Close

Complete Captcha

• 3 weeks - 206 posts - Latest - RSS
• Thread: 410 Goods vs 163 No Goods

1. Economist
   76b5
   

   From Institute for Replication's Twitter account:

   At our first Replication Games in Oslo, Montero (2022, JPolEcon) was assigned to Kjelsrud et al.

   In their new DP (accepted in JPE) Kjelsrud et al. discuss issues with one of the two set of results in Montero (2022). They find that “three-quarters of the observations are duplicates and income inequality is calculated over too few workers to be meaningful.”

   We [Institute for Replication] provide detailed information below [thread] on the replication, our rep games & the editorial decision.

   3 weeks ago # QUOTE 11 Volod 0 Vlad !

2. Economist
   bba5
   

   Doesn’t seem like the replicators are alleging fraud.

   3 weeks ago # QUOTE 5 Volod 4 Vlad !

3. Economist
   76b5
   

   75% duplicates and results are driven by those duplicates. Seems from the DP that this is obvious when opening data editor in stata. See Figure 2 in their DP:
   https://www.econstor.eu/bitstream/10419/268943/1/I4R-DP020.pdf

   3 weeks ago # QUOTE 0 Volod 0 Vlad !

4. Economist
   f9df
   

   Interesting! Good job (not) replicating this, sad state of affairs

   3 weeks ago # QUOTE 1 Volod 0 Vlad !

5. Economist
   c9a4
   

   Chicago Faculty and JPE. What else did you expect?

   3 weeks ago # QUOTE 2 Volod 3 Vlad !

6. Economist
   a567
   

   What are the consequences for this?

   If one of my students was caught doing this, they'd be kicked from the program.

   3 weeks ago # QUOTE 0 Volod 1 Vlad !

7. Economist
   1d5c
   

   Doesn’t seem like the replicators are alleging fraud.

   The sample restriction explicitly stated in the text also was not done, which was crucial to the significance of the results.

   3 weeks ago # QUOTE 0 Volod 0 Vlad !

8. Economist
   9446
   

   How can this happen as a honest mistake?

   75% duplicates and results are driven by those duplicates. Seems from the DP that this is obvious when opening data editor in stata. See Figure 2 in their DP:
   https://www.econstor.eu/bitstream/10419/268943/1/I4R-DP020.pdf

   3 weeks ago # QUOTE 1 Volod 0 Vlad !

9. Economist
   cb67
   

   This is so egregious that if action is not taken, what reputation should U Chicago or J Pol Econ have?

   3 weeks ago # QUOTE 0 Volod 1 Vlad !

10. Economist
    6519
    

    What does DP mean here?

    3 weeks ago # QUOTE 0 Volod 0 Vlad !

11. Economist
    1d5c
    

    Wow, Montero really trying to drown this. Bumping this thread, even with unrelated posts about sexual harrassment, doesn't help you.

    3 weeks ago # QUOTE 0 Volod 1 Vlad !

12. Economist
    04e1
    

    Karrrrllll

    3 weeks ago # QUOTE 0 Volod 1 Vlad !

13. Economist
    648e
    

    HRM execution, different quality

    3 weeks ago # QUOTE 0 Volod 0 Vlad !

14. Economist
    c43c
    

    Harvard development crowd. Same old cr@p. Sexyness ahead of science

    3 weeks ago # QUOTE 5 Volod 1 Vlad !

15. Economist
    bcb1
    

    DP means discussion paper. Seems like Institute for Replication has its own discussion paper series.

    3 weeks ago # QUOTE 0 Volod 0 Vlad !

16. Economist
    6519
    

    DP means discussion paper. Seems like Institute for Replication has its own discussion paper series.

    Thanks!

    3 weeks ago # QUOTE 0 Volod 0 Vlad !

17. Economist
    cb67
    

    lol @ the duplicates. professionally written but a clear middle finger between the lines.

    3 weeks ago # QUOTE 0 Volod 0 Vlad !

18. Economist
    6874
    

    LRMs that cannot write papers spend time on meaningless comments. Boring

    3 weeks ago # QUOTE 2 Volod 13 Vlad !

19. Economist
    35c3
    

    How can this happen as a honest mistake?

    75% duplicates and results are driven by those duplicates. Seems from the DP that this is obvious when opening data editor in stata. See Figure 2 in their DP:
    https://www.econstor.eu/bitstream/10419/268943/1/I4R-DP020.pdf
    

    To me, figure 2 is strong evidence of sloppiness rather than fraud. If you tried to cheat, the last thing you would do is to generates hundreds of identical copies. However, it could well happen if you don't bother looking at the data and just start regressing stuff. If I had to guess: a bad 1:m merge that gave all these units identical values.

    3 weeks ago # QUOTE 10 Volod 1 Vlad !

20. Economist
    92e3
    

    LRMs that cannot write papers spend time on meaningless comments. Boring

    RIP, Montero

    3 weeks ago # QUOTE 0 Volod 0 Vlad !

Reply »