Method understanding samples

Cash under pressure

The challenge of risk literacy

Risk literacy can be understood as a collection of abilities and skills that enable a critical understanding of uncertain events and the use of these in a way that is beneficial to one's own life. These abilities and skills include:

  • to correctly distinguish risk from certainty,
  • to correctly distinguish problems of risk from problems of uncertainty,
  • to correctly interpret numeric probability expressions,
  • to correctly calculate conditional probabilities,
  • to critically use graphical representations of risk (charts),
  • to question reference conditions of probability expressions,
  • to correctly distinguish between representations of absolute and relative risk,
  • to understand limitations of samples.

Why is it relevant to improve consumers' risk literacy?

However, when searching for information, consumers encounter statistics obtained by sampling. Samples never correspond 1:1 to the basic population (e.g. the total population) from which they were drawn. The quality of these samples therefore has a significant influence on the reliability of a conclusion about the total population. Different samples may lead to different conclusions. Thus, if consumers do not understand what constitutes a sample, scientific contradictions, which contribute little to their decision, are interpreted as incompetence and the role of science may be rejected.

 

Why is it problematic to improve risk literacy regarding the comprehension of samples?

If you want to increase consumers' risk competence by conveying the basics of sampling statistics, you will face the following challenges:

  • conveying the differentiation between sample and basic population,
  • conveying variations due to repeated drawing of samples,
  • recognising the relevance of engaging with reports on studies,
  • How can interest in abstract learning content like this be sparked`?
What is a suitable scientific approach?

Learning through experience. Like a scientist, the learner empirically collects observations within the framework of visualisation. This shows that a selection of individual observations is never perfect and that each new collection of observations produces different results than the previous ones.

The second decisive adjusting is storytelling in visualisations (Krzywinski & Cairo, 2013). The visualisation is "worked through" as a linear story with a problem at the beginning and a resolution at the end. This not only strengthens the motivation for learning, but also facilitates a permanent contextualisation of the current learning content into the overall framework. Interactivity can be used to nurture interest. It should be used in an economical, targeted and coherent way with the necessary controls and content.

In order to convey the understanding of samples, the basic population must first be explained. It must also be emphasised how characteristics are distributed within it. Here the exemplary counting of a fictitious basic population is possible. After this point, sampling can be introduced.

The technique of visualisation allows a comparison of the results of the sample with the previously counted results of the population in real time. This illustrates the character of a study in relation to the underlying actual distribution. Deviations can be identified as estimation errors.

The result of this is an immediate insight into how the sample size negatively affects the size of the estimation error. By allowing learners to "carry out" several studies, they not only learn that the uncertainty about estimated values remains, but that contradictory study results are possible. In addition, test questions with feedback continuously facilitate the internalisation of important aspe

The technical model was a visualization on the topic of election polls (Rock’n Poll, 2017).

Recommended literature on methodological basics
  • Krzywinski, M., & Cairo, A. (2013). Storytelling: Relate your data to the world around them using the age-old custom of telling a story. Nature Methods, 10(8), 687–688.
  • Rock’n Poll (2017). Polls explained with interactive graphics, http://rocknpoll.graphics/ (letzter Aufruf am 27.09.2019).
How can you implement the method?

Option 1: You can embed the given visualisation
It is possible to embed the visualisation from our website including the frame text via iframe. To do this, use the following html-code for your website: <iframe frameborder="0" height="800px" src="https://static.risikoatlas.de/modules/understanding-sampling/index.html" width="1024px"></iframe>

2

Option 2: You can adapt the given visualisation
If you use your own data as a multiplier, your web developers can enter it into your own consumer fact box with experience-based learning.

We will provide the person responsible for your website with the documented code for download via github. You can then edit the material. The link to the repository is available on request. Contact details can be found here.

Option 3: You can apply the scientific principle independently
If you require assistance, please consult the final report on the RiskAtlas project from July 2020 or contact us. Contact details can be found here.

When using the instruments, please mention the funding agency, which is the German Federal Ministry of Justice and Consumer Protection, and the Harding Centre for Risk Literacy as the responsible developers.

Logos can be downladed here.

Links to other methods

Methode Nichtlinearität verstehen

Method understanding relative risk

Visualisation with frame text

Cash under pressure

We often hear that Germans cling to their cash. In fact, in our European neighbouring countries it is much more common to pay even the smallest sums by card. But how important is cash really to Germans? Researchers determine the opinions of a large group of people through random sample surveys. How exactly does this work? And how representative are these samples really, and how distorted is the result when another - equally randomly selected - group of people is polled? With our visualisation you can learn how random samples are created, and thus improve your risk literacy!

 

When do I need this visualisation?

If you want to learn something about how to draw conclusions about reality from collected observations.
The purpose of this learning visualisation is to help you understand why different scientific studies based on samples never produce the exact same results, but support the same conclusion under appropriate conditions. This enhances your own risk competence. However, this learning visualisation is not supposed to support you in making a concrete decision.

 

What does the visualisation show?

The visualisation is accompanied by a sequence of comprehension questions. In addition to the correct answers, you will also receive feedback on how often other learners answered the questions correctly.

Sources and quality of the data

Where are the numbers coming from?

 

What is the quality of the data?

The underlying data is a learning example. The quality of the survey data used in the example is moderate as they are not based on household surveys with a national representative sample.

 

Last update: 14 Oktober 2019.

Links to other topics