The impact of using artificial data in undergraduate statistics students’ projects due to COVID-19 lockdowns

• Peter K. Dunn ^[1]
  1. [1] University of the Sunshine Coast

     University of the Sunshine Coast

     Australia

     
• Localización: International journal of mathematical education in science and technology, ISSN 0020-739X, Vol. 55, Nº. 8, 2024, págs. 1759-1768
• Idioma: inglés
• DOI: 10.1080/0020739X.2022.2056095
• Enlaces
  • Texto completo
• Resumen

  • The use of group work projects is common in introductory statistics courses, including projects where students collect their own data. However, the COVID-induced lockdown at the start of 2020 meant that data collection was compromised. In this study, we examine a situation where students were permitted to use artificial (made-up) data for their project during the COVID lockdown. We examine features of these project reports and the data the students used before, during and after the lockdown. We find evidence that during-COVID projects (based on artificial data) were more likely to study people but less likely to involve interaction with people, and were less likely to require t-tests for analysis. We also found that the artificial data were less likely to include outliers, but no evidence that the artificial data were contrived to produce ‘small' p values.