forked from Stat585-at-ISU/blog-2019
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathGaniAgadilov.Rmd
21 lines (13 loc) · 2.7 KB
/
GaniAgadilov.Rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
---
title: "Ethics and Reproducibility..."
author: "Gani Agadilov"
topic: "03"
layout: post
root: ../../../
---
Write a blog post addressing the questions:
1. **Pick one of the papers Retraction Watch features on their website and describe what went wrong**.
I chose the paper of a group of researchers that retracted from The Journal of Rheumatology. The researchers from Egypt have conducted a scientific study where the several issues had occurred. The scientists tried to conduct the experimental study where the control group had the same baseline characteristics as a treatment group. In addition, the researchers had statistical errors in some steps of the organization part of the study. The randomization part of the experimental units stated that 125 subjects randomized on a single day and enrolled at the same time. Moreover, the dataset in one table had taken from other previous paper that published in 2014. This was not the only reason to retract the paper. The researchers did not follow the ethical aspect of conducting the experimental research. The registration of the study at the website was June 12, 2012. It was a day when the study was finished. Overall, the paper had a large number of errors that violate not the only standard of publishing the research paper but also disregard the principles of the general scientific norms.
2. **After reading the paper by Sandve et al. describe which rule you are most likely to follow and why, and which rule you find the hardest to follow and will likely not follow in your future projects.**
These are very useful rules that I would like to follow while I am conducting computational research. However, it seems to me that Rule 1 and 4 are one of the most important rules that I want to take into account. The reason that Rule 1 is valuable because you should be able to interpret and record the output in such a way that you could reproduce the result after a certain amount of time. Paying attention to the details and following each sequence of the script which allows reproducing the results for a small part of the analysis is the critical part of Rule 1. In addition, Rule 4 is one of the most convenient rules that allow you to keep track of the evolution of the team based research project. The control version and remote version of the project give a more reliable tool to share, develop and combine the script files from different sources.
From my point of view, Rule 5 is difficult to follow. We can save the outcomes for each part of the analysis that allows us to monitor the overall draft of the project. However, standardizing results might be time-consuming. Therefore, it would be helpful to save the outcome of the processes in the accessible and easily tractable format.