GGRAPH/GPL vs. ggplot2

[Jignesh Sutar]

I've been trying to visualize more and more of my work recently and am finding many benefits in doing so.

Given the recent post from Albert-Jan which demo'd using ggplot2 the creation this:

http://spssx-discussion.1045642.n5.nabble.com/Overlaying-graphs-tp5727813p5727817.html





and a multitude of great examples from Andy here:

https://andrewpwheeler.wordpress.com/category/spss/


I'm wondering which graphic tool I should invest my time learning. My R knowledge is very limited so of course I am biased to perhaps take the easy option and learn GGRAPH/GPL rather than ggplot2 but is there more in the way of functionality and customization that at a more advance level ggplot2 can provide over ggraph/gpl and worth the investment to switch?

I am considering purchasing either one of these books to help aid the learning curve? Are there other books others would recommend?

The Grammar of Graphic

Elegant Graphics for Data Analysis
 Use R!


Many thanks in advance,
Jignesh

[Andy W]

Wilkinson's book is a tough read. I was already pretty familiar with SPSS's GGRAPH simply through practical experience before reading it, and while informative, I would not recommend it as an introduction to the material. Wilkinson's book is more theoretically orientated than a programming tutorial.

Hadley's book is a great, practical introduction. (I debated at one point of making SPSS replications for several visualization books, all of the graphs [save one I think] in Hadley's book can be recreated in SPSS.)

The graph builder in SPSS can write much of the code for you, so that helps quite a bit in learning the code. If you do most of the work in SPSS I see no inherent reason to learn only ggplot, but they are both largely exchangeable means to the same end. (R in general for everything is more flexible, but the vast majority of graphs you will ever want to make can be made in SPSS.)

More generally, learning the workflow in code in one program helps you think about it in another. So if you  learn one pretty well that knowledge can be more readily adapted to the other, despite the syntax looking so different.

[Jon K Peck]

I would not recommend Lee's book as a way to learn GPL.  It might help you think about what graphics would be appropriate in a given context, however.  And although the grammar of graphics is what lies behind GPL, the book is by  no means a manual for it.  Hint, BTW, "gg" stands for grammar of graphics.

There is also the new ggvis package which has moved beyond ggplot2.  (Not to mention lattice and a zillion other graphics packages in R.)


Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
phone: [phone]




From:        Jignesh Sutar <[hidden email]>
To:        [hidden email]
Date:        11/06/2014 10:07 AM
Subject:        [SPSSX-L] GGRAPH/GPS vs. ggplot2
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

I've been trying to visualize more and more of my work recently and am
finding many benefits in doing so.

Given the recent post from Albert-Jan which demo'd using ggplot2 the
creation this:

http://spssx-discussion.1045642.n5.nabble.com/Overlaying-graphs-tp5727813p5727817.html
<http://spssx-discussion.1045642.n5.nabble.com/Overlaying-graphs-tp5727813p5727817.html>  

<http://spssx-discussion.1045642.n5.nabble.com/file/n5727826/RGraph.png>



and a multitude of great examples from Andy here:

https://andrewpwheeler.wordpress.com/category/spss/
<https://andrewpwheeler.wordpress.com/category/spss/>  


I'm wondering which graphic tool I should invest my time learning. My R
knowledge is very limited so of course I am biased to perhaps take the easy
option and learn GGRAPH/GPL rather than ggplot2 but is there more in the way
of functionality and customization that at a more advance level ggplot2 can
provide over ggraph/gpl and worth the investment to switch?

I am considering purchasing either one of these books to help aid the
learning curve? Are there other books others would recommend?

The Grammar of Graphic
<http://www.springer.com/statistics/computational+statistics/book/978-0-387-24544-7>  

Elegant Graphics for Data Analysis
Use R!
<http://www.springer.com/statistics/computational+statistics/book/978-0-387-98140-6>  


Many thanks in advance,
Jignesh



--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/GGRAPH-GPS-vs-ggplot2-tp5727826.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD


===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

[Jignesh Sutar]

Thanks Andy. Do you know if the source data and the code is available for examples in Hadleys' book?

I could find any source data for either books?

[Andy W]

Hadley has code snippets right in the book and I'm pretty sure the data come loaded with the ggplot2 package. Even if if the text snippets aren't available online he does a good job of taking it slow and in small steps, so typing the commands as you follow along is not that bad.

The only downside for the book now is that some of the code is not applicable, particularly things like the global chart aesthetics and templates. But he has a quite nice website to help along with the newer code.

[Albert-Jan Roskam-2]

 
Learning R is a good strategic choice (it's popular language and a skill you can put on your CV), and ggplot2 is  a nice starting point. In a few lines of code you get actual, visual results, which is rewarding. One small downside of ggplot is that the parametrization might differ slightly between versions (e.g. linetype expects an integer in one version, and a string (like 'dash') in the other. I only want to learn ggplot (preferably the 'long' versions, not qplot) and the builtin graphs, not trellis, lattice and all the others. Ok, maybe also maptools :-)

Regards,

Albert-Jan




~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

All right, but apart from the sanitation, the medicine, education, wine, public order, irrigation, roads, a

fresh water system, and public health, what have the Romans ever done for us?

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~




----- Original Message -----

> From: Jignesh Sutar <[hidden email]>
> To: [hidden email]
> Cc:
> Sent: Thursday, November 6, 2014 6:04 PM
> Subject: [SPSSX-L] GGRAPH/GPS vs. ggplot2
>
> I've been trying to visualize more and more of my work recently and am
> finding many benefits in doing so.
>
> Given the recent post from Albert-Jan which demo'd using ggplot2 the
> creation this:
>
> http://spssx-discussion.1045642.n5.nabble.com/Overlaying-graphs-tp5727813p5727817.html
> <http://spssx-discussion.1045642.n5.nabble.com/Overlaying-graphs-tp5727813p5727817.html>  
>
>
> <http://spssx-discussion.1045642.n5.nabble.com/file/n5727826/RGraph.png>
>
>
>
> and a multitude of great examples from Andy here:
>
> https://andrewpwheeler.wordpress.com/category/spss/
> <https://andrewpwheeler.wordpress.com/category/spss/>  
>
>
> I'm wondering which graphic tool I should invest my time learning. My R
> knowledge is very limited so of course I am biased to perhaps take the easy
> option and learn GGRAPH/GPL rather than ggplot2 but is there more in the way
> of functionality and customization that at a more advance level ggplot2 can
> provide over ggraph/gpl and worth the investment to switch?
>
> I am considering purchasing either one of these books to help aid the
> learning curve? Are there other books others would recommend?
>
> The Grammar of Graphic
> <http://www.springer.com/statistics/computational+statistics/book/978-0-387-24544-7>  
>
>
> Elegant Graphics for Data Analysis
> Use R!
> <http://www.springer.com/statistics/computational+statistics/book/978-0-387-98140-6>  
>
>
>
> Many thanks in advance,
> Jignesh
>
>
>
> --
> View this message in context:
> http://spssx-discussion.1045642.n5.nabble.com/GGRAPH-GPS-vs-ggplot2-tp5727826.html
> Sent from the SPSSX Discussion mailing list archive at Nabble.com.
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> [hidden email] (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD
>  

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD