Article numbers as a leading indicator of publication time

Keywords: time series, auto-correlation function, cross-correlation function, leading indicator

Abstract

Aims: in health professions education (HPE), the use of statistics is commonly associated with somewhat larger samples, whereas smaller samples or single subjects (i.e., N = 1) are usually labelled as needing some kind of ‘qualitative’ approach. However, statistical methods can be very useful in small samples and for individual subjects as well, especially where we have time series of repeated measurements of the same outcome variable(s) of interest. The aim of this article is twofold: to demonstrate an example of a cross-correlation function for single subjects in a HPE context and to suggest a few settings in HPE where this cross-correlation function can be of use.
Method: the example uses data from a recent Open Access publication on among others article numbers and publication time in a number of major HPE journals to examine the relation between the number of articles published and median publication time over time in the zero-cost Open-Source statistical program R version 4.0.5.
Results: as to be expected, the number of articles published appears somewhat of a leading indicator of publication time: both number of articles in year ‘y’ and number of articles in year ‘y minus 1’ correlate > 0.6 with median publication time in year ‘y’, while correlations of other time differences (e.g., number of articles in year ‘y minus 2’ and median publication time in year ‘y’, or median publication time in year ‘y’ and number of articles in year ‘y plus 1’) are substantially smaller.
Conclusion: in line with recent literature, this article demonstrates that the cross-correlation function can be used in the context of small samples and single subjects. While the example focusses on article numbers and publication times, it can equally be applied in for example studying relations between knowledge, skills and attitude in individuals, or relations between behaviors of individuals working in pairs or small groups.

Downloads

Download data is not yet available.

Author Biography

Jimmie Leppink, University of York, Hull York Medical School, York, North Yorkshire, United Kingdom.

PhD in Statistics Education, LLM in Forensics, Criminology and Law, and MSc in Psychology and Law from Maastricht University, the Netherlands; MSc in Statistics from Catholic University of Leuven, Belgium; currently Research Director at Hospital Virtual Valdecilla (HvV), Santander, Spain.

References

Van de Schoot R, Miocevic M. Small sample size solutions: A guide for applied researchers and practitioners. New York: Routledge; 2020. 285 p. https://doi.org/10.4324/9780429273872

Brockwell PJ, Davis RA. Time series: Theory and methods (2nd ed.). New York: Springer Verlag; 1991.

Maggio LA, Bynum WE, Schreiber-Gregory DN, Durning SJ, Artino AR. When will I get my paper back? A replication study of publication timelines for health professions education research. Perspect Med Educ. 2020;9:139-46. https://doi.org/10.1007/s40037-020-00576-2

R Core Team. R: A language and environment for statistical computing [Internet] Vienna: R Foundation for Statistical Computing (version 4.0.5). Available from: https://www.r-project.org

Published
2021-11-05
How to Cite
Leppink, J. (2021). Article numbers as a leading indicator of publication time. Scientia Medica, 31(1), e41065. https://doi.org/10.15448/1980-6108.2021.1.41065
Section
Education in Health Sciences