# Why do histograms have bars?

Watch
Announcements
#1
Hi,

On one of my university assignments, a histogram was given in the form of a line graph, rather than using bars. The question asked why histograms should be presented as bars rather than as a line graph, but I'm not sure exactly how I would phrase an answer to that. I've only ever seen histograms presented with bars and I've always just accepted that's the way things are, so I'm not sure how to justify any reasoning.

Any help would be greatly appreciated!
0
1 year ago
#2
(Original post by Ash2810)
Hi,

On one of my university assignments, a histogram was given in the form of a line graph, rather than using bars. The question asked why histograms should be presented as bars rather than as a line graph, but I'm not sure exactly how I would phrase an answer to that. I've only ever seen histograms presented with bars and I've always just accepted that's the way things are, so I'm not sure how to justify any reasoning.

Any help would be greatly appreciated!
You should read up about the origins of histograms, but sometimes bars are more appropriate than a line.
It has the advantage of simplicity, more accurate and visibly more effective.

For instance if the data I am collecting is on an annual basis, then each bar can represent a year. As the data is for a whole year then there is no difference between beginning and end just the data being within that year i.e 1977.
A line wouldn't really work because you would be relying on any pint of the line to be reflective.
1
1 year ago
#3
(Original post by Ash2810)
Hi,

On one of my university assignments, a histogram was given in the form of a line graph, rather than using bars. The question asked why histograms should be presented as bars rather than as a line graph, but I'm not sure exactly how I would phrase an answer to that. I've only ever seen histograms presented with bars and I've always just accepted that's the way things are, so I'm not sure how to justify any reasoning.

Any help would be greatly appreciated!
A histogram is an approximation to the density function (when normalized to unity) of a continuous random variable. So one of the major principles of constructing a histogram is that the area of a bar represents the relative probability (or frequency) of the observations falling within the bin that defines the bar. Let's go back to when histograms were invented: no computers, no fancy graphics! Pencil, ruler and paper. Getting the area accurate for each bar is much easier with these tools if you use rectangular bars!

Of course, now we do have the fancy graphics capabilities, and can represent relative probabilities, or relative frequencies in many diverse and attractive ways. However, the histogram is still attractive, as it provides a good approximation (as the number of bins increases, and the area of each bin gets smaller) to the kernel density estimate of the underlying probability distribution. When you start going to this limit, then the importance of using bars for a histogram somewhat fades, as you're aiming to get a curve as the density estimate.
1
X

new posts
Back
to top
Latest
My Feed

### Oops, nobody has postedin the last few hours.

Why not re-start the conversation?

see more

### See more of what you like onThe Student Room

You can personalise what you see on TSR. Tell us a little about yourself to get started.

### Poll

Join the discussion

#### Would you give consent for uni's to contact your parent/trusted person in a mental health crisis?

Yes - my parent/carer (67)
33.5%
Yes - a trusted person (54)
27%
No (52)
26%
I'm not sure (27)
13.5%