Histogram Grouping

Go to the WEBSITE LINK on your browser copy and paste it

http://www.shodor.org/interactivate/activities/Histogram/?version=1.5.0_16&browser=safari&vendor=Apple_Inc.&flash=10.0.22

On the website go to the graph

Look at how different groupings in histograms on SAT scores can lead to different conclusions

Put in interval values of 10, 20, and 50

And Then

Answer these 2 questions

1: Which do you think is most appropriate and why?

2: Would you choose different parameters depending on whether you wanted to make a point about scores being relatively uniform, or that higher scoring students are relatively unusual

some examples answers could be like this

the graph provided uses interval values of roughly 42. This allows for the data to be displayed on the graph showing all the scores, from lowest to highest. If the interval value is changed to ten, the graph becomes much more complex. Rather than grouping scores in categories displayed by 10 bars, as is the case with an interval value of 42, entering 10 as the interval value now displays over 30 bars.

As the interval increases, the distribution becomes less right skewed and more uniform. The smallest interval is most appropriate due to it most accurately showcasing the skewedness of the data. Larger intervals can hide skewedness and showcase a fake uniform distribution. When you use larger intervals, evidence for relatively uniform distribution is more easily shown. In another sense, to show that higher scoring students are relatively unusual, a smaller interval forces the high scoring students to look like outliers rather than being grouped into a larger group from larger interval numbers. I

The initial bar graph was at 42 but when I switched the interval to 10, I noticed first that the minimum and maximum frequency changed but the graph was more detailed meaning it illustrated many smaller bars and the data is spread out. At the 20 interval, the bars are medium sized which I assume the bars are put into slightly bigger groups. At the interval 50, I happen to notice the bars are a lot bigger which possibly indicates the overall data is combined into larger groups. I found the interval 10 to be narrowed down meaning a lot of factors are considered into this data because the amount of bars provided.

The interactive tool sheds light on how the interpretation of data may be affected by various interval values in histograms. The graph becomes more intricate when interval values of 10 are used, with narrower bars showing a dispersed distribution of scores. Although examining individual data points may benefit from this degree of information, it might also make the graph seem visually crowded. Interval values of 50, on the other hand, result in wider bars, which represent bigger data groups. This makes the display simpler, but it could also hide subtleties in the data.