Home Forums From Data to Dashboards Summer 2021 Discussion 4 (DTDS21)

Viewing 11 reply threads
  • Author
    Posts
    • #8798
      Liz Lewis
      Moderator
      • Have you encountered many/any box-and-whisker plots in the course of your work? On other dashboards?

      • What about histograms?

      • If you have little or no prior experience with them, do you think they are easy to understand? If not, can you think of some alternate ways to display distribution data?

    • #9012
      Veronica Topp
      Participant

      I’ve only encountered box-and-whisker plots in stats or other quant classes that describe them but I’ve rarely seen them used outside of that. I’ve seen histograms used more frequently and I would say they are a better, if not the best, way to represent distribution data because you can quickly and easily interpret them even if you’re not familiar with histograms in general. Box-and-whisker plots on the other hand require remembering what each aspect means and so if you are not used to interpreting them, it can be difficult to understand what they are trying to show.

    • #9013
      Kate Hamdan
      Participant

      I also have mostly seen box-and-whisker plots in my coursework and in certain situations where audiences had more of a statistical background. I have seen histograms more frequently, particularly in publicly facing documents and dashboards as they are much easier to interpret at a glance. I think each of them can be easy to understand if presented with clear labelling and context.

    • #9018
      Scott Troppy
      Participant

      Who doesn’t enjoy a good box and whisker plot ….if explained clearly. I have seen box-and-whisker plots in certain statistical analysis/papers (and back in the day at school…). I think more folks are used to histograms in papers and dashboards. They are easier to use if described with legends and labels.

    • #9020
      Karen Murphy
      Participant

      My exposure to box-and-whisker plots is extremely limited. I would not have known how to interpret the data from them before this lesson.
      I have seen histograms, but I have not had occasion to create them. I hope to be able to use them in my work going forward.
      In my work, I have not had to display information beyond aggregate summaries. There has been very limited reason to display the information in dashboard formats. However, I am taking this class in order to show information to my boss and others how my department is doing, and to better represent information that can lead to better decisions.

    • #9021
      Emily Leonard
      Participant

      I’ve experienced histograms far more frequently than box and whisker plots. When explained well, box and whisker plots can be highly effective ways of conveying information. However, I find that histograms are more intuitive and effective for public-facing work.

    • #9028
      Lydia Merritt
      Participant

      I have never seen a histogram used in a presentation or explained in my 20+ year career! Nor have I seen anyone present with a histogram. However, I just think I have never been in a field that required use of either of the charts. I am certain if I spoke with our SPSS instructor, I could find a couple in his box of tools, though.

    • #9031
      Reed Sherrill
      Participant

      I don’t often encounter box & whisker plots at work (or in dashboards related to my work), although I have seen them in some research literature. Histograms, on the other hand, are common both at work and in dashboards related to my work.

      While box & whisker plots are uncommon in my day-to-day work, I’m familiar with them generally. Once you understand what each component represents, they become a powerful tool for consolidating several descriptive statistics into a single plot.

      For a public audience, histograms are far easier to interpret when communicating distributions. I would generally steer clear of box & whisker plots for non-technical audiences.

    • #9054
      Marilyn Romero
      Participant

      I have only encountered these in statistic courses and when I did research. They are easy to understand and use. The challenge is using them where it makes the most sense to show a distribution like the data set we are using. At my job I rarely see these utilized and that maybe lack of understanding of when to use them/ why they would be useful for utilization management.

    • #9068
      Elizabeth Mello
      Participant

      I have mostly encountered box and whisker plots in research articles, not often in dashboards. I don’t think they are easy to understand and to communicate data to a public audience. It requires the audience to have an understanding of several statistical terms and know what their interpretation are. I think they usually require more interpretation than the concise data summaries that are usually communicated in dashboards.

      However, I frequently encounter histograms. Epi curve, the graph of disease onset during an outbreak, is usually a histogram. They can work well for communicating counts over time.

    • #9192
      Sonnie Davies
      Participant

      I’ve encountered a few box and whisker plots only when meeting with statisticians, but I’ve seen them kind of hit a blank wall when presented to non-statisticians on the team. Histograms are more commonly presented in my experience to show a basic distribution of the data. But even then, histograms aren’t immediately recognizable in the way that my non-statistician PI’s usually want to see.

      Since I joined a team of social scientists who have little background in quantitative data, I’ve started to present what little quant data we have in relation to other data. Presenting just values of data don’t seem to mean much, at least not to my team members. Since box and whisker plots and histograms are extremely useful for me to translate data though, they’ll probably be common on my analysis pages. But not the final dashboard.

      The one exception I can think of would be a histogram (which is easier to understand immediately than box and whiskers) with a very explicit title describing what piece of data they’re supposed to get from the plot.

    • #9220
      Molly McKenna
      Participant

      I remember seeing a few box and whisker plots in grad school but I do not encounter them very often now. I have used and see histograms more often. I personally think histograms are easier to understand for most people but I would consider the target audience when picking out one of these graphs. I might choose a box and whisker plot if I was presenting to a group of statisticians because they might want to see the interquartile range and would be familiar with this type of graph. For most other audiences I would probably stick with a histogram because it is a simple graph and is easier to understand.

Viewing 11 reply threads
  • You must be logged in to reply to this topic.