We chose our company name, Creascience to emphasize that creativity can mix efficiently with science. In this blog, we share illustrations of how a correct use of statistics leads to drastic improvements of scientific research. Come back often to read our latest findings or simply follow this link to get notified by email (guaranteed 100% ad-free).
-
The Primary Outcome Is Positive — Is That Good Enough?
Clinical trial findings are often simplified into a binary conclusion, focusing on a P value of less than 0.05 for a treatment difference. However, a more nuanced interpretation requires examining the total evidence, including secondary end points, safety issues, and trial size and quality. This article aims to facilitate a more sophisticated and balanced interpretation…
-
Effect Size, Confidence Interval & Statistical Significance: A Practical Guide for Biologists
Main Ideas Reference Nakagawa, S., & Cuthill, I. C. (2007). Effect size, confidence interval and statistical significance: a practical guide for biologists. Biological Reviews, 82(4), 591–605. doi:10.1111/j.1469-185x.2007.00027.x
-
Power Failure: Why Small Sample Size Undermines the Reliability of Neuroscience
The issues discussed in this paper are clearly not only prevalent in neuroscience… I know that this is a 2013 paper and hopefully the situation has improved, but allow me to seriously doubt it. Nevertheless, I decided to post it because of the general nature of the problems and also because this is an easy…
-
Reporting Experimental Designs in Scientific Publications
Scientific rigour is a major concern and most scientific journals have a policy on statistical analysis and data representation to ensure that issues of experimental design can be reviewed thoroughly. Experimental design is one crucial component of a scientific method. “A well-designed, properly conducted experiment aims to control variables in order to isolate and manipulate…
-
The Winner’s Curse: Effect Size Inflation as a Function of Statistical Power
Consequences Illustration The figure shows simulations of the winner’s curse (expressed on the y‑axis as relative bias of research findings). These simulations suggest that initial effect estimates from studies powered between ~ 8% and ~31% are likely to be inflated by 25% to 50% (shown by the arrows in the figure). Inflated effect estimates make…
-
Error Bars in Science – Erroneous Display?
Today I would like to focus on the confusing reporting of findings using mean +/- error bars. Error bars are graphical representations that can help convey information about the variability or uncertainty in data. They are used in charts, graphs, or plots to show the range of values within which the true value of a…