How can you help non-statisticians understand statistical concepts?

Helping non-statisticians understand statistical concepts involves breaking down complex ideas into simpler terms, using relatable examples, and avoiding jargon as much as possible. Use Everyday Language: Explain statistical concepts using simple language, avoiding technical terms and jargon. Use analogies and metaphors that are relatable to everyday experiences. Provide Real-Life Examples: Illustrate statistical concepts with real-world examples that people can relate to. Show how statistics are used in various fields such as sports, healthcare, finance, and social sciences. Visualize Data: Use visualizations like charts, graphs, and diagrams to represent statistical concepts. It can make abstract ideas more easier to grasp.

I don't pitch statistical concepts to managers, scientists and medical liaisons in the same way. It is paramount to understand what the needs of the client are and to tailor the service (consulting/training/coaching) accordingly. For instance, managers and MLS are more focused on decision-making. These people do not see the data nor do they carry out statistical analysis. I provide them with tools to determine the validity, precision, scope and limitations of published results.

The target of your statistical explanations influences the description of your task. A student would need a reference to your code, you could provide a link to a GitHub repository for your project and some code screenshot which could make technical concepts clear to digest. In the case of companies, you should focus more on showing results you achieved. Use summary plots with the most interesting numbers and statistics to do that.

Al presentar resultados, es aconsejable simplificar la conversación; la complejidad del proceso no a?ade tanto valor en la presentación. No es imprescindible recurrir a gráficos o conceptos complicados para mostrar los resultados. En la mayoría de los casos, se pueden exponer de forma sencilla, ya sea mediante gráficos simples o a través de una narrativa, siempre enmarcados en un contexto de negocio.

I've always appreciated the saying "be clear, be brief, be gone." As statisticians, we possess a wealth of technical expertise in our methods and tools for extracting information and knowledge. However, to effectively convey this knowledge, we must also excel as storytellers. This means simplifying our language and visualizations to succinctly communicate the essential insights to stakeholders, without burdening them with the intricacies of our journey and technical requirements.

Use plain language and avoid jargon. Such as,the standard deviation measures the variation in data values from the average, with small deviations indicating more consistency and large deviations indicating more spread-out data. Choose the appropriate level of detail and complexity for your audience. Focus on conveying the main message or key findings by providing context and interpretation for the results, rather than just raw numbers. Use visual aids to supplement your verbal or written explanation. Using charts, graphs, tables, and diagrams can effectively display data patterns, trends, and relationships, but it's crucial to choose the appropriate type for the data and purpose.

Many believe to understand probability pretty well but struggle to understand it conceptually. Like, if we toss a fair coin 10 times, head comes thrice then we should say P(H) is 0.3 instead of 0.5. That's why we define probability with infinite tosses with infinite coins, so that P(H) converges to 0.5. Probability is accompanied with a limiting factor, the chance of head converges to half as number of trials are being repeated indefinitely. A simple concept like probability isn't only the ratio of favourable outcomes to total outcomes, seldom explained by roll of a fair dice or a fair coin toss; it's beyond that. One needs to understand the philosophy of the idea to get a logical understanding of it.

Find the context that matters to a person and creatively apply statistics to that problem. Take that to them, with lots of visualisations and step by step work that they can discuss/perform/get involved with. Demystify by creating contextualised meaning.

When I teach my course understanding the Understanding Biostatistics in Medicine to Medical Science Liaisons and Medical Writers, I make sure that after we cover a statistical concept, we immediately explore their suggested scientific publications so that they can relate and keep them engaged. One needs to keep in mind that not everybody can access the raw data and carry out statistical analyses.

One way to help non statisticians understand statistical concepts is through intuitive examples. For example here is how I explain degrees of freedom. Imagine being asked to choose 5 numbers that sum to 100. You say 20, 20, 20, 20. Before you utter the 5th number, I tell you that 5th number is 20. This is because the first 4 numbers chosen by you summed up to 80 and the condition was to chose 5 numbers which sums to 100. So 100 -80 = 20. The fifth number is 20. The fifth number in a sense chose itself due to the condition specified. You had 4 degrees of freedom alone. We needed the first 4 numbers only. This also intuitively explains the n-1. Here n being 5 and n-1 =4. You lost 1 dof to the condition of “5 numbers should sum to 100”

One thing I would like to add, rather suggest statisticians is not to suggest a book to non statisticians who has shown interest to learn the core concepts for the first time. The first step should be an attempt to explain the concept in very simple language with more than one real life examples if possible.I have noticed not always everyone will relate to just one example from one particular domain. Statisticians should also direct the learners to resources and content which is verified as lot of misconceptions regarding theories exist nowadays.

Taking a sample from a population to study a specific phenomenon is like to take a small sip from a cup of tea to test the taste of the tea. This sip is considered representative of the cup of tea and therefore gives me a clear idea of the distribution of the taste of tea. If the sip is acceptable to me, then my decision is that the cup does not need additional sugar. However, if I taste a different taste, my decision is that the cup of tea needs an additional spoonful of sugar, for example, and not the sip I took from the cup. Thus, we have generalized the sample results to the population.

Over the last 30 years, whenever a client wants to attend a statistical training, I design a unique training curriculum based on the case studies of the client. This way the hands-on application resonate with the client's needs. The sequence of the statistical concepts introduced in the training curriculum is also key.

I am going to start at the ground floor. I’m going to a high school and explain and demonstrate randomness. Many students believe that there is a reason for everything, without the concept random events.