Tableau Visualizations and Group Project Overview

Tableau Public Installation and Setup

To begin working with visualizations, Tableau Public Desktop needs to be installed. It is important to:

  1. Sign up for Tableau Public using an email address (preferably the student email).

  2. Download the desktop edition after signing up.

Tableau Public is free, but it requires saving work to the web instead of locally on the machine. This is the only difference from the version used in the OGGB computer labs.

Group Project

Forming Groups

Students need to form groups of two for the group project. This can be done via the "People" tab under "Project Groups".

Project Overview

The project involves:

  1. Selecting one data set from a provided list of six public data sets (accessible via Kaggle). Each data set pertains to a different area or topic, such as Spotify, video games, or Netflix.

  2. Formulating five questions related to the chosen data set.

  3. Using Tableau to answer these questions by extracting hidden information and visualizing it.

  4. Creating a website to showcase the data set, questions, answers, and their significance.

Data Sets

The provided data sets are public and mostly unmodified, with links to their original sources on Kaggle. A Kaggle account may be required to access detailed information.

Question Formulation

Questions should be well-thought-out and not too simple. They should demonstrate the use of Tableau's features and functionalities. Students can discuss questions with the instructor or tutors before finalizing them. Tutors are experienced students who have previously excelled in the course.

Website Theme

The website should have a theme that provides context to the analysis. For example:

  • If using the video games data set, the theme could be understanding the competitive environment for creating and marketing a new game.

  • If using the Netflix data set, the theme could be promoting a new TV show, examining comparable shows in the domain.

Website Creation with Wix

Wix, a user-friendly website builder, can be used to create the website. The website should apply visual design principles taught in week 10.

Project Components

The project has four parts:

  1. Data set selection.

  2. Question formulation.

  3. Answering questions using Tableau.

  4. Creating a website to demonstrate questions and answers.

Introduction to Visualizations and Business Intelligence

The course is transitioning to visualizations to understand how to generate insights from data (business intelligence). The focus will be on Tableau this week and Power BI next week.

Business Intelligence and Storytelling

Business intelligence involves understanding and extracting information from data. Telling a story with data is crucial, as the analysis part is becoming increasingly automated. Humans need to interpret findings and create narratives.

Tableau and Power BI

Tableau and Power BI are key skills for business intelligence roles. The course aims to provide a basic understanding of both tools.

Tableau Interface and Data Connection

After opening Tableau Public Desktop:

  1. Connect to Microsoft Excel.

  2. Select the provided "Superstore" data set.

Tableau can import multiple related data tables (Excel or CSV files) and is generally intuitive at detecting connections between them.

Data Types

Tableau automatically recognizes data types (numeric, text, dates) and assigns icons to each field.

  • Numeric values are recognized as numeric.

  • Text values are labeled with "ABC".

  • Dates have a calendar icon.

Data types can be manually adjusted if Tableau makes a mistake.

Interface Overview

  • Worksheets: Primary area for creating visualizations.

  • Dashboards: Combine multiple worksheets into a single view.

  • Stories: Not covered in detail in this course.

Dimensions and Measures

Data is divided into two categories:

  • Dimensions (blue): Qualitative, categorical data that provides context.

  • Measures (green): Quantitative data used for calculations.

Creating Basic Visualizations

Pills (data fields) can be dragged into the columns and rows sections to generate visualizations.

Aggregation

By default, Tableau aggregates measures (e.g., using "Sum"). This aggregation can be changed by clicking the down arrow on the pill and selecting "Measure (Sum)", then choosing different aggregates.

Sorting and Filtering

Visualizations can be sorted and filtered like spreadsheet data. Filters can be added by dragging dimensions to the filters card, then clicking the down arrow and selecting "Show Filter".

Show Me Tab

The "Show Me" tab suggests appropriate visualization types based on the selected dimensions and measures. This allows users to experiment with different visualizations.

Building Visualizations

The following steps and visualizations that can be constructed within Tableau are:

  1. Showing Sales for Each Subcategory Broken Down by Order Date (Year) vs. Category

    • Use color to show the sum of profit.

  2. Showing Sum of Sales for Each Category and Subcategory Broken Down by Order Date (Month)

    • Use size to indicate the sum of profit for each month.

    • Label only the minimum and maximum values.

  3. Showing Sum of Profit for Each Category and Subcategory Broken Down by Order Date (Year) for Each Customer Segment

    • Use color to indicate the customer segment.

Enhanced Analysis Through Sorting and Filtering

Sorting and filtering can refine visualizations and find specific data points.

  • Sorting can be done by clicking the column or axis and selecting the sort direction (ascending or descending).

  • Filters can be applied to focus on specific regions, categories, or other dimensions.

Highlighting Patterns and Trends

Highlighting allows focusing on specific subsets of data within a visualization.

  1. Create a line chart with order date (in months) on the x-axis and sales on the y-axis.

  2. Add profit to the tooltip to show it when hovering over data points.

  3. Use region for color and subcategory for detail.

  4. Show the subcategory highlighter to filter data.

Data Aggregation and Calculations

Quick Table Calculations

Quick table calculations can be applied by right-clicking a measure and selecting "Quick Table Calculation". These calculations include running totals, ranks, and percentages of total.

Calculated Fields

Calculated fields allow creating new measures using formulas or functions.

  1. Create a calculated field (right-click in the measures area and select "Create Calculated Field").

  2. Write a formula using existing measures and operators (e.g., Sales divided by Quantity to create "Average Unit Cost").

  3. The calculation is validated to check for accuracy via Tableau.

Once created, these fields can be used in visualizations.

Formatting Calculated Fields

Calculated fields can be formatted (e.g., as currency) by right-clicking, selecting "Default Properties," and then "Number Format."

Example

  • Create a field to find the unit price to shipping cost ratio.

  • Create a field to find the number of distinct orders.

Sales Spotlight

Sales Spotlight is a report which highlights when sales are not doing as well as expected (below 10,000).

  1. Create a calculated field called "Sales Spotlight" with an if condition.

  2. IF Sum(Sales) > 10,000 THEN "Acceptable" ELSE "Poor" END

  3. Display segment in columns, category and subcategory in rows, and sales as text in marks card.

  4. Add sales spotlight to the color mark to highlight the sales numbers.

Grouping Members

Grouping creates new dimension with group members.

  1. Select the required office supplies (e.g., envelopes, rubber bands, labels).

  2. Right-click and group.

Creating a Dashboard

A dashboard combines multiple visualizations into a single view for the website.

  1. Create a new dashboard.

  2. Drag existing worksheets onto the dashboard.

  3. Filters and legends remain interactive within the dashboard. The dashboard formatting can be managed in various ways.

  4. Add text or objects by dragging them onto the dashboard.

Saving Work

To save to file, "Save to Tableau Public" must be used. You will be prompted to sign in using your email address, then proceed to save to the web page. Then your file will be accessed as a webpage TWBX. This method is used with Tableau Public Desktop. In licensed versions, the Tableau Workbooks can be saved.