Close

Complete Captcha

• 2 years - 4567 posts - Latest - RSS
• Thread: 6908 Goods vs 3026 No Goods

1. Economist
   bb84
   

   No one has ever disclosed this info, except for PD. She had suspicious papers and suspicious data in a PS paper, so she was being "monitored", let's say. But yes, she basically took the surveys herself in the lab (probably so the IP numbers come from inside the university lab). It's a pretty dumb move considering there's also timestamps that can be verified in the data.

   Does anyone know exactly what sorts of data anomalies were in papers by Ping Dong and/or Nicole Coleman? In Lawrence Sanna's case, I read that the means and standard deviations were pretty much identical across conditions, indicative of copying and pasting with minor changes (read: fabrication) of data to make it look somewhat different. In other fields, people have been caught copying cells directly without any changes. How were Ping and Nicole caught and what were the data anomalies? Anyone know? Insider?

   4 months ago # QUOTE 0 Volod 0 Vlad !

2. Economist
   5ad1
   

   So I had my PhD student e-mail LN for data for his superhero priming paper. LN responded kindly to her that his data "has been lost and not retrievable." Hypocrites.

   4 months ago # QUOTE 5 Volod 0 Vlad !

3. Economist
   bb84
   

   Hahahahahaha

   Soulja Boy off in this ho
   Watch me crank it, watch me roll
   Watch me crank dat Soulja Boy
   Then Superman that ho
   Now watch me yuuuuuuu (Crank dat Soulja Boy)
   Now watch me yuuuuuuu (Crank dat Soulja Boy)
   Now watch me yuuuuuuu (Crank dat Soulja Boy)
   Now watch me yuuuuuuu (Crank dat Soulja Boy)

   4 months ago # QUOTE 1 Volod 0 Vlad !

4. Economist
   0c3a
   

   Partially correct.

   No one has ever disclosed this info, except for PD. She had suspicious papers and suspicious data in a PS paper, so she was being "monitored", let's say. But yes, she basically took the surveys herself in the lab (probably so the IP numbers come from inside the university lab). It's a pretty dumb move considering there's also timestamps that can be verified in the data.

   Does anyone know exactly what sorts of data anomalies were in papers by Ping Dong and/or Nicole Coleman? In Lawrence Sanna's case, I read that the means and standard deviations were pretty much identical across conditions, indicative of copying and pasting with minor changes (read: fabrication) of data to make it look somewhat different. In other fields, people have been caught copying cells directly without any changes. How were Ping and Nicole caught and what were the data anomalies? Anyone know? Insider?
   

   4 months ago # QUOTE 0 Volod 0 Vlad !

5. Economist
   bb84
   

   Pls tell what's incorrect and what's remaining

   4 months ago # QUOTE 0 Volod 0 Vlad !

6. Economist
   0c3a
   

   No one was monitoring Ping. Sleuths on Twitter just found a lot of fckery in her data, got the PS paper retracted, and it all unraveled after that.

   Pls tell what's incorrect and what's remaining

   4 months ago # QUOTE 0 Volod 0 Vlad !

7. Economist
   3ad4
   

   Did LN manipulate or even fake data for his superhero paper? I mean, finding something to be significant after 90 days for a priming manipulation, it's very unlikely that p hacking would even work. i bet you his results are fraudulent and that's why he won't share his data.

   4 months ago # QUOTE 2 Volod 0 Vlad !

8. Economist
   bb84
   

   But it's just a binary choice thing. Even if he faked, he would never get caught. It's not like there's a lot of irregularities that can be found (or inferred) based on comparing 2 proportions.

   4 months ago # QUOTE 0 Volod 0 Vlad !

9. Economist
   5ad1
   

   So moral of story, if people want to fabricate data, use binary choice as outcome. No one can "detect" irregular means and distributions because there aren't any!

   4 months ago # QUOTE 0 Volod 0 Vlad !

10. Economist
    5ad1
    

    Aaron Charlton is onto Leif's data, according to his twitter.

    4 months ago # QUOTE 0 Volod 1 Vlad !

11. Economist
    c710
    

    Hi this is Aaron Charlton. Super Hero priming looks like p-hacking to me. P-hacking is so common in marketing that it's crazy that people would go after somebody so hard for p-hacking 20 years ago. Leif Nelson was recently an outside expert on a data anomalies case and I have reason to believe the anomalist is attempting to defame him and me using multiple anonymous accounts. If everyone could ignore that for a while and wait patiently for the big reveal that'd be great.

    4 months ago # QUOTE 3 Volod 2 Vlad !

12. Economist
    562e
    

    So moral of story, if people want to fabricate data, use binary choice as outcome. No one can "detect" irregular means and distributions because there aren't any!

    I mean if you were really going to look at it I guess you’d look for anomalies in the tetrachoric corrrlations with other measures and weird issues with covariates but I haven’t though that hard about it.

    4 months ago # QUOTE 0 Volod 0 Vlad !

13. Economist
    a4e7
    

    fu

    Great paper

    Suspect paper:
    Andrews, Michelle, Xueming Luo, Zheng Fang, and Anindya Ghose. "Mobile ad effectiveness: Hyper-contextual targeting with crowdedness." Marketing Science 35, no. 2 (2016): 218-233.

    their code is available on the MNKSC website. Try running it and replicate the results from the paper. You cannot. But, MKSC does not care. Because XL got a lot of papers with the INFORMS brass, they don't want to embarrass themselves.
    

    moron

    FUM

    4 months ago # QUOTE 0 Volod 0 Vlad !

14. Economist
    bd88
    

    It’s not fair but people should be realistic about old data. 25 years ago was a different world - hard drives crashed and the cloud wasn’t a thing.

    I remember CB students collecting data on paper surveys and then entering the data into excel. Was any of that real?

    So I had my PhD student e-mail LN for data for his superhero priming paper. LN responded kindly to her that his data "has been lost and not retrievable." Hypocrites.

    4 months ago # QUOTE 0 Volod 0 Vlad !

15. Economist
    562e
    

    I carried boxes of paper around for 8 years across two jobs

    It’s not fair but people should be realistic about old data. 25 years ago was a different world - hard drives crashed and the cloud wasn’t a thing.
    I remember CB students collecting data on paper surveys and then entering the data into excel. Was any of that real?

    So I had my PhD student e-mail LN for data for his superhero priming paper. LN responded kindly to her that his data "has been lost and not retrievable." Hypocrites.
    

    4 months ago # QUOTE 2 Volod 0 Vlad !

16. Economist
    39c4
    

    Why do people (here and elsewhere) vilify authors with anomalies in their data? Obviously, there are some cases that show clear fraud. But people here seem to vilify even those whose data is just "weird." In the other sciences like chemistry and physics, people have retracted papers because their data were (shown/proven by others to be) weird. Yet, these people aren't vilified. Some in fact go on to win Nobel prizes. The social sciences is wonky. Data is weird. LN likes to look at last digits, which should be random, but in one or two cases, it's bound to be non-randomly distributed. Many of these cases, there is no fraud (at least not until it's proven). It's just how it is. Other fields accept retractions as normal (unless, again, it's fraud). But not CB. We automatically believe DA to be a fraud when data is just plain weird. Grow up.

    4 months ago # QUOTE 0 Volod 2 Vlad !

17. Economist
    562e
    

    Why do people (here and elsewhere) vilify authors with anomalies in their data? Obviously, there are some cases that show clear fraud. But people here seem to vilify even those whose data is just "weird." In the other sciences like chemistry and physics, people have retracted papers because their data were (shown/proven by others to be) weird. Yet, these people aren't vilified. Some in fact go on to win Nobel prizes. The social sciences is wonky. Data is weird. LN likes to look at last digits, which should be random, but in one or two cases, it's bound to be non-randomly distributed. Many of these cases, there is no fraud (at least not until it's proven). It's just how it is. Other fields accept retractions as normal (unless, again, it's fraud). But not CB. We automatically believe DA to be a fraud when data is just plain weird. Grow up.

    The probability that the insurance data is “real” and not fabricated is basically nil. The only question is who is responsible. And if we ever get to the bottom of it, they deserve to be vilified.

    As for other fields not vilifying fakers and frauds. The people who flecked with photoshop to publish Alzheimer’s papers are being vilified. Anil Potti, vilified. The guy who faked the contact hypothesis is totally vilified. There are whole fields in electrical engineering where no one will cite Chinese papers for fear the data are fake. There’s a difference between weird data and fraud and other fields definitely anathematize frauds.

    4 months ago # QUOTE 2 Volod 0 Vlad !

18. Economist
    bb84
    

    What is the big reveal? Pls

    I'm curious and looking forward to the next retraction lol

    4 months ago # QUOTE 0 Volod 0 Vlad !

19. Economist
    39c4
    

    Why do people (here and elsewhere) vilify authors with anomalies in their data? Obviously, there are some cases that show clear fraud. But people here seem to vilify even those whose data is just "weird." In the other sciences like chemistry and physics, people have retracted papers because their data were (shown/proven by others to be) weird. Yet, these people aren't vilified. Some in fact go on to win Nobel prizes. The social sciences is wonky. Data is weird. LN likes to look at last digits, which should be random, but in one or two cases, it's bound to be non-randomly distributed. Many of these cases, there is no fraud (at least not until it's proven). It's just how it is. Other fields accept retractions as normal (unless, again, it's fraud). But not CB. We automatically believe DA to be a fraud when data is just plain weird. Grow up.

    The probability that the insurance data is “real” and not fabricated is basically nil. The only question is who is responsible. And if we ever get to the bottom of it, they deserve to be vilified.
    As for other fields not vilifying fakers and frauds. The people who flecked with photoshop to publish Alzheimer’s papers are being vilified. Anil Potti, vilified. The guy who faked the contact hypothesis is totally vilified. There are whole fields in electrical engineering where no one will cite Chinese papers for fear the data are fake. There’s a difference between weird data and fraud and other fields definitely anathematize frauds.

    Read what I said. I said if there's fraud, then yes vilify them. But if there are just weird data, that's not (necessarily) fraud and the researchers should not be vilified.

    4 months ago # QUOTE 0 Volod 2 Vlad !

20. Economist
    bb84
    

    I really wanted to know how to find some anomalies (and what do these mean). I have some suspicions regarding 2 recent JCR papers. Is it about finding nonsensical correlations among variables in the dataset? How can I learn more?

    4 months ago # QUOTE 0 Volod 1 Vlad !

Reply »