Introduction
For this week’s challenge, we wanted to hit two birds with one stone: try out a new feature from the Tableau 2020.4 and celebrate Black History Month with an interesting dataset on Historical Black Colleges and Universities (HBCU).
Predictions, table calculations, and parameter actions…oh my! Consider this a feature sampler, building on some previous challenges and adding a predictive element into the mix.
Viz Responsibly: While predictive analysis in Tableau has never been easier, we encourage you to be intentional in how you use this powerful feature because not doing so (e.g., selecting a model type not appropriate for your data) could lead to inaccurate results. Click here, to learn more about predictive modeling in Tableau.
Requirements
Requirements:
- Create a view that shows the actuals (1993-2018) and predicted values (through 2023) for the measure selected by the user
- Measures: Total Enrollment (All Students), % of Black Students Enrolled, % of Non-Black Students Enrolled
- Dashboard Size: 1200 X 600
- Number of Sheets: 2
- Color
- Actuals Area chart: #00a2b3, opacity: 75%
- Prediction Values Line Chart: #5c6068
- Selection Options: #00a2b3
- No drop-down menus or hard-coded years in calculations
- Match tooltips and number formatting
Additional Information:
- Predictive model type used: Gaussian process regression “model=gp”
- Prediction Residual = Actual Value-Predicted Value
Dataset
This week uses U.S. Dept. of Education National Center for Education Statistics. You can get it here at data.world
Attribute
When you publish your solution on Tableau Public make sure to take the time and include a link to the original inspiration. Also include the hashtag #WOW2021 in your description to make it searchable!
Share
After you finish your workout, share on Twitter using the hashtag #WOW2021 and tag @AnnUJackson, @ItsCandraM, @LukeStanke, @_Lorna_Brown and @HipsterVizNinja
Solution
Interactive
Hi, I will admit that I am still learning how to use Tableau, I love the WOW because it really pushes my understanding and teaches me a bunch of new skills every week. This week you have really stumped me. When I downloaded the work book a number of the calculated fields were broken (Red exclamation points) which I have tried to fix (mostly by replacing renamed calculations) but there is one that I cant work out how to fix so I was wondering if you could cast some light? its the calculated field called “Prediction: Gaussian Prediction” which reads //predicts future median values via Gaussian regression model
MODEL_QUANTILE(“model=gp”,0.5, [Select Measure], attr(datetrunc(‘year’,[Year 1])))
under “model+gp”,0.5, there is a red line that says “Argument 1 to MODEL_QUANTILE must be a(n) float literal.
does this mean that I need to update something to be a float?
Hi.. I would like to download the workbook, in order to see the solution, but I couldn’t, It seems the workbook is corrupted.