# [R] Calculating column differences

William Michels wjm1 @end|ng |rom c@@@co|umb|@@edu
Wed Mar 24 19:40:46 CET 2021

```More correctly, with an initial "NA" value in the "diff" column:

> df <- data.frame(ID=1:5,Score=4*2:6)
> df1 <- rbind(c(0,0), df)
> cbind(df1, "diff"=c(NA, diff(df1\$Score)) )
ID Score diff
1  0     0   NA
2  1     8    8
3  2    12    4
4  3    16    4
5  4    20    4
6  5    24    4
>

HTH, Bill.
On Wed, Mar 24, 2021 at 10:59 AM William Michels <wjm1 using caa.columbia.edu> wrote:
>
> Dear Jeff,
>
> Rather than diff-ing a linear vector you're trying to diff values from
> two different rows. Also you indicate that you want to place the
> diff-ed value in the 'lower' row of a new column. Try this (note
> insertion of an initial "zero" row):
>
> > df <- data.frame(ID=1:5,Score=4*2:6)
> > df1 <- rbind(c(0,0), df)
> > cbind(df1, "diff"=c(0, diff(df1\$Score)) )
>   ID Score diff
> 1  0     0    0
> 2  1     8    8
> 3  2    12    4
> 4  3    16    4
> 5  4    20    4
> 6  5    24    4
> >
>
> HTH, Bill.
>
> W. Michels, Ph.D.
>
>
>
> On Wed, Mar 24, 2021 at 9:49 AM Jeff Reichman <reichmanj using sbcglobal.net> wrote:
> >
> > r-help forum
> >
> >
> >
> > I'm trying to calculate the diff between two rows and them mutate the
> > difference into a new column. I'm using the diff function but not giving me
> > what I want.
> >
> >
> >
> > df <- data.frame(ID=1:5,Score=4*2:6)
> >
> >
> >
> > What a want  where
> >
> >   ID Score  diff
> >
> > 1  1     8      8
> >
> > 2  2    12     4
> >
> > 3  3    16     4
> >
> > 4  4    20     4
> >
> > 5  5    24     4
> >
> >
> >
> > What I am getting
> >
> >   ID Score  diff
> >
> > 1  1     8      NA
> >
> > 2  2    12     4
> >
> > 3  3    16     4
> >
> > 4  4    20     4
> >
> > 5  5    24     4
> >
> >
> >
> > Jeff
> >
> >
> >
> >
> >         [[alternative HTML version deleted]]
> >
> > ______________________________________________
> > R-help using r-project.org mailing list -- To UNSUBSCRIBE and more, see
> > https://stat.ethz.ch/mailman/listinfo/r-help
> > and provide commented, minimal, self-contained, reproducible code.

```