On Mon, 11 Nov 2013 18:02:19 +0300
Keniajin Wambui <kiangati at gmail.com> wrote:
> I am using R 3.0.2 on a 64 bit machine
>
> I have a data set from 1989-2002. The data has four variables
> serialno, date, admission ward, temperature and bcg scar.
>
> serialno admin_ward date_admn bcg_scar temp_axilla yr
> 70162 Ward2 11-Oct-89 y 38.9 1989
> 70163 Ward1 11-Oct-91 y 37.2 1991
> 70164 Ward2 11-Oct-92 n 37.3 1992
> 70165 Ward1 11-Oct-93 y 38.9 1993
> 70166 Ward1 11-Oct-94 y 37.7 1994
> 70167 Ward1 11-Oct-95 y 40 1995
>
>
> I want to do a bar graph of total data (serialno) vs *(data of one of
> the variables) to show the available data vs total data over the years
>
> i am using
>
> gplot(dta, aes(temp_axilla, fill=admin_ward)) + geom_bar() +
> facet_grid(. ~ yr, scales = "free",margins=F) +
> geom_histogram(binwidth=300)
>
> But can include the serialno which shows the data. how can I achieve
> this
>
You really need to pay better attention somewhere. There are six
variables in your example table for instance, not four. You state you
want to do something with the serialno variable, but it is not used
in your bit of code. Also, given that serialno appears to be
unique in your example, a bar graph would be singularly uninformative.
You need to provide a better example of the data, and an actual example
of the code you are trying to use, including the libraries you've
loaded.
jwdougherty