Author Topic: Improved Replication  (Read 865 times)

Jasonb4165

  • Full Member
  • ***
  • Posts: 12
    • View Profile
Improved Replication
« on: May 03, 2017, 03:20:19 PM »
I am using my replication from earlier in the semester as part of my argument in my paper, and while I got a very high grade on my replication, it is still not perfect with regards to the coefficients and the statistical significance levels. How can I make my statistical significance levels and coefficients closer to what they are in the paper?


Now that the replications are graded, can someone give tips to improve my replication?

Roric Brown

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 43
    • View Profile
Re: Improved Replication
« Reply #1 on: May 03, 2017, 06:07:31 PM »
No one has coefficients and significance levels that are completely perfect in their replication, so I would simply acknowledge that your replication was not perfect and move forward with your analysis from there.  The imprecision of your regression may matter more or less depending on what particular facet of the paper you are critiquing, as long as you are transparent with your original results, you should be ok.  Perhaps when you put the tables in your paper, include the original results of your replication so that the grader can specifically see what changes. 
Sorry we cannot give you any precise directions to yield a "perfect" replication, but as long as you include a full discussion and analysis you should be alright.

Jasonb4165

  • Full Member
  • ***
  • Posts: 12
    • View Profile
Re: Improved Replication
« Reply #2 on: May 04, 2017, 11:15:35 AM »
Can you give some tips none-the-less?

Scott Onestak

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 40
    • View Profile
Re: Improved Replication
« Reply #3 on: May 04, 2017, 08:21:24 PM »
Hey,


So, it's really hard to help with getting your replication closer to the researcher's values because we don't know what you have already done.  Also, nobody has the perfect solution.


What I would say is: if your replication was close enough to receive near full marks on the assignment, then it should be good enough to use on the critique.  Even if one or two of the coefficients is off, just ask yourself if that variable is an important variable in your critique.  If it isn't the subject of your paper, then I don't think it is that big of a deal.