persons with the same hh_id;
so my data would look like:
| hh_id | indl | ind_salary | hh_income |
|---|---|---|---|
| 1 | 1 | 200 | 650 |
| 1 | 2 | 450 | 650 |
| 1 | 3 | 00 | 650 |
| 2 | 4 | 1232 | 1655 |
| 2 | 5 | 423 | 1655 |
Any ideas please;
Using dplyr:
data %>% group_by(hh_id) %>% mutate(hh_income = sum(ind_salary))
You can use R base function ave to generate sum of ind_salary grouped by hh_id and get a vector of the same length of ind_salary
> df$hh_income <- ave(df$ind_salary, df$hh_id, FUN=sum)
> df
hh_id indl ind_salary hh_income
1 1 1 200 650
2 1 2 450 650
3 1 3 0 650
4 2 4 1232 1655
5 2 5 423 1655
Using only base R:
hh_id <- c(1, 1 ,1, 2, 2)
indl <- c(1, 2, 3, 4, 5)
ind_salary <- c(200, 450, 0, 1232, 423)
hh_df <- data.frame(hh_id, indl, ind_salary)
hh_income <- tapply(hh_df$ind_salary, hh_df$hh_id, sum)
hh_income <- as.data.frame(hh_income)
hh_income$hh_id <- rownames(hh_income)
hh_df <- merge(hh_df, hh_income, by = 'hh_id')
View(hh_df)
Just to add more explanation to KacZdr's answer which would have helped me immensely as a beginner. Also, this is more in line with standard tidyr pipe code standards.
new_data <- data %>% # This creates a new dataset from the original so you don't alter the original, I find this much easier
group_by(hh_id)%>% # obviously groups the data by the variable that has duplicate values within the column that you want to apply a summary function , in this case sum
mutate(income = sum(ind_salary))# mutate creates a new column "income" and fills it with the sum of ind_salary for all with the same hh_id. This would be what you have called hh_income in your table.
?mutate.