Removing entire observation with a condition-CodePudding

I have made a survey and would like to remove all answers from people who answered ''no'' to being a parent?

The dataset is called ''tillid''

The variable is called ''Er du forældre'' and the answer is either ''Ja'' or ''Nej''

CodePudding user response：

Question

Welcome to SO. You should read the How do I ask a good question advice shared by @user2974951 to learn how to ask questions in a way that helps the community respond.

From the [r] tag guidance:

Please use minimal reproducible example(s) others can run using copy & paste. Show desired output. Use dput() for data & specify all non-base packages with library(). Don't embed pictures for data or code, use indented code blocks instead.

In this case you have a data.frame that looks something like this:

> Tillid

  Id ... Er du forældre
1  1                 Ja
2  2                Nej
3  3  
4  4                Nej
5  5                 Ja
...

To create a minimal reproducible example use dput on a subset of rows and columns:

> dput(Tillid[1:5, c('Id', 'Er du forældre')])

structure(list(Id = 1:5, `Er du forældre` = c("Ja", "Nej", "",
"Nej", "Ja")), class = "data.frame", row.names = c(NA, -5L))

Anyone can copy this line code and create a dataset that looks like yours.

Next you want to show what you are trying to achieve (based on the minimal example):

> <insert code here>

  Id ... Er du forældre
1  1                 Ja
3  3  
5  5                 Ja

Answer

In this case, the code to remove rows with the value 'Nej' is as follows (you might want to assign this to a new variable using <-)

> Tillid[Tillid$`Er du forældre` != 'Nej', ]

  Id Er du forældre
1  1             Ja
3  3
5  5             Ja

If you also want to exclude missing answers (of which you have 114), you could slice to only those rows with the value "Ja":

> Tillid[Tillid$`Er du forældre` == 'Ja', ]

  Id Er du forældre
1  1             Ja
5  5             Ja

As mentioned by @mat in his answer, it's good practice to avoid special characters, and spaces, in column names.

CodePudding user response：

tillid <- tillid[tillid$`Er du forældre` != "Nej"]

The != means "not equal to".

Alternatively:

tillid <- tillid[tillid$`Er du forældre` == "Ja"]

Note that if you have missing values (NA), the first alternative will preserve them, whereas the second option will exclude everything that is not equal to Ja.

I would suggest to avoid special characters (e.g., æ) in your variable names as this can cause some bugs in R.