# 9 Lab [en] - Data visalization, LOD, Quick Calculations
###### tags: `Data Visualization` `level of details` `quick calculations`

[TOC]

# Introduction - exercise objective
The purpose of this class is to perform exercises using Tableau's advanced calculation formulas.

# 1 Source data

The exercise is based on the *"Sample EU SuperStore "* data source built into Tableau.

# 2.1 Calculation formulas in Tableau

## 2.1 Quick Table Calculations
The system has a built-in library of calculation functions that can be used to define calculation fields, as shown below.


The most frequently used calculation formulas have been gathered in a pop-up menu, which allows to quickly call up a predefined calculation formula in the context of a selected measure.


The following video demonstrates the technique of using *Quick Table Calculations* along with converting predefined formulas into regular calculation fields and editing them.

{%youtube 3o_3WRT9OWI %}

## 2.2 Pop-up menus - defining the context in which calculation formulas work
When using the built-in, handy *Quick Table Calculations* formulas, we have the ability to define the scope -- context of the formulas used. Using the *Compute using* list available in the pop-up menu (figure below), we can define how the calculation formulas work.


Slightly more possibilities than the above list are given to us by using the *Edit Table Calculation...* option from the popup menu, which brings up the dialog as shown below.


An example use of the options described above is illustrated in the video below.

{%youtube UeFOCJJmq4I %}

## 2.3 Level of details - level of details analysis in Tableau
There are three Level of details (*LOD*) expressions available in Tableau:
* FIXED,
* INCLUDE,
* EXCLUDE.

These levels allow you to define how formulas are calculated.

### 2.3.1 FIXED LOD
The FIXED level of detail calculates values using selected dimensions, regardless of the dimensions presented in the visualization.
:::info
**Example**.
{FIXED [Region] : SUM([Sales])}
:::
shows the sales value regardless of the dimensions shown in the selected view - figure below. The figure shows a table where the sales value is expressed in the context of (for) the region regardless of the State/Province dimension.


When changing the FIXED attribute to State/Province, we get a visualization as in the next figure.


### 2.3.2 INCLUDE LOD
INCLUDE LOD calculates values using the indicated dimension (in INCLUDE) in addition, relative to the dimensions currently used in the visualization.

:::info
**Example**.
AVG({ INCLUDE [State] : SUM(Sales)})
:::
The expression will calculate the average value of sales in the context of each state. This will be a completely different value than: AVG(Sum(Sales)) - compare in the figure below.

The top panel of the above figure shows the average calculated from the LOD expression with the include option set to *State/Province* - the result is the average sales values for the state. The bottom panel plots the relationship *AVG(Sales)*, which is the average sales value calculated from all sales records (without looking at *State/province*).

### 2.3.1. EXCLUDE LOD
EXCLUDE LOD indicates the dimensions to be excluded from the visualization.

:::info
**Example**.
{EXCLUDE [Region]: SUM([Sales])}
:::
The above example bypasses *Region* when counting the sum. In the figure below, you can see the effect of using such a defined field to improve chart coloring.

In the figure above, the values for the *region* data series are colored using *SUM(sales)* -- but this does not allow for proper coloring using the full color palette, because the same scale is used for each region.

Using *{EXCLUDE [Region]: SUM([Sales])}* for coloring allows you to draw an independent scale for each region, as shown below.

:::spoiler
EXCLUDE LOD is a useful expression for constructing a visualization showing a relationship like "percent of total" or "difference/difference from average". I ask those interested to review the materials available at: [help.tableau.com](https://help.tableau.com/current/pro/desktop/en-us/calculations_calculatedfields_lod.htm)
:::

# 3 Exercise - mini project
> **Packacking waste - analysis of the structure and recycling rates for plastic and glass waste**
>
> * data source: Eurostat (sample: [Packaging waste statistics](https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Packaging_waste_statistics))
> * sample analysis/report: [BAD-MOUTHING PLASTIC DOESN’T SOLVE ANY PROBLEMS](https://kampus.umcs.pl/pluginfile.php/765865/mod_page/intro/raport%20plastics%20en.pdf)

Prepare the short project (3-4 dashboards) in gropus of two that will analyze packaging waste statistics provided by Eurostat and answer research questions:
1. Which packaging waste streams are dominant and how has this changed over last 10 years?
2. Which type of packaging waste: glass or plastic is more common and better recycled in EU countries?
3. Is there a correlation between economic and demographic factors and waste levels in EU countries?