Original subtitles

Narrator: Now that we've seen the different types of data

and levels of measurement we can have,

we are ready to explore different graphs and tables,

which will allow us to visually represent

the data we are working with.

Visualizing data is the most intuitive way to interpret it,

so it's an invaluable skill.

It is much easier to visualize data

if you know its type and measurement level.

As you may recall, there are two types of variables,

categorical and numerical.

In this video, we will focus on categorical variables.

Some of the most common ways to visualize them

are frequency distribution tables, bar charts, pie charts,

and Pareto diagrams.

First, let's see

what a frequency distribution table looks like.

It has two columns, the category itself

and the corresponding frequency.

Imagine you own a car shop and you sell only German cars.

The table below shows the categories of cars,

Audi, BMW, and Mercedes,

and their frequency,

or in plain English, the number of units sold.

By organizing your data in this way,

you can compare the different brands

and see that Audi has been sold the most,

so that is a frequency distribution table.

However, tables aren't much fun, are they?

Using the same table,

we can construct a bar chart, also known as column chart.

The vertical axis shows the number of units sold,

while each bar represents a different category

indicated on the horizontal axis.

In this way, it is much, much clearer

that Audi is the best selling brand.

Okay, let's represent the same data as a pie chart.

In order to build one, we need to calculate

what percentage of the total each brand represents.

In statistics, this is known as relative frequency.

Naturally, all relative frequencies add up to 100%.

Pie charts are especially useful

when we want to not only compare items among each other,

but also see their share of the total.

Okay, this example could be easily transformed

into a business example of market share.

Market share is so predominantly represented by pie charts

that if you search for market share in Google Images,

you would only get pie charts.

Imagine that the data in our table

is representing the sales of Audi, BMW, and Mercedes

in a single German city, say, Bonn.

The chart will show us the market share

that each of these brands has.

Lastly, we have the Pareto diagram.

In fact, a Pareto diagram

is nothing more than a special type of bar chart

where categories are shown in descending order of frequency.

By frequency,

statisticians mean the number of occurrences of each item.

As we said earlier, in our example,

that's exactly the number of units sold.

Let's go back to our frequency distribution table

and order the brands by frequency.

Now we can create the bar chart

based on the reordered table,

and voila, we almost have a Pareto diagram.

There is one last touch to make it one,

a curve on the same graph, showing the cumulative frequency.

The cumulative frequency

is the sum of the relative frequencies.

It starts as the frequency of the first brand

and then we add the second, the third, and so on,

until it finishes at 100%.

This polygonal line is measured by a different vertical axis

on the right of the graph.

At each of its vertices,

it shows the subtotal of the categories to its left.

See how the Pareto diagram

combines the strong sides of the bar and the pie chart?

It is easy to compare the data both between categories

and as a part of the total.

Furthermore, if this was a market share graph,

you could easily see the market share

of the top two or top five companies.

Finally, it is named after Vilfredo Pareto.

You may have heard of another idea of his,

the Pareto Principle, also known as the 80-20 rule.

It states that 80% of the effects

come from 20% of the causes.

A real life example is a statement by Microsoft

that by fixing 20% of its software bugs,

they managed to solve

80% of the problems customers experience.

A Pareto diagram can reveal information like that.

It is designed to show how subtotals change

with each additional category

and provide us with a better understanding of our data.

Okay, these are the main ways

in which we can visually represent categorical data.

In our next lecture, we will get acquainted

with the most useful graphs and tables for numerical data.

Thanks for watching.