Forum Replies Created
-
Posts
-
I used box and whisker plots frequently in undergrad. I find them really helpful in understanding a lot of data at one time in a visually engaging way. I never really know when to use a histogram or how to glean sufficient information from it, while a box and whisker plot allows me to easily compare means, medians, quartiles, and standard deviations across multiple data sets. It is really helpful for me in understanding overall differences in datasets.
I found the same outcomes in rankings for Northern and Eastern as the above comments (most frequent being Curative Care in both zones; least frequent in Eastern being Inpatient Obstetrics and Inpatient medical; least frequent in Northern being speech therapy). I agree with Elizabeth that the vertical bar graph would be the most useful display – by having the service category on the y axis and having the data organized from most frequent to least frequent, the graph visually aligns with our word association of “highest” or “most” being on top and “lowest” or “least” being on bottom.
Which type of graph do you think is best for showing Flores del Mundo’s membership trends over the course of the year?
A line graph would be the best option for showing membership trends over time as we are more interested in seeing an overall shape of the total year as opposed to individual months. This would allow us to see if there are any particularly interesting changes between months, as well (such as a spike or drop in membership).
Does your answer change if you’re graphing total membership versus employee/spouse/dependent?
A line graph could also be useful here with different colors representing each category, if the goal is to again visualize the whole year as opposed to smaller scale. However, I would argue that a bar graph might be better here in certain instances. For example, the person viewing the dashboard may be interested in seeing if there are any particular times of year where major changes occur in addition or removal of household members from the plan, as that could give some information on planning ahead for seasonal fluctuations.
Does your answer change if you want to show Eastern and Northern on the same graph?
I think in this case, a bar graph would be most visually useful to indicate what portion of memberships come from each location. If each bar is organized by Total/Location1/Location2, we can see the annual trends of each but more importantly we can see if there are patterns of membership at different locations (e.g., does one location tend to host more memberships than the other?). A line graph would make it harder to visualize the part-to-whole relationship.
The example shared in the WHO article meets the definition of a dashboard per Few’s article, consolidating the most important information selected by the hospital and color coded for easy recognition of health trends requiring urgent intervention.
However, I am curious whether all the data is actually laid out on one page given that the title reads “Maternity Dashboard (front page)”. I wonder whether any key information is missing on this page. Additionally, as mentioned in the answer sheet for the Graph Design IQ test, having bright colored backgrounds and a table highlighting all of the information makes the data potentially harder to read, especially if anyone needing the information experiences red/green colorblindness. It might be more legible and therefore more effective to only highlight the data that requires the attention of the viewers – in this case, numbers indicating poor outcomes.
