R create variable IF ELSE leads to wrong values - r

Question

I have a dataframe with:
"serial" the number of households, each one with a variable number of components "head, spouse, parent and child or grandchild" and total number of children in the house "nchild"
I want to create a new variable (in the dput I added an example for clarity: withCM 'living with male child' and withCF). I have tried various combinations but I cannot discriminate on the sex of the child within the same "serial", so that for withCM=1 only when relate=="child"&sex==1, but the 1 would appear on a different row (that of the head, spouse or parent)
mydata$withCM<- ifelse(mydata$nchild>0&mydata$relate!="child",1,0)
mydata <- structure(list(serial = c(12345L, 12345L, 12345L, 12345L, 12346L,
12346L, 12347L, 12347L, 12347L, 12348L, 12348L, 12348L, 12348L,
12348L, 12348L, 12348L, 12349L, 12350L, 12350L, 12351L, 12351L,
12351L, 12352L, 12352L, 12352L, 12352L, 12352L, 12353L, 12354L,
12354L), age = c(45L, 44L, 13L, 11L, 29L, 28L, 65L, 61L, 35L,
68L, 61L, 35L, 34L, 6L, 2L, 1L, 62L, 54L, 52L, 67L, 67L, 12L,
49L, 50L, 28L, 21L, 22L, 70L, 89L, 55L), sex = c(1L, 2L, 2L,
1L, 1L, 2L, 1L, 2L, 1L, 1L, 2L, 1L, 2L, 2L, 2L, 1L, 2L, 2L, 1L,
1L, 2L, 1L, 1L, 2L, 1L, 2L, 2L, 2L, 2L, 2L), relate = structure(c(4L,
7L, 1L, 1L, 4L, 7L, 6L, 6L, 4L, 4L, 7L, 1L, 2L, 3L, 3L, 3L, 4L,
4L, 7L, 4L, 7L, 3L, 4L, 7L, 1L, 5L, 5L, 4L, 6L, 4L), .Label = c("child",
"childinlaw", "grandchild", "head", "nonrelative", "parent",
"spouse"), class = "factor"), nchild = c(2L, 2L, 0L, 0L, 0L,
0L, 1L, 1L, 0L, 1L, 1L, 3L, 3L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L,
0L, 1L, 1L, 0L, 0L, 0L, 0L, 1L, 0L), conhija = c(1L, 1L, 0L,
0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 1L, 1L, 0L, 0L, 0L, 0L, 0L, 0L,
0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 0L, 1L, 0L), conhijo = c(1L,
1L, 0L, 0L, 0L, 0L, 1L, 1L, 0L, 1L, 1L, 1L, 1L, 0L, 0L, 0L, 0L,
0L, 0L, 0L, 0L, 0L, 1L, 1L, 0L, 0L, 0L, 0L, 0L, 0L)), .Names = c("serial",
"age", "sex", "relate", "nchild", "conhija", "conhijo"), class = "data.frame", row.names = c(NA,
-30L))

Answer 1

You can tabulate the gender, family, and role-within-family as:
xtab <- table(mydata$serial, mydata$sex, mydata$relate)
And then choose the heads of the families (or, in the commented line, anyone who has the specific relationship), and alter their tallies as follows:
mydata$sex1 <- 0
mydata$sex2 <- 0
ind <- mydata$relate=="head"
#ind <- mydata$relate %in% c("head","spouse","parent")
mydata$sex1[ind] <- xtab[as.character(mydata$serial[ind]), "1", "child"]
mydata$sex2[ind] <- xtab[as.character(mydata$serial[ind]), "2", "child"]

Answer 2

Use lapply to split into families, then test if they are an adult, and there is at least one male child in the unit.
lives_with_boy <- function(serial)
{
unit <- mydata[mydata$serial==serial,]
as.character(unit$relate) %in% c("head","spouse","parent") & any(unit$relate == "child" & unit$sex==1)
}
mydata$withCM <- unlist(lapply(unique(mydata$serial),lives_with_boy ))