Introduction

1.1 Background

As society becomes increasingly data-driven, the ability to understand and evaluate statistical information becomes more crucial to navigating everyday life. Statistics are present in every facet of our lives including in important topics such as healthcare, economics, politics, and education. The inability to engage with this information can be severely detrimental to our success as a citizen within society (Ipsos, 2013).

Multiple organisations have stated the importance of ‘statistical literacy’. Valerie Isham, when President of the Royal Statistical Society (RSS), said that “Statistical literacy is an essential life skill: the need to make decisions based on numerical data confronts us all in every aspect of our professional and personal lives” (2012). Achieving and/or advocating for improved Statistical Literacy is included in the targets of the Government, in their 2021 Life Sciences Vision, and the UK Statistics Authority’s (UKSA) 2020-2025 strategy “Statistics for the public good”.

The consequences of poor statistical comprehension are wide-ranging. This can include having difficulty handling personal finances as well as being more susceptible to misperceptions and misinformation. Ipsos has been conducting a global research project since 2012 titled “Perils of perception” which records the gap between public perception and reality on a wide range of subjects. The overall findings of this project demonstrate widespread misperception on numerous topics including climate change, causes of death, and health. Bobby Duffy, while Managing Director of Ipsos Social Research Institute London, stated that one reason underlying these misperceptions is “our struggle with maths and proportions” (Duffy, 2016).

But what is statistical literacy? A 2017 review concluded that “statistics educators, statisticians and researchers around the world have not reached a consensus…and hence numerous definitions of statistical literacy abound” (Sharma, 2017).

Without a clear definition of statistical literacy, there are difficulties in both establishing the current levels within the general public and implementing strategies to improve it. When establishing strategies to improve statistical literacy, education-based initiatives may be considered. It is possible that this could place undue responsibility on the data consumer, we also do not know whether it is suitable or possible for everyone to invest the necessary time to develop complex statistical literacy. Evidence continues to emerge regarding how data producers can tailor their communication to achieve more widespread understanding and engagement for a population with varying levels of statistical literacy.

1.2 Aims of the review

This report presents the findings from a literature review to establish the current landscape of statistical literacy research. Research conducted in the area of statistical literacy is multifaceted. This report includes evidence on several specific topics.

The review was commissioned by the Office for Statistics Regulation (OSR), the regulatory arm of the UK Statistics Authority, to help shape its future work in this area.

The first section of this report explores how a consensus has not been reached on a single definition of statistical literacy. The section includes a review of the definitions that have been proposed thus far, to determine commonalities and differences, as well as the reasons underlying differences between definitions. The first section also covers the tools that have been applied in the measurement of statistical literacy as these require the developer to include an explicit or implicit definition of statistical literacy to determine how the tool should be constructed.

The second section outlines how researchers have attempted to measure components related to statistical literacy. This includes not only skills directly related to statistics, but also more foundational abilities such as general literacy and numeracy.

The third section provides evidence on the topic of statistical communication, including research on how to communicate effectively to non-specialist audiences to maximise understanding and engagement. There is a vast amount of research published on communicating different types of statistics, and this review is not intended to be exhaustive. Instead, the aim is to achieve a broad overview of research findings on several key topics.

Lastly, the report summarises actions taken in the UK and internationally to improve statistical literacy and statistical communication.