How to loop with two lists in R - r

Question

I have a dataset with demographic information and with questions.
DF<-(Participant = c(1,2,3,4,5,6,7,8,9,10)
Male = c(1,0,1,1,0,1,0,0,1,0)
Female = c(0,1,0,0,1,0,1,1,0,1)
Q1 = c(9,6,5,4,5,1,3,5,5,2)
Q2 = c(2,4,5,4,2,1,3,5,4,2)
Q3 = c(6,8,2,7,5,2,1,1,6,3))
I have two lists (made from column titles), one of demographic information (Males, Females, age group etc) and one of questions with their associated response.
Demographic <- c(“Male”, “Female”, “Age_group_1”, “Age_group_2”…)
Questions<- c(“Q1”, “Q2”, Q3”, “Q4”…)
I need something along the lines of- if value in demographic column is equal to 1 then sum scores in all separate question columns. But I want to do this is a loop so I have the separate question scores (~300) for all columns in the demographic list (~80). Plus I want to save the output. I have no idea how to do this and I’m getting into a loop of bad programming myself!
The end result should resemble this:
M F
Q1 20 21
Q2 16 16
Q3 23 18
I would be grateful for any help!
Thanks in advance.
UPDATE:
With help from a friend, I have found a work around my problem. How do you make this more efficient though?
df.list <- list()
for(question in questions){
question.df <- (DF[, lapply(.SD,sum, na.rm=T), by=question,
.SDcols=c(demographic)])
df.list <- append(df.list, question.df)}
list_new <- bind_cols(df.list, .id = "column_label")

Answer 1

library(tidyr)
library(dplyr)
df <- data.frame(
Participant = c(1,2,3,4,5,6,7,8,9,10),
Male = c(1,0,1,1,0,1,0,0,1,0),
Female = c(0,1,0,0,1,0,1,1,0,1),
Q1 = c(9,6,5,4,5,1,3,5,5,2),
Q2 = c(2,4,5,4,2,1,3,5,4,2),
Q3 = c(6,8,2,7,5,2,1,1,6,3)
)
df %>%
mutate(sex = ifelse(Male == 1, "M", "F")) %>%
select(-Male, -Female) %>%
pivot_longer(cols = starts_with("Q"), names_to = "Q") %>%
group_by(sex, Q) %>%
summarise(value = sum(value)) %>%
pivot_wider(names_from = sex)
gives:
Q F M
<chr> <dbl> <dbl>
1 Q1 21 24
2 Q2 16 16
3 Q3 18 23

Answer 2

Depending on what you want to do with the output, another approach is to use tables::tabular(), which can be used to generate additional statistics (e.g. percentages), as well as customizing row and column headings.
We'll generate a simple table using the data provided in the question.
df <- data.frame(Participant = c(1,2,3,4,5,6,7,8,9,10),
Male = c(1,0,1,1,0,1,0,0,1,0),
Female = c(0,1,0,0,1,0,1,1,0,1),
Q1 = c(9,6,5,4,5,1,3,5,5,2),
Q2 = c(2,4,5,4,2,1,3,5,4,2),
Q3 = c(6,8,2,7,5,2,1,1,6,3))
df$sex <- ifelse(df$Male == 1,"M","F")
library(tables)
tabular((Q1 + Q2 + Q3)~Factor(sex)*(sum),data=df)
...and the output:
> tabular((Q1 + Q2 + Q3)~Factor(sex)*(sum),data=df)
sex
F M
sum sum
Q1 21 24
Q2 16 16
Q3 18 23
Processing multiple demographic variables
In the comments to my answer a question was asked about how to use tabular() with more than one demographic variable.
We can use a combination of lapply(), paste(), and substitute() to build the correct formula expressions for `tabular().
To illustrate the process we will add a second demographic variable, Income to the data frame listed above. Then we create a vector to represent the list of demographic variables for which we will generate tables. Finally, we use the vector with lapply() to produce the tables.
df <- data.frame(Participant = c(1,2,3,4,5,6,7,8,9,10),
Male = c(1,0,1,1,0,1,0,0,1,0),
Female = c(0,1,0,0,1,0,1,1,0,1),
Income = c(rep("low",5),rep("high",5)),
Q1 = c(9,6,5,4,5,1,3,5,5,2),
Q2 = c(2,4,5,4,2,1,3,5,4,2),
Q3 = c(6,8,2,7,5,2,1,1,6,3))
df$Sex <- ifelse(df$Male == 1,"M","F")
library(tables)
tabular((Q1 + Q2 + Q3)~Factor(Sex)*(sum),data=df)
demoVars <- c("Sex","Income")
lapply(demoVars,function(x){
# generate a formula expression including the column variable
# and use substitute() to render it correctly within tabular()
theExpr <- paste0("(Q1 + Q2 + Q3) ~ Factor(",x,")*(sum)")
tabular(substitute(theExpr),data=df)
})
...and the output:
> lapply(demoVars,function(x){
+ # generate a formula expression including the column variable
+ # and use substitute() to render it correctly within tabular()
+ theExpr <- paste0("(Q1 + Q2 + Q3) ~ Factor(",x,")*(sum)")
+ tabular(substitute(theExpr),data=df)
+ })
[[1]]
Sex
F M
sum sum
Q1 21 24
Q2 16 16
Q3 18 23
[[2]]
Income
high low
sum sum
Q1 16 29
Q2 15 17
Q3 13 28
Note that we can enhance the solution further by saving the tables to an output object and rendering them in a printer friendly format as needed.