T3.1 Systematic literature review dataset

This document presents the T3.1 Systematic Literature Review Dataset, which was collected for the VERITY project, to identify and categorise the evidenced-based approaches to tackle mistrust in science. This dataset includes reference information of 19 articles that were systematically selected by conducting the Preferred Items for Systematic Reviews and Meta-Analyses (PRISMA) search strategy.

The PRISMA strategy was driven by a rigorously designed search string, which was devised following the Population, Intervention, Comparison, Outcome (PICO) guidelines. PICO guidelines worked as our framework for creating the necessary keywords associated with society (Population), and intervention studies (Intervention) measuring trust (Comparison) in science (Outcome). We searched three databases (PubMed, Scopus, and Web of Knowledge); the search string was adjusted to correspond to the specific requirements of each database.

The search was completed on 03 May 2023 and was restricted to titles and abstracts, with no restrictions on publication date. The search strategy yielded a total of 2835 results. To proceed with PRISMA, these results were uploaded on CADIMA (www.cadima.info), an open-access web tool for conducting systematic literature reviews (see Kohl et al., 2018). After removing duplicates, 2097 studies remained for title and abstract screening. To identify the studies that used an intervention (or manipulation) to increase trust in science, title and abstract screening was based on the following pre-defined criteria:

1. abstract written in a language that can be understood by reviewers (English or Greek)
2. involve trust of any kind
3. include an intervention (either measuring trust pre- vs. post-intervention or between an experimental vs. control group)
4. original piece of research (present primary data)
5. focus on science.

Through title and abstract screening, 1846 records were excluded. Respectively, 251 records followed to the next PRISMA phase, that of full-text screening. Full-text screening was guided by the following three criteria:

1. investigation of people's trust in science
2. investigation of a manipulation designed to influence people's (mis)trust
3. typically developed and healthy (control) sample.

The reasons of these adjustments were twofold: first, full-text screening requires stricter criteria that reflect the objectives of the task, and second, some pre-set exclusion options are already provided by CADIMA (full-text not-available/assessable, duplicate, no-primary data presented). Following these criteria, full-text screening excluded 232 records. As such, 19 studies were identified that were eligible for the purpose of this report of identifying and categorising strategies, methods, and tools that intervene on societal (mis)trust in science.