Subsetting data based on a value within ids in r - r

Question

I'm trying to subset a dataset based on two criteria. Here is a snapshot of my data:
ids <- c(1,1,1,1,1,1, 2,2,2,2,2,2, 3,3,3,3,3,3)
seq <- c(1,2,3,4,5,6, 1,2,3,4,5,6, 1,2,3,4,5,6)
type <- c(1,1,5,1,1,1, 1,1,1,8,1,1, 1,1,1,1,1,1)
data <- data.frame(ids, seq, type)
ids seq type
1 1 1 1
2 1 2 1
3 1 3 5
4 1 4 1
5 1 5 1
6 1 6 1
7 2 1 1
8 2 2 1
9 2 3 1
10 2 4 8
11 2 5 1
12 2 6 1
13 3 1 1
14 3 2 1
15 3 3 1
16 3 4 1
17 3 5 1
18 3 6 1
ids is the student id, seq is the sequence of the questions (items) students take. type refers to the type of the question. 1 is simple, 5 or 8 is the complicated items. What I would like to do is to generate 1st variable(complex) as to whether or not student has a complicated item(type=5|8). Then I would like to get:
> data
ids seq type complex
1 1 1 1 1
2 1 2 1 1
3 1 3 5 1
4 1 4 1 1
5 1 5 1 1
6 1 6 1 1
7 2 1 1 1
8 2 2 1 1
9 2 3 1 1
10 2 4 8 1
11 2 5 1 1
12 2 6 1 1
13 3 1 1 0
14 3 2 1 0
15 3 3 1 0
16 3 4 1 0
17 3 5 1 0
18 3 6 1 0
The second step is to split data within students.
(a) For the student who has non-complex items (complex=0), I would like to split the dataset from half point and get this below:
>simple.split.1
ids seq type complex
13 3 1 1 0
14 3 2 1 0
15 3 3 1 0
>simple.split.2
ids seq type complex
16 3 4 1 0
17 3 5 1 0
18 3 6 1 0
(b) for the students who have complex items (complex=1), I would like to set the complex item as a cutting point and split the data from there. So the data should look like this (excluding complex item):
>complex.split.1
ids seq type complex
1 1 1 1 1
2 1 2 1 1
7 2 1 1 1
8 2 2 1 1
9 2 3 1 1
>complex.split.2
ids seq type complex
4 1 4 1 1
5 1 5 1 1
6 1 6 1 1
11 2 5 1 1
12 2 6 1 1
Any thoughts?
Thanks

Answer 1

Here's a way to do it using data.table, zoo packages and split function:
library(data.table)
library(zoo)
setDT(data)[, complex := ifelse(type == 5 | type == 8, 1, NA_integer_), by = ids][, complex := na.locf(na.locf(complex, na.rm=FALSE), na.rm=FALSE, fromLast=TRUE), by = ids][, complex := ifelse(is.na(complex), 0, complex)] ## set data to data.table & add a flag 1 where type is 5 or 8 ## carry forward and backward of complex flag ## replace na values in complex column with 0
data <- data[!(type == 5 | type == 8), ] ## removing rows where type equals 5 or 8
complex <- split(data, data$complex) ## split data based on complex flag
complex_0 <- as.data.frame(complex$`0`) ## saving as data frame based on complex flag
complex_1 <- as.data.frame(complex$`1`)
split(complex_0, cut(complex_0$seq, 2)) ## split into equal parts
split(complex_1, cut(complex_1$seq, 2))
#$`(0.995,3.5]`
# ids seq type complex
#1 3 1 1 0
#2 3 2 1 0
#3 3 3 1 0
#$`(3.5,6]`
# ids seq type complex
#4 3 4 1 0
#5 3 5 1 0
#6 3 6 1 0
#$`(0.995,3.5]`
# ids seq type complex
#1 1 1 1 1
#2 1 2 1 1
#6 2 1 1 1
#7 2 2 1 1
#8 2 3 1 1
#$`(3.5,6]`
# ids seq type complex
#3 1 4 1 1
#4 1 5 1 1
#5 1 6 1 1
#9 2 5 1 1
#10 2 6 1 1

Answer 2

If you prefer using the tidyverse, here's an approach:
ids <- c(1,1,1,1,1,1, 2,2,2,2,2,2, 3,3,3,3,3,3)
seq <- c(1,2,3,4,5,6, 1,2,3,4,5,6, 1,2,3,4,5,6)
type <- c(1,1,5,1,1,1, 1,1,1,8,1,1, 1,1,1,1,1,1)
data <- data.frame(ids, seq, type)
step1.data <- data %>%
group_by(ids) %>%
mutate(complex = ifelse(any(type %in% c(5,8)), 1, 0)) %>%
ungroup()
simple.split.1 <- step1.data %>%
filter(complex == 0) %>%
group_by(ids) %>%
filter(seq <= mean(seq)) %>% #if you happen to have more than 6 questions in seq, this gives the midpoint
ungroup()
simple.split.2 <- step1.data %>%
filter(complex == 0) %>%
group_by(ids) %>%
filter(seq > mean(seq)) %>%
ungroup()
complex.split.1 <- step1.data %>%
filter(complex == 1) %>%
arrange(ids, seq) %>%
group_by(ids) %>%
filter(seq < min(seq[type %in% c(5,8)])) %>%
ungroup()
complex.split.2 <- step1.data %>%
filter(complex == 1) %>%
arrange(ids, seq) %>%
group_by(ids) %>%
filter(seq > min(seq[type %in% c(5,8)])) %>%
ungroup()