To Be a Frequentist or Bayesian? Five Positions in a Spectrum

2Column Editors' Note: The new column, Meta Data Science, presents short articles that discuss philosophical issues around the practice and theory of data science, ranging from ethical to epistemological and metaphysical questions.It aims to promote philosophy-minded data science and data-science-minded philosophy, and we welcome proposals for future columns on these issues.In this inaugural article, one of this column's editors, Hanti Lin (University of California, Davis), revisits a traditional debate: frequentism versus Bayesianism in the foundations of statistics.Instead of two dichotomized views, he proposes a spectrum with five positions-from radical frequentism to radical Bayesianism with intermediate views in between-on the basis of the existing practice of statistical science.