statistics
Can We Smell A Trillion Odors?
Can We Smell A Trillion Odors?
Kerry Grens | Jul 8, 2015
A reanalysis calls into question a year-old claim that humans can decipher at least 1 trillion different scents.
WHO: TB’s Toll Worse Than Thought
WHO: TB’s Toll Worse Than Thought
Bob Grant | Oct 22, 2014
A new report from the World Health Organization finds that tuberculosis has infected hundreds of thousands more people around the world than was estimated a year ago.
Week in Review: October 13–17
Week in Review: October 13–17
Jef Akst | Oct 17, 2014
Snail not extinct after all; results too good to be true?; mice need myelin production for motor learning; keeping the brain young; the evolution of archaea
Epigenetics Paper Raises Questions
Epigenetics Paper Raises Questions
Kate Yandell | Oct 16, 2014
GENETICS publishes a commentary criticizing a Nature Neuroscience paper claiming that mice can inherit smell sensitivities that their parents acquired during life.
Screen the Healthy?
Screen the Healthy?
Tracy Vence | Mar 12, 2014
In light of a study that showed blood-based biomarkers could predict future cognitive impairment, researchers discuss statistical analyses and the problem of false positives.
BPTW: By the Numbers
BPTW: By the Numbers
The Scientist Staff | Aug 1, 2013
Take a closer look at some of the statistics generated by The Scientist's Best Place to Work Academia 2013 survey.
Opinion: Statistical Misconceptions
Opinion: Statistical Misconceptions
Vladica M. Veličković | Jul 31, 2013
Researchers must be wary of the common mistakes of correlation analysis when drawing conclusions about the nature of their data.
Worried Sick
Worried Sick
Megan Scudellari | Jul 1, 2013
Expectations can make you ill. Fear can make you fragile. Understanding the nocebo effect may help prevent this painful phenomenon.
Bad Stats Plague Neuroscience
Bad Stats Plague Neuroscience
Bob Grant | Apr 16, 2013
A new study blames the unreliable nature of some research in the field on underpowered statistical analyses.
Stats Are Right Most of the Time
Stats Are Right Most of the Time
Beth Marie Mole | Jan 28, 2013
A new analysis suggests that only 14 percent of published biomedical results are wrong, despite prominent opinions to the contrary.