the plot thickens

Any thoughts on the new Scatterplot banner above versus the old one? I’ve been bothered that the blue in the old plot didn’t match the blue of the rest of the theme, and then while I was at it I also decided to try using smaller-sized points.

People have asked and, yes, the graphic is an actual scatterplot generated using Stata, although the data are random (and uniformly distributed on both the abscissa and ordinate.) Note the relatively large spaces of white adjacent to places where the points are densely bunched on top of one another. That’s what true randomness really looks like, and the reason people are bad when they try to fake random numbers (or coin flips, or financial transactions, or scatterplot points) is that they think randomness should be more evenly spread.

Secret bonus for those nice enough to follow the jump: Regarding my use of “abscissa” and “ordinate” above, I’ve seen a couple talks in other disciplines in the past couple years where researchers have used these terms instead of the more familiar and ordinary-sounding “x-axis/y-axis” (or just “horizontal” and “vertical”).  As far as academic affectations to make a speaker sound more posh, I can affirm that this did, indeed, work to make the speakers sound more posh. So, shhh, we should also start using these terms in our sociology talks and it can be a special Scatterplot inside-joke academic affectation, as inside-joke affectations are really the only kind that are worthwhile.

Author: jeremy

I am the Ethel and John Lindgren Professor of Sociology and a Faculty Fellow in the Institute for Policy Research at Northwestern University.

13 thoughts on “the plot thickens”

  1. I like the new one.
    I did my my favourite assignment in undergrad about how people suck at making up random data and how that made it easier for codes to be broken in world war two.

    Like

  2. I will admit, I thought about putting in a giant hole of white space on the left side and writing the same little didactic thing about randomness, just to screw with people’s heads.

    Like

  3. Anyone who plays Settlers of Catan realizes the unfortunate reality of randomness.

    The blue didn’t bother me before, but I like how, with the smaller points, “scatterplot” stands out more.

    Like

  4. i may be the only one who likes the old one better. i like how in the old one you can easily see the different shapes and colors of the data points. but the new one does make the ‘scatterplot’ title stand out more.

    Like

  5. Sewenz: Thanks!

    yli: I’m sort of torn, but let’s go with the new one for awhile.

    Tina: I need to look at the CSS to figure out what the width needs to be. The new one is the same width as the old one. I’ll look into fixing whenever the wind moves me to redraw the plot.

    Like

  6. I like the former one better for the reason yli cited, that you can actually tell what those little things are. This way, it just looks like a mix of little circles, as if we were looking at tiny organisms moving around in a Petri dish. Of course, that has its own charm if you’re going for that kind of an effect.;)

    Isn’t the title itself coming through better because you increased its size?

    Like

  7. I like it, but related to blog mainainence issues, I think you need to install pingomatic or some such, because scatterplot is very rarely updated on my technorati faves list. I forget to visit (perish the thought!).

    Like

  8. Really, Kieran? It works for most of my favorited blogs (about 85% of the 40 or so) – I just always assumed the others weren’t pinging — my error.

    Like

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.