Researchers from the University of Amsterdam’s (UvA) Psychology Department have shown that willingness to share research data is related to the strength of the evidence and the quality of the statistical results.
To what extent are academics willing to share their research data with fellow scientists and scholars? Researchers from the University of Amsterdam’s (UvA) Psychology Department have shown that such willingness is related to the strength of the evidence and the quality of the statistical results. Their findings were published in the scientific journal PLoS ONE on 2 November.
A lack of willingness to share research data is often ascribed to fear among academics that independent reanalysis of the data will point up errors in their work or lead to contrasting conclusions. UvA researchers Jelte Wicherts, Marjan Bakker and Dylan Molenaar conducted the first-ever systematic investigation of this assumption, correlating unwillingness to share data for reanalysis to 1,148 statistically significant research results in 49 academic articles published in two leading psychology journals.
They found that unwillingness to share data is related to weaker evidence and a higher prevalence of apparent errors in the reporting of statistic results. This link was particularly apparent where such errors concerned the statistical significance of the published research results. The researchers concluded that statistical results are especially difficult to verify in cases where reanalysis poses a greater chance of arriving at a contrasting conclusion. According to the team, this finding highlights the critical need to lay down mandatory policies on data archiving.