This note outlines the design of dashboards using Tableau. The data set is to be used in this case study. The visualization with be created using Tableau.
We first load the working data to R and perform a simple exploratory data analysis and then decide what specific visualizations will be created.
The description of the data can be found at: https://github.com/pengdsci/sta553/raw/main/dash/mushroom-description.pdf
The data set can be found at: https://github.com/pengdsci/sta553/raw/main/dash/mushroom-data.csv
## [1] "class" "cap.diameter" "cap.shape"
## [4] "cap.surface" "cap.color" "does.bruise.or.bleed"
## [7] "gill.attachment" "gill.spacing" "gill.color"
## [10] "stem.height" "stem.width" "stem.root"
## [13] "stem.surface" "stem.color" "veil.type"
## [16] "veil.color" "has.ring" "ring.type"
## [19] "spore.print.color" "habitat" "season"
Three numerical variables are summarized in the following.
## cap.diameter stem.height stem.width
## Min. : 0.380 Min. : 0.000 Min. : 0.00
## 1st Qu.: 3.480 1st Qu.: 4.640 1st Qu.: 5.21
## Median : 5.860 Median : 5.950 Median : 10.19
## Mean : 6.734 Mean : 6.582 Mean : 12.15
## 3rd Qu.: 8.540 3rd Qu.: 7.740 3rd Qu.: 16.57
## Max. :62.340 Max. :33.920 Max. :103.91
## [1] "class" "cap.shape" "cap.surface"
## [4] "cap.color" "does.bruise.or.bleed" "gill.attachment"
## [7] "gill.spacing" "gill.color" "stem.root"
## [10] "stem.surface" "stem.color" "veil.type"
## [13] "veil.color" "has.ring" "ring.type"
## [16] "spore.print.color" "habitat" "season"
## $class
##
## e p
## 27181 33888
##
## $cap.shape
##
## b c f o p s x
## 5694 1815 13404 3460 2598 7164 26934
##
## $cap.surface
##
## d e g h i k l s t w y
## 14120 4432 2584 4724 4974 2225 2303 1412 7608 8196 2150 6341
##
## $cap.color
##
## b e g k l n o p r u w y
## 1230 4035 4420 1279 828 24218 3656 1703 1782 1709 7666 8543
##
## $does.bruise.or.bleed
##
## f t
## 50479 10590
##
## $gill.attachment
##
## a d e f p s x
## 9884 12698 10247 5648 3530 6001 5648 7413
##
## $gill.spacing
##
## c d f
## 25063 24710 7766 3530
##
## $gill.color
##
## b e f g k n o p r u w y
## 954 1066 3530 4118 2375 9645 2909 5983 1399 1023 18521 9546
##
## $stem.root
##
## b c f r s
## 51538 3177 706 1059 1412 3177
##
## $stem.surface
##
## f g h i k s t y
## 38124 1059 1765 535 4396 1581 6025 2644 4940
##
## $stem.color
##
## b e f g k l n o p r u w y
## 173 2050 1059 2626 837 226 18063 2187 1025 542 1490 22926 7865
##
## $veil.type
##
## u
## 57892 3177
##
## $veil.color
##
## e k n u w y
## 53656 181 353 525 353 5474 527
##
## $has.ring
##
## f t
## 45890 15179
##
## $ring.type
##
## e f g l m p r z
## 2471 2435 48361 1240 1427 353 1265 1399 2118
##
## $spore.print.color
##
## g k n p r u w
## 54715 353 2118 1059 1259 171 182 1212
##
## $habitat
##
## d g h l m p u w
## 44209 7943 2001 3168 2920 360 115 353
##
## $season
##
## a s u w
## 30177 2727 22898 5267
The above frequency table indicates that several categorical variables have a significantly high percentage of missing values. Since we only perform visual analytics to illustrate how to use Tableau to create dashboards, we will not perform any data management for modeling purposes.
We briefly introduced the basic statistics charts using Tableau. In this note, we choose both categorical and quantitative variables in the working data set to construct individual charts with Tableau and then use these charts to construct a dashboard.
We will construct five descriptive charts: a two-way contingency table, donuts chart (a variation pie chart), scatter plot, box plots, and histogram.
We will use four individual charts to construct a dashboard that includes a reactive filter to update all charts in the dashboard.
Tableau can create a form presentation of the existing individual chart so we can tell the story based on the Tableau charts.
This assignment focuses on Tableau dashboards and story points. You can choose a data set that has both numerical and categorical variables so you can create several charts such as histograms, scatter plots, frequency tables, pie charts, etc. using both types of variables.
Choose a data set with at least two numerical variables so you can construct a histogram and scatter plots.
The data set should have at least three categorical variables so you can create pie charts, donuts charts, and box plots.
You should use at least one filter that updates all charts in the dashboard (reactive dashboard).
Use appropriate tiles and axis labels for each chart and a meaningful title for the dashboard.
Adjust the margins of the individual chart and make the dashboard tidy and informative.
Create a story point and provide informative captions for storytelling.
(optional) if you have geo-information in your data set, you can create maps including animated maps.
(optional) if you want to use more advanced topics such as derived variables and parameters to enhance your dashboard, you are very welcome to do so.
Publish your dashboard and story point on the Tableau Public server.
Write a simple plain HTML OR RMD document to include both Tableau dashboard and story-point on the same web page.
You have two options to prepare a single document to publish on either RPubs or Github. As usual, you need to submit the RMD source to the D2L dropbox if you use RMD to generate the web.
Option 1: HTML Template and Publish It on GitHub - Embedding Tableau Charts on Web
The HTML template of the assignment: https://raw.githubusercontent.com/pengdsci/sta553/main/dash/index-html-template
After finishing writing the HTML, you need to save it as index.html
in a local folder and upload it to the root directory, say, STA553
.
Configuration: settings ==> pages ==> main ==> root. The URL will then be something like
https://pengdsci.github.io/sta553/.
Option 2: Writing Narrative Using RMD and Publish HTML on RPubs
Write a brief explanation (mini-report) in RMD and simply include the links to your Tableau charts in the RMD.
Publish your viz report on RPubs.
Submit the web link and the RMD source to D2L.