Search

Sound experimental design and analysis require improved statistical training.

Peer review is at the heart of publishing scientific papers. In this first installment of a two-part Editorial, we explain how we manage the process at Nature Methods.

Bar charts are too frequently used to communicate data that they cannot represent well. We strongly encourage the use of more appropriate plots to display statistical samples.

The meaning of error bars is often misinterpreted, as is the statistical significance of their overlap.

All animals are equal, but some animals are more equal than others. —George Orwell

We’ve made some recent updates to our various content type offerings at Nature Methods. Here is a cheat sheet.

The reliability and reproducibility of science are under scrutiny. However, a major cause of this lack of repeatability is not being considered: the wide sample-to-sample variability in the P value. We explain why P is fickle to discourage the ill-informed practice of interpreting analyses based predominantly on this statistic.

We shed some light on how the Nature Methods editorial team evaluates papers submitted to the journal.

Statistics does not tell us whether we are right. It tells us the chances of being wrong.

Nature uses only the longest threads to weave her patterns, so that each small piece of her fabric reveals the organization of the entire tapestry. —Richard Feynman

It is the mark of an educated mind to rest satisfied with the degree of precision that the nature of the subject admits and not to seek exactness where only an approximation is possible. —Aristotle

Nature is often hidden, sometimes overcome, seldom extinguished. —Francis Bacon

Decision trees are a simple but powerful prediction method.

If you sit on the sofa for your entire life, you’re running a higher risk of getting heart disease and cancer. —Alex Honnold, American rock climber

Use box plots to illustrate the spread and differences of samples.

Love’s the only engine of survival. —Leonard Cohen

The exception proves the rule.

I do not think you can start with anything precise. You have to achieve such precision as you can, as you go along. —Bertrand Russell

“We demand rigidly defined areas of doubt and uncertainty!” —D. Adams

All that glitters is not gold. —William Shakespeare

Many heads are better than one.

Constraining the magnitude of parameters of a model can control its complexity

PCA helps you interpret your data, but it will not always find the important patterns.

Supervised learning algorithms extract general principles from observed examples guided by a specific prediction objective.

Clustering finds patterns in data—whether they are there or not.

With four parameters I can fit an elephant and with five I can make him wiggle his trunk. —John von Neumann

Machine learning extracts patterns from data without explicit instructions.

Tabulating the number of objects in categories of interest dates back to the earliest records of commerce and population censuses.

A P value measures a sample's compatibility with a hypothesis, not the truth of the hypothesis.

Statistics draws population inferences from a sample, and machine learning finds generalizable predictive patterns.

It is important to understand both what a classification metric expresses and what it hides.

The P value reported by tests is a probabilistic significance, not a biological one.

When some factors are harder to vary than others, a split plot design can be efficient.

Good experimental designs mitigate experimental error and the impact of factors not under study.

The ability to detect experimental effects is undermined in studies that lack power.

Today's predictions are tomorrow's priors.

Incorporate new evidence to update prior information.

Quality is often more important than quantity.

Robustly comparing pairs of independent or related samples requires different approaches to the t-test.

When multiple variables are associated with a response, the interpretation of a prediction equation is seldom simple.

When a large number of tests are performed, P values must be interpreted differently.

Good experimental designs limit the impact of variability and reduce sample-size requirements.

Nonparametric tests robustly compare skewed or ranked data.

Residual plots can be used to validate assumptions about the regression model.

For studies with hierarchical noise sources, use a nested analysis of variance approach.

When multiple factors can affect a system, allowing for interaction can increase sensitivity.