When talking about histograms, why do we refer to the x-axis as "bin"?
$\begingroup$
$\endgroup$
2
-
3$\begingroup$ In vernacular English, a "bin" is a container for holding things. On the horizontal axis, we have a number bins, each of which holds data points that are sufficiently similar. We are essentially sorting the data into box. $\endgroup$– Xander HendersonApr 12, 2018 at 16:51
-
1$\begingroup$ google.com/… $\endgroup$– whuber ♦Apr 12, 2018 at 21:57
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
2
When wanting to create a histogram of a continuous variable, you first need to split those into bins (sometimes referred to as buckets). Subsequently, this procedure is called binning or bucketing.
So the x-axis of a histogram represents the bins of the continuous variable.
-
2$\begingroup$ You could also bin discrete/count data, in whole or in part. For example, number of sexual partners often has a long right tail that researchers top code at X+. Histogram construction is largely a tradeoff between detail and generalization (or variance and bias), and this tradeoff does not depend on the type of data. $\endgroup$– dimitriyApr 12, 2018 at 22:33
-
$\begingroup$ Yes you are correct, binning works on discrete data as well, however I wanted to point out that you can't have a histogram from a continuous variable without binning it first. $\endgroup$– Djib2011Apr 12, 2018 at 22:36