[BlindRUG] newbie question about seeing relationships between variables

[Godfrey, Jonathan]

Hello Mike et al.,

The reason these two functions are masked is that the BrailleR functions were written so that the objects created are given a class and the information about the axis labels etc. are retained; this doesn't happen with the standard functionality. The masking functions do use the standard function to do the plotting though so the outcome is the same as anyone else would get.

The benefit of this approach is that you can wrap the call to the boxplot() or hist() commands with a call to VI() and get a text description of the graph's content.

I've started out with these two plot types because they are among the first ones a student would need. It will be difficult to do this for the scatter plotting functions so there is a WhereXY() function you might want to look into.

I think your idea about subsetting the possible predictor and using a summary of the response variable is sound. You might create this using the existing boxplot(y~x) construct once you make x using cut().

Cheers,
Jonathan

 





-----Original Message-----
From: BlindRUG [mailto:blindrug-bounces at nfbnet.org] On Behalf Of Mike Gorse via BlindRUG
Sent: Sunday, 8 March 2015 12:00 p.m.
To: blindrug at nfbnet.org
Subject: [BlindRUG] newbie question about seeing relationships between variables

Hi all,

First of all, thanks for the list and for the work to make R easier to use! It's good that some people write statistics software and work to make it more accessible, so that other people can maybe learn to do things like analyze realtime bus data, so that other people can get around the city more easily...

I'm very new to R and to statistics in general, so maybe there's an easy answer to this that I don't know about. I'm starting a Coursera course on statistics and data analysis, and the course had a lab that involved investigating some survey data. In a few places, it advised generating a plot in order to see the relationship between two variables, but I'm blind, meaning that I can't see, meaning that I can't see the nice plots that R generates for me. Eventually I figured out that I could more or less do what I needed to do by subseting the data set based on the explanatory variable, taking summaries of the response variable for the different subsets, and comparing the results, but I'm curious whether this is the best way to do it / what other people would do in order to get an idea of the relationship between two variables in a data set.

Also, when I run
library("BrailleR")
I get a message saying that hist and boxplot are masked from package:graphics, so it looks as though BrailleR has its own implementation/wrapper for boxplot, but, when I call boxplot(), it seems to behave the same as it would if BrailleR weren't loaded (ie, put a boxplot on the screen). Is there something else that it should be doing? I don't see any new files being generated in my working directory (I'm not sure if I should expect them or not).

Thanks,
-Mike

_______________________________________________
BlindRUG mailing list
BlindRUG at nfbnet.org
http://nfbnet.org/mailman/listinfo/blindrug_nfbnet.org
To unsubscribe, change your list options or get your account info for BlindRUG:
http://nfbnet.org/mailman/options/blindrug_nfbnet.org/a.j.godfrey%40massey.ac.nz
The list archive can be viewed at:
http://www.nfbnet.org/pipermail/blindrug_nfbnet.org
More information and useful links about using R as a blind person can be obtained at:
http://R-Resources.massey.ac.nz

Look for help using R commands by reading the accessible e-book "Let's Use R Now" compiled by Jonathan Godfrey at:
http://R-Resources.massey.ac.nz/lurn/front.html