The only real mistake is the one from which we learn nothing

Naomi Havron is currently a postdoc at the LSCP in Paris, France. She is investigating syntactic adaptation and syntactic category learning in children and infants.

I spent four years writing my thesis, four years of ups and downs, p values smaller than .05, but also some t values smaller than .05. At times, I felt confident and optimistic, at times less so – this was somewhat correlated with the p values, but not significantly so.

Then it came time to start gathering everything I did into an article thesis. In the Hebrew University, at least in my year, not all of these articles had to be published articles, and you could also include manuscripts you did not plan to submit for publication. I thought I would write up everything, even my null results, because whoever reads my dissertation (well, at least all those 1st year PhD students and my grandfather) could probably learn just as much from my failed attempts than from my success stories. Continue reading The only real mistake is the one from which we learn nothing

Advertisements

How to run a meta-analysis? A video tutorial

Christina, Page and I like meta-analyses. We are convinced they are a great tool to leverage past research in order to move forward: To gain an overview of the state of a field, to get an idea of research practices, to plan new experiments, and even to get novel theoretical insights.

Continue reading How to run a meta-analysis? A video tutorial

Find a significant effect in any study

Stephen Politzer-Ahles is Assistant Professor at the Department of Chinese and Bilingual Studies of The Hong Kong Polytechnic University. He is committed to finding solutions to current challenges in the cognitive sciences. For instance, he is developing efficient and transparent strategies to empty out his own file drawer.

p>.05. We’ve all been there. Who among us hasn’t had a student crying in our office over an experiment that failed to show a significant effect? Who among us hasn’t been that student?

Statistical nonsignificance is one of the most serious challenges facing science. When experiments aren’t p<.05, they can’t be published (because the results aren’t real), people can’t graduate, no one can get university funding to party it up at that conference in that scenic location, and in general the whole enterprise falls apart. The amount of taxpayer dollars that have been wasted on p>.05 experiments is frankly astounding. Continue reading Find a significant effect in any study

7 small things you can do for science in 2017

The new year is here, and many of us start off with some resolutions. Following this trend, I thought it would be fun to share some things I’ve been doing and will continue to do that mostly (with the exception of point 3) require only little effort on my side and which positively impact my sciencing and that of those around me. Continue reading 7 small things you can do for science in 2017

R-Ladies: Coding sans prejudice

Recently, we (that is Page and Christina) successfully launched the Parisian installation of R-Ladies Global. It’s a meetup group and at the same time a non-profit coding club for all R proficiency levels, whether you’re a new or aspiring R user, or an experienced R programmer interested in mentoring, networking, and maybe picking up some new skills. We are a community designed to encourage, support and ultimately drive the development of our own R skills through a range of events, including meetups where members tackle hands-on tutorials and exercises to learn specific functionalities, informal gatherings, talks about latest trends, and debates. Our goal is to promote access to STEM (Science, Technology, Engineering, Mathematics) careers and tools for women (trans and cis) and gender-variant people. Men are welcome, too, by the way. We just need a member to bring them to the next meetup. In other words, we try to be a harassment-free zone. Sadly, that’s easier to do when men are screened beforehand.

Continue reading R-Ladies: Coding sans prejudice

R Course: Lesson 4

Today we’ll learn how to take an old statistics test (logistic regression) but expand it to when you have two variables (multiple regression). The package purrr is introduced and, as always, we’ll expand our knowledge of dplyr and ggplot2.

For full materials, see the course website for Lesson 4.

Here you can revisit Lesson 0, Lesson 1Lesson 2, and Lesson 3.

Continue reading R Course: Lesson 4

R Course: Lesson 2

Guest post by Page Piccinini

With some basics under our belt from Lesson 1, in this lesson we’ll continue working with dplyr and ggplot2, while also learning about the math behind linear regression and how to implement it in R. Plus you get to finish with a report about how the popularity of your name changes over time.

For full materials, see the course website for Lesson 2.

Here you can revisit Lesson 0 and Lesson 1.

Continue reading R Course: Lesson 2

Datasharing – yes, please! An attempt at a beginner’s guide

Science is becoming more and more open and transparent, and I think that’s awesome. An important aspect is sharing whatever information is necessary to reproduce results, usually that includes data and scripts. While open science can be beneficial for a researcher, this practice is still being met with some (justified) skepticism, but has become more and more accepted and common in research; in fact PLOS One for example made it a requirement for publication (how well that’s going is a different story). Funding agencies across the globe are quickly following suit, so chances are high you either already have to or will in the near future think about data sharing. But what does it entail? Continue reading Datasharing – yes, please! An attempt at a beginner’s guide