7 Replies Latest reply on Jul 20, 2016 6:26 AM by Michael Hesser

    #MakeoverMonday Week 29 - The Next to Die

    Nicholas Hura

      Hi all, I'm new to the forums and Tableau in general. I was just hoping to get some brutally honest feedback on my most recent viz to see where I can improve. I appreciate the community being very nice and well-receptive, but you always have to keep yourself grounded and try to get better. Any and all critiques are welcome.





        • 1. Re: #MakeoverMonday Week 29 - The Next to Die
          Michael Hesser

          Hi Nick!

          First-- big kudos for a striking presentation! It looks sharp and it certainly engaged me. Very inspiring! Please don't think by my critiques that I didn't enjoy it. Here goes!


          • While your title is certainly catchy, I'm not sure it matches your data. It seems you're reporting on how many people are executed by state/county/type, rather than how many are awaiting execution. Your title (to me) suggests the latter, rather than the former. I'm uncertain what a more fitting title would be... maybe simply changing it into a question? What about "How We Kill Our Prisoners" ?
            I realize The Next to Die is the name of a larger project... so maybe a subtitle explaining what the user is seeing?
            As a footnote, it would be interesting to be able to see how many are waiting on Death Row... but that's just me.
          • Executions remains pluralized when there is only 1 (see CT). I told you I was picky!
          • I really like your "blocky state map"-- a cool way of representing all the states in roughly their correct locations. I'm a Kansas boy, so I know Colorado is directly to the West, Missouri to the East, and so on (so the map is a little hinky). Overall, it's still pretty effective.
          • It proved very difficult for me to hover on southernmost states (FL and HI). The active area for these seems about 1/2 as sensitive as others. Has anyone else had this problem?
          • There were some other problems with the state hover: when there are no executions, neither the County Breakdown or Yearly Executions vizes are affected... it seems to me like graying them all out would be best, rather than showing all values.
          • Clicking a state causes the County Breakdown to default to the state (with no border)... I'm not sure how effective this is... you really need to know the geography of the state, I guess.
          • Similarly, I was unable to access the all county data. After clicking on LA I noticed the upper-left diamond corresponded with two counties. No matter how I hovered, I could only see the data from one (Bossier Parish). How likely is someone going to notice this? Very, very unlikely.
          • I really like the diamonds on the County Breakdown (hearkening back to the squares on the state view). Nice touch.
          • It was difficult for me to use the Yearly Execution view, as I first tried to use it by just cruising around (with little luck). It seems to work best when you hover on the state to see the aggregations, but I had to figure this out for myself. How is the best way to utilize that view? Personally, I needed some instructions
          • I'd like a brief note as to why you chose TX, OK, and VA as a group in the Yearly Execution view (I know they are the top 3, but I think a small note, even if it is on hover, would be nice)
          • Call me a stickler, but I think you may want to rename your "Methods" viz to "Methods of Execution." Why shy away from the fact that people are being put to death by the state? You might want to include Executions in every title...
          • I really like your little execution graphics. The gas mask is cute (did I say "cute" referring to a viz about death?!?) however it's not technically representative of happening
          • Hanging and Firing Squad have the same tally (for USA), yet are not on the same level. I know you're trying to maintain your slope in the viz, but a user might incorrectly assume more people are executed by firing squad than the noose...
          • The active link to The Marshall Project was great.
          • Lastly (and I think the most important): your viz is too long. This interferes with the way I could interact with it: it was difficult to scroll over states and see how it affects Yearly Executions, for example, and impossible to see how it affected Methods (unless I clicked). You have a lot of great graphics here; if you're not able to scrunch them all into one happy screen, perhaps you could look into pop-up windows?
            If you find-- after collapsing these-- you have a little extra real estate, maybe you can add a cold-but-neutral description, such as the dictionary definition of "execution"?


          THANK YOU for sharing your inspiring viz! --Michael

          1 of 1 people found this helpful
          • 2. Re: #MakeoverMonday Week 29 - The Next to Die
            Nicholas Hura

            Hi Michael,


            Thanks for the critique! This is exactly what I was hoping for as I've probably developed a few bad habits because people didn't call me out on them. Nice to know there are still some people out there with a very good eye for detail.


            • I never really put too much thought into the title and how much of an importance it really is. Go figure my most successful viz has the titled posed as a question with a short description as well.
            • I noticed the 'executions' afterwards, I'll make sure to use 'execution(s)' in the future to avoid this annoyance.
            • The blocky-map definitely is a little bit off, but I like to use them in situations like this where the geographical size and shape of the state isn't the important. It's the number of executions in each state that matters so the blocks give each state the same size and shape so the focus more on the shear numbers. This is why I've found these maps to be extremely situational and would not recommend them for every analysis that uses a map.
            • I'm not sure why the Southern-most states are having issues with hovering and selecting as they worked fine in my workbook, oh Tableau public...
            • In general, it seems that I need to provide more context on each graph which was hard with the limited space (I was trying to go for more of a mobile-friendly dashboard in this case) but I can quickly see where the confusion lies in every case.
            • The hanging and firing squad slope was definitely an issue that could have been easily avoided, partially some laziness on my part. There always that rush to finish when you create the last graph for the dashboard.
            • Pop-up windows seem like a great alternative to this dashboard as I agree it did become too long with all the actions. I wasn't planning on it being quite this long so I think I lost touch with how effective the overall user experience would be with universal actions.


            Overall, I think this might be an example of how my focus on graphic design exceeded the actual analysis and left out pretty basic information that would help move this viz along smoothly. I'll be sure to incorporate your advice into my next viz and keep building off of the successful parts of this one.


            Thank you again!



            1 of 1 people found this helpful
            • 3. Re: #MakeoverMonday Week 29 - The Next to Die
              Nicholas Hura

              By the way...


              I found that the issue with Florida and Hawaii was that the floating divider I had was 'covering' the most of those two states even though the divider has a transparent background.

              • 5. Re: #MakeoverMonday Week 29 - The Next to Die
                Michael Hesser

                I *really* liked your "squap" (square + map ).


                I don't work much with maps in Tableau, but is this a regular map choice? Maybe you can point me in the direction where you found this?


                Glad I could be of help! --Michael

                • 6. Re: #MakeoverMonday Week 29 - The Next to Die
                  Nicholas Hura

                  I first became interested in the 'squap' from Matt Chambers IronViz entry here: Workbook: red vs blue-Republican & Democrat Voting


                  Basically you will blend from an Excel spreadsheet on the State field. Here is the sheet for the setup Matt used in his Viz: Square Map Positions - Google Sheets


                  You may also need to reverse the Y-Axis if things do not look right the first time you drag the Row and Column fields onto the shelf.


                  Also, if you do ever want to play around with maps in Tableau, I highly recommend joining the #makeovermonday crew on Twitter. Andy Kriebel and Andy Contgreave run it and we remake visualizations from a different source every week. It's a fun time to experiment and try things, both visualizations and data sources, that you might not typically use. It's been a great learning experience for me, and I'd recommend it to anyone. You can also use the datasets from pass weeks as examples to practice with as well because we all know how hard it is to find a good dataset sometimes. You can read more about it here: Makeover Monday Challenge



                  • 7. Re: #MakeoverMonday Week 29 - The Next to Die
                    Michael Hesser

                    Wonderful and Thank You! As a curmudgeon who prefers sitting on my porch and yelling at the kids to stay off my lawn, maybe this is "Twitter-thing" is just what I need to finally drag myself into the 21st century!