Bridging statistics and life sciences undergraduate education

Figures & data

Figure 1. Statistics requirement in life science programs of U15 Universities A. A systematic search demonstrated that 21.1% of programs do not require statistics courses (n = 70) and that 45.2% of programs only require one statistics course (n = 150). B. The majority (58.6%) of required or elective statistics courses are at the second-year (i.e. level 2) of undergraduate training (n = 194). Detailed methodology for panels A and B are described in the Supplemental Data.

Figure 2. Course model for Statistics and Scientific Inquiry in the Life Sciences. Course learning outcomes were designed to mirror the scientific inquiry process, and were mapped to course activities and assessments to enable dissemination, practice and application of statistical principles in a collaborative classroom environment.

Table 1. Common statistical errors in scientific inquiry. Examples of common statistical errors made at each stage of the scientific inquiry process from the literature.

Stage of Scientific Process	Common Statistical Errors	Reference
Identifying the Problem	Unable to critically assess existing literature and build research questions on non-robust findings	Abdullah et al. 2015 Janick-Buckner 1997 Rauschert et al. 2011
Planning the study	Failing to understand biases of different sampling methods and study designs Using the wrong study design (e.g. cohort vs. cross-sectional) for proposed research question Not recognising and controlling for confounds	Williamson 2003 Ercan et al. 2007 Allen-Mersh 1997
Collecting the data	Not reporting sample drop-outs or documenting outliers Converting continuous data into categorical data Using misleading data visualisations	Ercan et al. 2007 Lang 2004 Murphy 2004
Analysing the data	Failing to check model assumptions (e.g. independence, normality) prior selecting statistical test	Weissgerber et al. 2016 Lazic 2010 Thiese, Arnold, and Walker 2015
Interpretation of data	Interpretating observed differences or trends as significant Confusing clinical significance with statistical significance	Ercan et al. 2007 Weissgerber et al. 2016 Ranganathan, Pramesh, and Buyse 2015

Figure 3. The relevance of statistics at all stages of the scientific inquiry process. Scientific inquiry relies on identification of a scientific problem worth studying, after identification of gaps in the scientific literature. A plan is constructed, with clear identification of a study hypothesis, study design and methodologies. Data collection ensues and is later compiled and analysed using data summaries, and descriptive and inferential statistics. Conclusions can be made, enabling contribution of knowledge to the life sciences field of interest. These data and conclusions can then be used as foundations for establishment of new plans for study.

Abdullah, C., J. Parris, R. Lie, A. Guzdar, E. Tour, and N. Pelaez. 2015. “Critical Analysis of Primary Literature in a Master’s-Level Class: Effects on Self-Efficacy and Science-Process Skills.” CBE Life Sciences Education 14 (3): ar34. doi:10.1187/cbe.14-10-0180.

 PubMed Web of Science ®Google Scholar

Janick-Buckner, D. 1997. “Getting Undergraduates to Critically Read and Discuss Primary Literature.” Journal of College Science Teaching 27 (1): 29–32.

 Google Scholar

Rauschert, E. S. J., J. Dauer, J. L. Momsen, and A. Sutton-Grier. 2011. “Primary Literature across the Undergraduate Curriculum: Teaching Science Process Skills and Content.” Bulletin of the Ecological Society of America 92: 396–405. doi:10.1890/0012-9623-92.4.396.

 Google Scholar

Williamson, G. R. 2003. “Misrepresenting Random Sampling? A Systematic Review of Research Papers in the Journal of Advanced Nursing.” Journal of Advanced Nursing 44 (3): 278–288. doi:10.1046/j.1365-2648.2003.02803.x.

 PubMed Web of Science ®Google Scholar

Ercan, I. Y., B. Yang, Y. Ozkaya, G. Cangur, S. Ediz, and B. Kan. 2007. “Misusage of Statistics in Medical Research.” European Journal of General Medicine 4 (3): 128–134.

 Google Scholar

Allen-Mersh, T. 1997. Cancer: Principles and Practice of Oncology. V. T. Devita Jr, S. Hellman, and S. A. Rosenberg eds, 286× 220 mm Illustrated 1996. Philadelphia, Pennsylvania: Lippincott-Raven.£ 160. In: Oxford University Press.

 Google Scholar

Lang, T. 2004. Twenty Statistical Errors Even You Can Find in Biomedical Research Articles. Citeseer.

 Google Scholar

Murphy, J. R. 2004. “Statistical Errors in Immunologic Research.” The Journal of Allergy and Clinical Immunology 114 (6): 1259–1263. quiz 1264 10.1016/j.jaci.2004.09.023.

 PubMed Web of Science ®Google Scholar

Weissgerber, T. L., V. D. Garovic, J. S. Milin-Lazovic, S. J. Winham, Z. Obradovic, J. P. Trzeciakowski, and N. M. Milic. 2016. “Reinventing Biostatistics Education for Basic Scientists.” PLoS Biology 14 (4): e1002430. doi:10.1371/journal.pbio.1002430.

 PubMed Web of Science ®Google Scholar

Lazic, S. E. 2010. “The Problem of Pseudoreplication in Neuroscientific Studies: Is It Affecting Your Analysis?” BMC Neuroscience 11: 5. doi:10.1186/1471-2202-11-5.

 PubMed Web of Science ®Google Scholar

Thiese, M. S., Z. C. Arnold, and S. D. Walker. 2015. “The Misuse and Abuse of Statistics in Biomedical Research.” Biochemia medica 25 (1): 5–11. doi:10.11613/BM.2015.001.

 Web of Science ®Google Scholar

Ranganathan, P., C. S. Pramesh, and M. Buyse. 2015. “Common Pitfalls in Statistical Analysis: “No Evidence of Effect” versus “Evidence of No Effect”.” Perspectives in Clinical Research 6 (1): 62–63. doi:10.4103/2229-3485.148821.

 PubMedGoogle Scholar