Ashley Holub

@ashtroid22

Epidemiology PhD student, using techno and tacos to get through training. // she/her

Rochester, NY
Vrijeme pridruživanja: studeni 2011.

Tweetovi

Blokirali ste korisnika/cu @ashtroid22

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @ashtroid22

  1. Prikvačeni tweet
    15. kol 2019.

    GUYS I WROTE THIS! check it out and let me know what you think. It's my favorite piece to date

    Poništi
  2. prije 10 sati

    All of these account for the survey design. For the modified poisson and log binomial this was through nesting of the strata and the clusters in the repeated statements

    Prikaži ovu nit
    Poništi
  3. prije 10 sati

    As an additional check I actually created a constant variable and used the proc surveyphreg to see the HR which is HR: 1.10 So logistic and log binomial match and correspond with survey features Modified poisson and cox are close and do not match with proc survey logistic

    Prikaži ovu nit
    Poništi
  4. prije 10 sati

    Alright stats people can someone explain what is happening here (using a sandwich estimator to correct for survey design) Crude logistic OR: 0.65 Crude survey logistic OR: 0.64 Crude log binomial (robust variances/sandwhich est) RR:0.70 Crude poisson (robust): RR 1.05

    Prikaži ovu nit
    Poništi
  5. prije 11 sati

    Today I am wearing a black turtle neck again, but with balloon sleeves So I am basically Steve Jobs again, but trendy

    Poništi
  6. proslijedio/la je Tweet
    2. velj

    So is here a part of a thing I'm working on that I'm sharing a bit prematurely, largely to help myself find typos and just to feel like I've "done" something. Any comments are very welcome.

    Prikaži ovu nit
    Poništi
  7. proslijedio/la je Tweet
    2. velj

    No more screen time, kids. It’s time to watch a 5-hour football game.

    Poništi
  8. prije 20 sati
    Poništi
  9. 3. velj

    Wow was not expecting this split

    Prikaži ovu nit
    Poništi
  10. 2. velj

    Reducing bias in a Risk ratio:

    Prikaži ovu nit
    Poništi
  11. 2. velj

    Can someone catch me up here, I thought we stopped naming viruses after places?

    Poništi
  12. proslijedio/la je Tweet
    1. velj

    This is going gently science-Twitter viral (I'd say an R0 of like 0.5) so I will now plug the fun, award-winning daily science podcast I co-host:

    Prikaži ovu nit
    Poništi
  13. 1. velj

    Psycho metrics has me feeling like the world is doomed

    Poništi
  14. 1. velj

    * Feel like I should mention...this measure was developed for longitudinal data where they are looking for within-subject change, thus the 0-3 bounds don't matter for that context. However, looking at the validity for a cross-sectional study is where I am stuck

    Prikaži ovu nit
    Poništi
  15. 1. velj

    I guess what I'm trying to get at is "every one unit increase in X, is assoc with a xx unit increase in Y..." wouldn't necessarily work here? Or am I thinking too hard? any thoughts?

    Prikaži ovu nit
    Poništi
  16. 1. velj

    so divided score again becomes 0-3. So my question is what plugging that domain score into a linear regression (as the DV) will do? Does the division throw off the whole interpretation? Because you’re dividing it’s become more of a percent rather then points?

    Prikaži ovu nit
    Poništi
  17. 1. velj

    to get the domain score. For example: domain 1 can have a score ranging from 0-18 in the numerator if you completed all the items then the denominator is 6 but if you completed half, then it’s 3. Therefore, the score range becomes 0-3. Domain two has 7 questions, max of 21 pets

    Prikaži ovu nit
    Poništi
  18. 1. velj

    Okay I’m stumped. I’m working a measure that has multiple domains, different number of questions per domain. To make domains directly comparable and account for missing the scoring divides the score for the domain/the number completed for the domain

    Prikaži ovu nit
    Poništi
  19. proslijedio/la je Tweet
    1. velj

    “Here comes the smolder” ~ Arthur Arthur has been crowned the Bestest Boy Ever & our staff can’t stop gushing about him. He’s so friendly & well-behaved, knows “sit” and “down, and seems to be housebroken. Visit the shelter today from 12-4PM to make Arthur your new best friend!

    Poništi
  20. proslijedio/la je Tweet
    1. velj

    Tweeps, I had the most wonderful evening hanging out with superstars & 🤩!! We had lots of laughs, plus I got advice on how to decide when to say ‘yes’ or ‘no’ that’s so great I had to make a graphic to share with y’all! 🥳🥰

    Poništi
  21. 1. velj

    To my earlier point last week, about discussion of the risks and benefits... don’t forget the do nothing chat! This is how you do it

    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:

    ·