- Details
- Category: Core Skills
- Published: Friday, 07 August 2026 10:50
- Written by: Matt Pye - HWD
Introduction to Statistical Thinking
Overview
This practical course provides an introduction to statistics and statistical thinking, helping participants develop the skills to think critically about data and use it appropriately to investigate questions and support decision-making.
The emphasis is not on complex statistical techniques. Instead, participants will learn how to formulate clear research questions, recognise common sources of bias, explore and understand data through visualisation, and use appropriate summary statistics to identify and describe patterns.
Throughout the course, participants will consider real-world examples of both good and poor statistical thinking and the consequences that can arise when data is misunderstood or misrepresented. Practical exercises and real datasets will be used to demonstrate how statistical thinking can challenge preconceptions and support more robust, evidence-based conclusions.
Who should attend?
The course will be particularly useful for:
- Analysts who are new to working with statistical data
- Researchers and analysts looking for a refresher in statistical thinking
- Policy and other professionals who use, interpret or commission analysis
- Those who need to critically assess data, evidence and statistical claims
- Anyone wanting to develop greater confidence in understanding and communicating information derived from data
No advanced statistical knowledge is required.
Learning outcomes
By the end of the course, participants will be able to:
- Understand what is meant by statistics and statistical thinking
- Formulate clear and appropriately defined research questions
- Understand the importance of collecting random and representative data
- Recognise common sources of bias and consider their potential impact on analysis
- Understand how missing data can affect conclusions
- Identify appropriate types of data visualisation for different questions and datasets
- Use visualisation to explore data, identify patterns, trends, outliers and potential errors
- Select appropriate summary statistics for different types and distributions of data
- Compare data appropriately across groups and over time
- Interpret relationships between variables
- Distinguish between what can be concluded from a sample and what can be inferred about a wider population
- Communicate statistical findings clearly, concisely and responsibly
Course content
Statistical thinking and working with data
- What do we mean by statistics and statistical thinking?
- Using data to investigate questions rather than relying on assumptions or preconceptions
- The role of critical thinking when working with data
- Examples of poor statistical reasoning and its consequences
- Formulating questions, collecting data, analysing results and communicating findings
The course deliberately focuses on statistical thinking rather than introducing complex analytical methods.
Developing effective research questions
- Why a well-defined research question is essential
- Moving from a general issue to a question that can be answered using data
- Using the PICO approach:
- Population
- Intervention or comparison
- Outcome
- Defining populations, groups and measurable outcomes
- Understanding numeric and categorical variables
- When categorisation of numeric variables may or may not be appropriate
- Identifying and improving poorly formulated research questions
- Practical exercises developing and refining research questions
Understanding bias
- Why sophisticated analytical methods cannot compensate for fundamentally biased data
- Selection bias and the importance of representative samples
- Recall bias
- Confirmation bias
- Recognising how analysts' expectations can influence interpretation
- Missing data and understanding why data may be missing
- Considering whether missing data could systematically affect results
- Assessing how bias can influence conclusions
Exploring data through visualisation
- The role of visualisation in data analysis
- Using visualisations to identify errors, outliers and unusual observations
- Identifying patterns and generating hypotheses
- Understanding trends and differences between groups
- Using visualisation to communicate findings clearly
- Selecting an appropriate visualisation for the type of data and analytical question
Participants will explore common approaches including:
- Histograms for understanding the distribution of numeric data
- Scatterplots for exploring relationships between variables
- Line graphs for examining change over time
- Maps for spatial data
- Boxplots for summarising distributions and comparing groups
- Lines of best fit and other useful annotations
Practical examples will be used to develop participants' ability to interpret what a visualisation is actually showing rather than simply accepting it at face value.
Summarising and describing data
- Why summary statistics are useful
- Understanding the difference between describing a sample and drawing conclusions about a wider population
- Choosing suitable measures for different types of data
- Mean and median
- Standard deviation, interquartile range and range
- Understanding distributions and when different summaries are appropriate
- Percentages, proportions and rates
- Comparing summaries between groups
- Differences in means and medians
- Correlation and describing relationships between numeric variables
- Comparing percentages and proportions
- Measuring percentage change over time
Practical application
The course uses practical exercises and real-world datasets to enable participants to apply the concepts throughout the session. Participants will practise improving research questions, identifying potential biases, interpreting visualisations and selecting appropriate methods for summarising and describing data.
The overall aim is to give participants greater confidence in questioning data, recognising its limitations and drawing well-supported conclusions from it, rather than simply applying statistical techniques mechanically.