MIS 501 Summer Tableau Workshop

Tableau Desktop
• can create workbooks comprising of worksheets, dashboards and stories.
• is a licensed product. Comes in two versions: Desktop Professional and Desktop Personal.
• allows workbooks to be stored locally.
• allows workbooks, dashboards and stories to be published to Tableau Server, Tableau Online and Tableau Public.

Tableau Public

• is a free product from Tableau.
• limits the amount of work with data (number of rows).
• can only connect to Excel, Access or text file and anything you save can downloaded by anyone.

Anyone can sign-up for via the Tableau Public site.
Once you have signed in, click the Create Tab > Web Authoring.

Managing Data Sources

One of the most overlooked aspects of creating visualizations is making sure the data you use is clean.  Depending on where the data originates, this may be easier to identify and correct. 

Numerical Conventions

When dealing with business/economic data, the sources can vary, the reporting authorities can have different standards or the numeric conventions may differ.  For example, if we are looking at a country's GDP or exports we need to know if those figures are in US$, or reporting countries' currency.  Are those values reported as the actual numbers, or are they represented as thousands ($10,000 = $10) or millions ($10,000,000 = $10).

There is no universal standardization and any country, or data vendor can represent their information however they want.

GIGO

The most important aspect of the data you use, is identifying where it came from.  The adage, "Garbage In, Garbage Out" is critical to your research. 

Many data vendors have extensive quality control measures in place to decrease the likelihood of erroneous values, they are no foolproof.  The larger the dataset, the higher the chances of errors. 

Considerations for identifying reliable resources:

Authority:  Who is the publisher of the data?  Government data is good source of information as there are strict criteria for the collection process.  Data that is the source of previously published articles is another great source.  The review process for academic and scientific journals are very rigid and demanding.  Any data that was used for articles must be fully explained and vetted.  Third party data vendors (library databases) are another good source of data.  **However, there will be copyright restrictions on what, where and how the information is used.

Timeliness:  Does the data reflect the period that your visualization will be focusing?  Is the data up-to-date?  Many resources in the library (secondary resources) have restrictions on the data.  We are an academic institution with academic licenses.  That means many of our vendors place an embargo on the data we are able to see.  Make sure you verify the date ranges used in research from any data vendor.

Things to look for

Does data represented match the expected data type

What numeric convention did the source use to display the data

Are there any duplicate values that need to be removed

Is the formatting acceptable, or will you need to alter it to meet your specifications

Tableau has several sections that are displayed as tabs in the lower left corner:

Data Source		This is where you import/manage data from an external source.
Sheet		Allows you to create visualizations from the data you imported.  Use the '+' icon to create as many scenarios as you wish.
Dashboard		Once you have created all your visualizations in Sheets, the Dashboard is the place to compile those into the visualization that you want displayed to the public
Story		This area creates the story you want to tell.  It can be a collection of Sheets, Dashboards, Text, and Icons that emphasize you findings from the Data Source you used.

Data Pane:

Located in the left, Data has two sections: 

Dimensions - The top section of the pane is the categorical items.  These can include non-numeric items such as Gender, Dates, Geographical etc.

Measures - The numeric items such as Sales, Employees, Prices and Quantity

Shelves:

Pages - Analyze how certain fields effect others.  Can break current view into different areas to focus on these affected items.

Filters - Allows you to restrict what data is visible from your data set.  ('only show Alabama, Iowa and Texas', or 'display areas where sales exceed $1  mil')

Marks - Display options such as color, size, adding text or displaying more detail on rollover.

Workspace

Drag items in the Data pane into rows and columns.  This information will become a visualization that can be customized.

Show Me

When you have selected your data items, the Show Me tab provides quick display options based on the data types you selected.