Daniël LakensVerified account

@lakens

Psychologist . Stats blog: Free Coursera course:

Netherlands
sites.google.com/site/lakens2/H…
Joined November 2008
886 Photos and videos Photos and videos

Tweets

You blocked @lakens

Are you sure you want to view these Tweets? Viewing Tweets won't unblock @lakens

  1. Pinned Tweet
    6 Oct 2016

    My Coursera course "Improving Your Statistical Inferences" is live! I really hope you enjoy it!

    29 replies 319 retweets 669 likes
    Undo
  2. 13 hours ago

    Today a reviewer took great care to explain why they suggested a self-citation in a review. I did the same today in another decision letter. Good to see caution in self citations (the effect?). But also, sometimes we are actually experts, and have truly relevant work.

    7 replies 6 retweets 62 likes
    Undo
  3. 17 hours ago

    It is my belief that sometimes *not* naming observed ethical violations (either to an organization that checks ethical violations, such as Dutch LOWI, or publicly (with transparant evidence)) is itself an ethical violation. Not naming names is not automatically ethical behavior.

    Not going to name the journal here, but will definitely never consider submitting to it again.
    Show this thread
    4 replies 3 retweets 24 likes
    Undo
  4. Retweeted
    20 hours ago

    Redefine the statistical significance threshold to p ≤ 0.005

    2 replies 16 retweets 78 likes
    Undo
  5. Retweeted
    19 hours ago

    Our most stats-relevant episodes are 4, 25, 26, 42, 43, 46, 49, and 52. Listen or download here or subscribe to the show in your favourite podcast app 🎧

    Replying to @CogNerd @lakens @hertzpodcast
    I’m diving into quant this semester. Any episode recommendations focusing on stats?
    1 reply 12 retweets 32 likes
    Undo
  6. Retweeted
    May 3

    For the scientists looking for compelling podcasts with a current and irreverent edge, check out Good stuff with on statistics and academic hipsters. Highly recommended.

    1 reply 3 retweets 20 likes
    Undo
  7. Retweeted
    May 3

    is looking for one more Associate Director to help run this very exciting operation. If you're outside of North America/Europe, like open science and crowdsourcing, and want to work with myself, , , and , shoot Chris a message/email!

    7 retweets 7 likes
    Undo
  8. Retweeted
    May 3

    In response to some general twitter outrage about the QRP of sequentially peeking at analyses and stopping data collection after a significant result, I made a shiny app to simulate different scenarios. See just how much the false positive rate increases:

    4 replies 32 retweets 58 likes
    Show this thread
    Show this thread
    Undo
  9. May 3

    An excellent idea to to fund refugees by Dutch science funder . However, as noted by only people who already have an NWO grant (like me) can submit for *this* NWO grant (= Matthew effect). So great idea, but please make it open for *all* Dutch academics.

    NWO launches new “Refugees in Science” pilot. The aim is to fund one-year appointments for academics who have fled their home country and wish to continue their scientific career in the Netherlands.
    3 replies 11 retweets 38 likes
    Undo
  10. May 2

    Another excellent episode - what to do if your PhD supervisor does not value good research practices. At 95% of workshops where I teach (your?) PhD students, the first question is: How do I convince my supervisor we should improve the way we currently work? Let that sink in.

    New episode: "The PhD Who Caught the Car!" We talk about starting your first faculty job - getting ready and getting organized for your new role and responsibilities
    Show this thread
    10 replies 50 retweets 163 likes
    Undo
  11. May 2

    Sequential analyses with the goal to use the fewest number of animals is very ethical and efficient. But you need to do it correctly and control your error rate. My 2014 paper explains how to get started with sequential analyses - the right way:

    Wow. It's 2018 and you can still publish your QRP's in a Nature journal. How can something like this happen, ? (HT )
    Show this thread
    1 reply 13 retweets 61 likes
    Undo
  12. Retweeted
    May 2

    A massive undertaking. Who says you can't publish null effects in good journals?

    Really proud of our new paper finding no evidence that hormonal status influences women’s face preferences. Online now (open access with open data, stimuli, and code). Easily the largest study I’ve done, so it’s great to see it out.
    Show this thread
    3 replies 30 retweets 113 likes
    Undo
  13. Retweeted
    May 1

    At Prolific we're currently designing functionality that will let you collect representative samples easily and quickly. This would help make research more generalisable. Would you be interested in using such a tool? We'd love to hear your thoughts

    4 replies 47 retweets 72 likes
    Show this thread
    Show this thread
    Undo
  14. Retweeted
    May 1

    Could psychological societies please stop press releasing conference abstracts (eg )? Conference abstracts are not peer reviewed. They should not be used as bait for unnecessary press attention.

    4 replies 2 retweets 43 likes
    Undo
  15. May 1

    Look at this, (together with and ) organizing an early career meta-scientists meet-up (no seniors involved or invited!)! This kind of bottom-up collaboration makes me think the kids will be alright.

    Attending awesome meta-science meetup in Utrecht today, hearing about the projects of a bunch of fellow meta-research ECRs. Big thanks to and the other organizers for putting this together!
    1 reply 13 retweets 60 likes
    Undo
  16. Retweeted
    May 1

    On the flip side of Kaufmann's point, I left academia because I didn't trust the field of psychology to care more about telling than the truth than about the wounded feelings of already successful psychologists.

    Attributing bad intent to punishers is common, see this great article "The reputation of punishers" It is also why so few people speak out when they see bad behavior in academics - their attempts to improve science will be seen as jealousy 1/2
    Show this thread
    1 reply 16 retweets 82 likes
    Undo
  17. Retweeted
    Apr 30

    Got an idea for a study that merits large-scale collaborative data collection? Consider submitting a proposal to the Psychological Science Accelerator. Proposals are due by June 20, 2018. For details, see (CC: )

    16 retweets 30 likes
    Undo
  18. Retweeted
    Apr 30
    Replying to and

    I think paper authorship should move towards a "credits" model, just like the one that was adopted by movies 100 years ago. Nowadays, it is rare that the director is *also* acting and *also* shooting the piece. Why expect this of us? Collaboration - properly rewarded - is key.

    2 retweets 10 likes
    Undo
  19. Apr 30

    The screenshot below predicts the response by . I understand it hurts to see a long-time collaborator criticized. Punishment is very messy (I wish it wasn't, but it is). But there is no need to accuse people who try to fix science of being jealous. 2/2

    3 replies 3 retweets 22 likes
    Show this thread
    Show this thread
    Undo
  20. Apr 30

    Attributing bad intent to punishers is common, see this great article "The reputation of punishers" It is also why so few people speak out when they see bad behavior in academics - their attempts to improve science will be seen as jealousy 1/2

    In seeing the people who are leading the witch hunt against Bob Sternberg, it makes me wonder what motivates investing so much time to try and hurt others -- jealousy? boredom? People wonder why people hate academia. I think it's not too hard to figure out why. Sometimes I do.
    3 replies 19 retweets 56 likes
    Show this thread
    Show this thread
    Undo
  21. Retweeted
    Apr 30

    Finished Coursera course today. Such an awesome (and free!) resource. Highly recommend it!

    1 reply 2 retweets 28 likes
    Undo

@lakens hasn't Tweeted yet.

Loading seems to be taking a while.

Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.

    You may also like

    ·