Jack Wilkinson

@jd_wilko

Centre for Biostatistics. Statistical Editor for Stats methods for & . . Tweets do not reflect employer opinion.

Manchester
Vrijeme pridruživanja: srpanj 2013.

Tweetovi

Blokirali ste korisnika/cu @jd_wilko

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @jd_wilko

  1. proslijedio/la je Tweet
    prije 13 sati
    Odgovor korisniku/ci

    If you want an "accept without revisions" then go for this: 😀😀

    Poništi
  2. proslijedio/la je Tweet
    Odgovor korisnicima

    "Theme to replicate the ugly monstrosity that was the old gray-background Excel chart. Please never use this."

    Poništi
  3. proslijedio/la je Tweet
    prije 13 sati
    Odgovor korisnicima i sljedećem broju korisnika:

    There is also a thing which tries to turn R into SPSS.

    Poništi
  4. proslijedio/la je Tweet
    Odgovor korisnicima

    ...and someone has made a ggplot2 theme to look like Stata

    Poništi
  5. proslijedio/la je Tweet

    My take on Trust in Algorithms in Harvard Data Science Review: "we need to be able to assess the trustworthiness of claims made both about an algorithm, and those made by an algorithm"

    Poništi
  6. proslijedio/la je Tweet
    prije 14 sati
    Odgovor korisniku/ci

    Use jerzpals: ! Inspired by amazing wardrobe.

    Poništi
  7. proslijedio/la je Tweet

    Teaching focussed lecturer post at Manchester. Reach out if you'd like to know more!

    Poništi
  8. proslijedio/la je Tweet
    prije 15 sati

    I’ve been told that yes, my (correct) analysis was right but the reviewers would’t understand it so change it. I was even told once to drop an age squared term because it didn’t make biological sense. Being a statistician in medical research could be infuriating.

    Poništi
  9. prije 15 sati
    Poništi
  10. prije 15 sati

    We all know that using the default grey backgrounds in Stata plots is a surefire way to make yourself look like a dork. What about if I use default ggplot2 plotting colours? Am I going to make myself look *like a bad dork*

    Poništi
  11. proslijedio/la je Tweet
    prije 17 sati
    Odgovor korisniku/ci

    I used to think this was a slightly cynical viewpoint, until I saw it first hand with this:

    Poništi
  12. proslijedio/la je Tweet
    Odgovor korisniku/ci

    You haven't tried Advanced Bayesian Imputation where you replace a missing value by the value that maximizes your subjective belief

    Poništi
  13. proslijedio/la je Tweet
    prije 20 sati
    Poništi
  14. proslijedio/la je Tweet
    prije 18 sati

    thought of the day: one cannot pretend to be interested in if one doesn't want to adjust for covariates in a randomized trial.

    Poništi
  15. prije 18 sati

    If you thought mean imputation for missing data was bad, let me introduce you to modal imputation for categorical variables.

    Poništi
  16. proslijedio/la je Tweet
    prije 20 sati
    Odgovor korisniku/ci

    I reviewed report of RCT recently, dynamite plots included. Recommended use better graph (included link below on what to do), response: these plots are what researchers in our area expect, so we refuse to change it.

    Poništi
  17. prije 20 sati

    Time to event analysis is very poorly understand in infertility research at present. In general, if you reduce the likelihood of an event occurring, you tend to increase the time to that event.

    Poništi
  18. proslijedio/la je Tweet
    Odgovor korisniku/ci

    As one statistician at a journal put it to me, stats education in clinical research too often turns out to be copying the methods of a similar study that got published.

    Poništi
  19. proslijedio/la je Tweet
    prije 21 sat
    Odgovor korisniku/ci

    Sharing this pain Jack...

    Poništi
  20. prije 21 sat

    People are very happy to adopt treatments without proper RCTs. But when an RCT suggests a treatment they use does not work, they suddenly become very interested in methodological rigour.

    Poništi

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·