CODEWITHSUNDEEP
rtidyr
pitosalas
pitosalas

Reputation: 10862

How to call ggplot in my scenario

Here's my data:

   Assignment final first resub
44       PA07 59.00  48.5  69.5
46       PA07 74.25  72.5  76.0
63       PA07 98.50  98.5    NA
36       PA07 76.25  71.0  81.5
32       PA07 84.50  77.5  91.5

I would like to make a scatter plot with ggplot where the X axis is the value in "first". There are two sets of dots, one set uses the value in final as the y, and the other uses the value in resub in y.

I tried this with the above layout as well as with this layout, and still cannot figure out how to do it. 

     Assignment Stage  Grade
719        PA06 final  78.50
860        PA06 resub     NA
1866       Exam final  53.24
665        PA05 resub     NA
1881       Exam final  99.34
269        PA03 final  66.00
1675       PA10 resub     NA
761        PA06 final  98.50
498        PA04 resub     NA
435        PA04 first 100.00

Continuing based on suggestion from @andrew:

So now my code looks like this, but I think I can refactor it down but not sure how:

gxg = gather(grades, key, Grade, -email)

gxg = separate(gxg, key, c("Assignment", "Stage"))

gxg = spread(gxg, Stage, Grade)

gxg = gather(gxg, key=Stage, value=Grade,-c(first,Assignment, email))

(btw starting str() is:)

'data.frame':   65 obs. of  30 variables:
 $ email       : chr  ...
 $ course.final: num  86.3 68.9 38.1 77.4 90 ...
 $ part.final  : num  78.2 71.5 54.2 98.2 99.8 ...
 $ PA01.final  : num  78 78 86.5 72.5 86.5 79.5 83 72 90 86 ...
 $ PA02.final  : num  88.5 93 93 93 91.5 90 93 93 86 85.5 ...
 $ PA03.final  : num  90 28 81.5 70 77 97 69 83 66 53 ...
 $ PA04.final  : num  88.8 55.2 0 70.2 73 ...
 $ PA04.first  : num  83.5 53.5 0 65 73 87 62.5 88 45 30 ...
 $ PA04.resub  : num  94 57 0 75.5 NA 94 83.5 NA NA 41.5 ...
 $ PA05.final  : num  96.5 81 0 87.8 96.5 ...
 $ PA05.first  : num  96.5 81 0 79 96.5 81 96.5 93.5 95 50.5 ...
 $ PA05.resub  : num  NA NA NA 96.5 NA 98.5 NA NA NA NA ...
 $ PA06.final  : num  98.5 82.5 0 78.5 95 98.5 95 95 100 83 ...
 $ PA06.first  : num  98.5 82.5 0 78.5 95 98.5 95 95 100 83 ...
 $ PA06.resub  : num  NA NA NA NA NA NA NA NA NA NA ...
 $ PA07.final  : num  97 63.5 0 63.5 86.2 ...
 $ PA07.first  : num  97 63.5 0 60 74 84.5 67 81 91.5 88 ...
 $ PA07.resub  : num  NA NA NA 67 98.5 95 NA 98.5 95 NA ...
 $ PA08.final  : num  100 95 0 97.5 95 97.5 90 95 97.5 95 ...
 $ PA08.first  : num  100 95 0 97.5 95 97.5 90 95 97.5 95 ...
 $ PA08.resub  : num  NA NA NA NA NA NA NA NA NA NA ...
 $ PA09.final  : num  98.5 33 0 69 64 98.5 74.5 81.5 100 93 ...
 $ PA09.first  : num  98.5 33 0 69 81.5 98.5 74.5 67.5 100 93 ...
 $ PA09.resub  : num  NA 33 NA NA 46.5 NA NA 95.5 NA 93 ...
 $ PA10.final  : num  90 90 NA 98 96 100 90 87 93 97 ...
 $ PA10.first  : num  90 90 NA 98 96 100 90 87 93 97 ...
 $ PA10.resub  : logi  NA NA NA NA NA NA ...
 $ PA11.final  : num  45.2 71.5 NA 34.5 91 ...
 $ EC.final    : num  78 15 20 26 10 155 30 90 47 30 ...
 $ Exam.final  : num  66.4 57.2 NA 63.3 87.5 ...

Upvotes: 0

Views: 36

Answers (1)

Andrew Gustar
Andrew Gustar

Reputation: 18425

The following line in tidyr will convert your first df to the correct long form. You can then send this to ggplot

df2 <- df %>% gather(key=Stage,value=Grade,-c(first,Assignment))

ggplot(df2,aes(x=first,y=Grade,colour=Stage)) + geom_point()

enter image description here

In response to follow-up, the following might work...

gxg <- grades %>% 
       gather(key=key,value=Grade,-email) %>% #convert to long format
       separate(key,c(Assignment,Stage)) %>% #separate assignment and stage
       spread(key=Stage,value=Grade) %>% #convert Stage back to wide format (as it includes first as well as final and resub)
       gather(key=Stage,value=Grade,c(final,resub)) #convert final and resub to long format leaving first intact

Hopefully this will result in gxg having columns email, Assignment, first, Stage, Grade

...then ggplot(gxg...) as before.

Upvotes: 1

Related Questions