Statistics in the media

“Measure twice, cut once” is an English expression that underscores the importance of information for our actions. It is best to measure a piece of wood twice before cutting it, just to make sure you don’t end up with two pieces you can’t use anymore.

Statistics are everywhere. We want to know what rating other customers gave a product before we buy it. On a larger scale, we demand that our governments do research into the impact of their policies before they implement them. Research is the cornerstone of the way our societies operate.

The data it generates helps us to understand the world better. But since the data doesn’t speak for itself, it needs to be presented and explained. Since data influences our decision-making, both on a personal and a societal level, it is important to understand how it can be used to inform and also mislead people. To make sense of numbers in the media means to understand how they are used and how to draw conclusions from them.

• Statistics
  The science of collecting, analyzing, presenting, and interpreting data.
  Encyclopædia Britannica

A couple of ways you might encounter statistics in the media are in averages/means, medians, percentages and survey/poll results. These numbers can in turn be represented by visuals and infographics (see New Media 3: Data Visualization), but we will not discuss this further in this lesson.

Some methods of data collection are through:

• Polls
  • Polls usually contain one or a couple of closed questions
  • For example: who will you vote for? Choose X, Y or Z
• Surveys
  • Surveys are longer, detailed lists of questions, often containing open and closed questions
  • For example, a survey on your relationship with social media in which you answer questions about your social media consumption and mental health
• Research
  • Data can be found in government records and publications, scientific publications, company records, etc.

The data collected can be presented in different ways. Let’s go over some concepts to gain an understanding of what they mean and how they differ:

Percentages

Percentages are everywhere, but they can get confusing.

A percentage shows how much of something you have compared to the whole thing. For example, if you have a cake with 8 slices and you eat 5 slices, you ate 50%, of half, of the cake.

Percentages can also used to compare two things.

What does it mean when a yearly festival increased the number of visitors by 30%?

It means that the number of visitors this year is 30% more than the number of visitors last year. If the festival welcomed 10.000 people last year, that means that this year they received 10.000 people + 10.000 * 0.3 (this is 30% of 10.000 people) = 13.000 people in total.

You can also calculate this number like this: 10.000 people * 1.3 = 13.000 people

Another way to talk about this increase is by saying the following:

The number of visitors to the yearly festival this year was 130% of last year’s visitors.

This means the same thing as the earlier statement about a 30% increase of visitors. The difference here is that the percentage does not describe the increase of visitors, but compares the total number of visitors to the total number of last year’s visitors.

If 100% of last year’s visitors showed up this year, that would mean that the number of visitors had stayed the same: 10.000 in both years.

If last year’s number of visitors increased by 100%, that means the number of visitors this year had doubled to 20.000.

Relationships: correlation and causation

Suppose a study finds that when ice cream sales are high, crime rates tend to be higher as well.

However, it would be a mistake to assume that ice cream sales cause crime or that crime causes ice cream sales. The true cause of the relationship between the two could be a third variable, such as temperature. It just so happens that on hot days, people are more likely to both buy ice cream and engage in more outdoor activities, which could lead to higher crime rates.

Correlation is a statistical relationship between variables.

Causation is a statistical relationship between variables in which one variable influences the other(s).

In this case, there is a correlation between ice cream sales and crime rates, not causation. Both ice cream sales and crime went up, but neither one caused the other to go up. Correlation does not imply causation.

Read this article to find some real world examples.

Students are divided in groups of 3-4 students. Groups will use the provided statistical data below to tell different stories about climate change and renewable energy.

1. Divide the students in groups of 3-4.
2. Groups go over the statistics and discuss what they mean.
3. Groups come up with three different titles and subtitles for a news article, taking information from the statistics to back up their point. Every article concept has a different goal. Additional research is welcomed.
   1. The first article concept is very optimistic about the energy transition from fossil fuels to renewable sources
   2. The second concept is very pessimistic about the energy transition from fossil fuels to renewable sources
   3. The third concept makes outrageous fake news claim
      
      For example:
      Title: One out of five ducks is unhappy
      Subtitle: New research shocks the chicken community
4. Groups present their article concepts to the class who give feedback.

Data

In 2020, renewable energy was responsible for:

• 22.1% of energy consumption in the EU
• 37.5% of electricity consumption in the EU
• 23.1% of energy used for heating and cooling in the EU
• 10.2% of energy used for transport
• Source

Of this renewable energy:

• 36% was wind-generated
• 33% was hydro-generated
• 14% was solar-generated
• 8% was generated with solid biofuels
• 7% was generated with other renewable sources
• Sources

A 2016 survey shows that Europeans believe that more of their energy should be generated using:

• coal — 13% (mean of 23 countries)
• gas — 31%
• nuclear — 19%
• hydro — 69%
• wind — 78%
• solar — 72%
• biomass — 45%
• Source (find more data here)

Every group presents its three article concepts to the class and other students give feedback.

1. Who is responsible to help people understand statistical data in the media?
2. Should statistics be taught in schools?
3. Are media allowed to use statistics in a sensationalist way?
4. Should social media fact-check statistics that are shared on their platforms?
5. How can data visualizations help to clarify statistics?