Annex A: Approach

The aim of this review was to gain insight into the current landscape of statistical literacy research. At the inception of the review, key terms were used to complete searches in Google and Google Scholar to identify how readily available evidence was in this area. It was established at an early stage that the terms associated with statistical literacy were not sufficiently narrow to filter records to a manageable number to review within the time frame of the review. The search terms were also not leading to a substantial number of highly relevant records.

Therefore, the decision was made to take a more pragmatic approach to the review. This entailed identifying highly relevant records from the initial searches and suggestions by members of the project group at OSR. Once a set of initial relevant articles were collected, their reference lists and papers that cited these articles were then screened in a snowballing approach.

Furthermore, a Twitter scraping tool, developed internally by OSR, was also applied using search terms including: “improving statistical literacy”, “communicating statistics”, “effective communication of statistics” and “general public statistics understanding”. This approach was used to pick up additional recent research in these areas and capture evidence that may not refer to “statistical literacy” specifically. Any relevant evidence detected using this approach had already been picked up in the initial search engine approach.

To gain an additional understanding of the OSR’s previous stance in the areas of statistical literacy that arose in this review, the OSR website was also used to identify any relevant correspondence including key terms. Searches were conducted on published correspondence which are made available on the OSR website.

The initial findings of the review were presented in an OSR-wide session, also attended by Johnny Runge (ESCoE). Attendees were asked for feedback on any potential gaps in the evidence base or misinterpretations of the data. This report also underwent a round of peer-review by researchers identified as key members of the Statistical Literacy community (Johnny Runge and David Spiegelhalter).

Overall, this review and the approach used were intended to provide a broad overview of the literature around statistical literacy rather than to provide a systematic or exhaustive summary of all relevant literature. The results of the review will therefore be limited by the search terms applied and the records identified in the initial stages of the review.