Class Journal Week 8 - Revision history

Ebachour: adding my responsees

2017-10-24T07:00:50Z

adding my responsees

Dbashour: added dina's responses

2017-10-24T06:59:14Z

added dina's responses

Nicolekalcic: added Nicole Kalcic

2017-10-24T06:52:00Z

added Nicole Kalcic

Hhinsch: added my answers to the class journal

2017-10-24T06:49:18Z

added my answers to the class journal

ArashLari at 06:48, 24 October 2017

2017-10-24T06:48:41Z

Aporras1: /* Aporras1 Response */ response

2017-10-24T06:48:18Z

‎Aporras1 Response: response

ArashLari: setting up my response

2017-10-24T06:38:29Z

setting up my response

Simonwro120: finished journal

2017-10-24T06:34:07Z

finished journal

Johnllopez616: Added my responses (John Lopez)

2017-10-24T06:26:22Z

Added my responses (John Lopez)

Simonwro120: answered 1 and 2

2017-10-24T06:20:16Z

answered 1 and 2

← Older revision		Revision as of 07:00, 24 October 2017
Line 147:		Line 147:
	[[User:Dbashour\|Dbashour]] ([[User talk:Dbashour\|talk]]) 23:59, 23 October 2017 (PDT)		[[User:Dbashour\|Dbashour]] ([[User talk:Dbashour\|talk]]) 23:59, 23 October 2017 (PDT)
	{{Template:dbashour}}		{{Template:dbashour}}
		+
		+	==Eddie Bachoura's Responses==
		+	# The biggest issue with the data analysis was that Baggerly and Coombs had data that was not reproducible. Also, their data was not clearly labeled. Both of these were practices that DataONE believes to be best practices, and both were violated. Baggerly agrees with the fact that his data wasn't documented well was a big issue.
		+	# Dr. Baggerly recommends that, when working with data analysis, one should very strictly follow these protocols, such as labeling data, documenting the process well, and most importantly have a reproducible workflow. DataOne agrees with all three of these, with more focus on the reproducibility aspect.
		+	# This week, we followed the best practices of documenting our data well as well as ensuring that our results were reproducible by others.
		+	# Same reactions as before, but it helps me understand a bit about the professional biological world and how important it is to follow these protocols.
		+	[[User:Ebachour\|Ebachour]] ([[User talk:Ebachour\|talk]]) 00:00, 24 October 2017 (PDT)

← Older revision		Revision as of 06:59, 24 October 2017
Line 134:		Line 134:

	[[User:Nicolekalcic\|Nicolekalcic]] ([[User talk:Nicolekalcic\|talk]]) 23:51, 23 October 2017 (PDT)		[[User:Nicolekalcic\|Nicolekalcic]] ([[User talk:Nicolekalcic\|talk]]) 23:51, 23 October 2017 (PDT)
		+
		+	==Dina Bashoura's Responses==
		+
		+	#What were the main issues with the data and analysis identified by Baggerly and Coombs? What best practices enumerated by DataONE were violated? Which of these did Dr. Baggerly claim were common issues?
		+	#* The main issue was probably the fact that data could not be reproduced. Beggarly and Coombs found errors, discrepancies, duplicate data, and poor documentation. The common issues Dr. Baggerly claimed surrounded the mislabeling of data.
		+	#What recommendations does Dr. Baggerly recommend for reproducible research? How do these correspond to what DataONE recommends?
		+	#* Dr. Baggerly recommended labeling data, documenting all the processes, and using templates to have reproducible data. DataONE recommends proper labeling as well to allow for clear, reproducible data.
		+	#What best practices did you perform for this week's assignment?
		+	#* This week, we performed the best practice of labeling the data. We highlighted every new section of data a different color to have clear and easily distinguishable columns. We also documented the process in our electronic notebooks as well as used a template, as we do every week.
		+	#Do you have any further reaction to this case after viewing Dr. Baggerly's talk?
		+	#* After completing this week's assignment, I can see how tricky it is to organize all the data and realized how easy it is to miss over incorrect data. But this video clearly shows how many discrepancies were found and how the university still overlooked these findings. I can understand a few mistakes when working with large data sets like the one we worked with this week, but to find huge discrepancies in data shouldn't have been overlooked.
		+	[[User:Dbashour\|Dbashour]] ([[User talk:Dbashour\|talk]]) 23:59, 23 October 2017 (PDT)
		+	{{Template:dbashour}}

← Older revision		Revision as of 06:52, 24 October 2017
Line 124:		Line 124:
	[[User:Hhinsch\|Hhinsch]] ([[User talk:Hhinsch\|talk]]) 23:49, 23 October 2017 (PDT)		[[User:Hhinsch\|Hhinsch]] ([[User talk:Hhinsch\|talk]]) 23:49, 23 October 2017 (PDT)
	{{hhinsch}}		{{hhinsch}}
		+
		+
		+	==Nicole Kalcic's Responses==
		+
		+	#The biggest issue with the data was that it was not able to be reproduced. After this week's assignment and readings, I feel like I have a much greater understanding on the importance of our electronic journals. If someone is not able to see exactly what steps I took to get my final product, then my data will not be able to be verified (making it somewhat useless). Several best practices enumerated by DataOne were broken. They include the lack of workflow documentation I just mentioned and a lack of labeling techniques. Dr. Baggerly claimed that improper labeling was the most common mistake.
		+	#Following that, it would be easy to guess that Dr. Baggerly recommends properly and carefully labeling for reproducible research. Again, DataONE has labeling as one of the best practices, meaning that Baggerly and DataONE are both strongly emphasizing the same thing.
		+	#This week, we used clear labeling on our data sets. We had consistent data entry and reproducible research.
		+	#I feel like the case is more shocking for me at this point, because it has been drilled into our class that our work needs to have a journal and that it needs to be checked/verified. How could so many professionals miss that in regards to the work behind the case?
		+
		+	[[User:Nicolekalcic\|Nicolekalcic]] ([[User talk:Nicolekalcic\|talk]]) 23:51, 23 October 2017 (PDT)

← Older revision		Revision as of 06:49, 24 October 2017
Line 115:		Line 115:
	[[User:ArashLari\|ArashLari]] ([[User talk:ArashLari\|talk]]) 23:38, 23 October 2017 (PDT)		[[User:ArashLari\|ArashLari]] ([[User talk:ArashLari\|talk]]) 23:38, 23 October 2017 (PDT)
	{{template:ArashLari}}		{{template:ArashLari}}
		+
		+
		+	== Hayden Hinsch's Responses ==
		+	#The main issues with the data and analysis identified by Beggerly and Coombs is that the data was not reproducible. They found many simple mistakes, mentioning that simple mistakes are the most common and that clear documentation of statistical analysis could be a way to fix this. The practices that were violated were that the data was not easily reproducible, the data had different names within the table, and data was not consistent. Baggerly claimed that the fact that the data was not well documented was a common issue. The indexes were off by one, and when reproducing the calculations, most of the heat maps were not even similar.
		+	#Baggerly recommends that everything is extremely well documented. This will help with the reduction of simple errors. This corresponds with the DataONE recommendations but DataONE specifies that all of the data should be on one table.
		+	#This week we created a description of the statistical analysis that we performed so that someone else may reproduce the results we obtained. We also maintained descriptive table names and had a singular table with multiple sheets for all of our data.
		+	#I am not very surprised that so many simple mistakes were made. It seems that humans make simple mistakes all the time, so it makes sense that it would happen. I am glad I am learning about this, but am sad that the habit was not formed within me much earlier in life.
		+	[[User:Hhinsch\|Hhinsch]] ([[User talk:Hhinsch\|talk]]) 23:49, 23 October 2017 (PDT)
		+	{{hhinsch}}

@@ Line 109: / Line 109: @@
 [[User:Johnllopez616|Johnllopez616]] ([[User talk:Johnllopez616|talk]]) 23:26, 23 October 2017 (PDT)
 == Arash Lari's Responses ==
+# The biggest problem that they had, which is a big sin in regards to the scientific method, is the fact that they couldn't reproduce their results, which is a big no-no in scientific research. They also mislabeled, didn't name stuff clearly. Another big instance of dishonesty and inaccuracy was the manipulation of data.
+# Dr. Baggerly recommends proper labeling and ensuring that the experiment is reproducible. DataONE also heavily emphasizes doing these things.
+# The whole point of this assignment was to create a reproducible step by step process and it revolved around proper labeling and regular saving and communication.
+# I'm still surprised that the original event happened, regardless of watching the talk, because I feel as though much of this information was common sense for acquiring legitimate data. Even though Dr. Baggerly made it seem like it should be obvious to spot phony data, I think the fact that they weren't able to is proof that it's not that easy. I would say that I'm most surprised that the country's government doesn't establish a part of the government for the advancement of science that could unbiasedly and objectively review research from different labs.
 [[User:ArashLari|ArashLari]] ([[User talk:ArashLari|talk]]) 23:38, 23 October 2017 (PDT)
 {{template:ArashLari}}

@@ Line 79: / Line 79: @@
 ==Aporras1 Response==
-# What were the main issues with the data and analysis identified by Baggerly and Coombs? What best practices enumerated by DataONE were violated? Which of these did Dr. Baggerly claim were common issues?
+#What were the main issues with the data and analysis identified by Baggerly and Coombs? What best practices enumerated by DataONE were violated? Which of these did Dr. Baggerly claim were common issues? They identified that a lot of the data wasn't reproducible and multiple aspects of the research was mislabeled and even unidentifiable. The practices which were violated was incomplete documentation, mixing up data labeling, and overall flaws in experimental design. The most common mistakes were the most simple mistakes.
-# What recommendations does Dr. Baggerly recommend for reproducible research? How do these correspond to what DataONE recommends?
+#What recommendations does Dr. Baggerly recommend for reproducible research? How do these correspond to what DataONE recommends? He recommends literate programming, reusable templates, report structure, executive summaries and appendices. Compared to DataONE, they recommend entering complete lines of data, validation of data as it's being entered and ultimately they both agree on reproducible workflow.
-# What best practices did you perform for this week's assignment?
+#What best practices did you perform for this week's assignment? From this assignment, making sure my work was reproducible using the electronic notebook was essential along with always ensuring the data I was entering was organized properly in the spreadsheet and easy to follow.
-# Do you have any further reaction to this case after viewing Dr. Baggerly's talk?
+#Do you have any further reaction to this case after viewing Dr. Baggerly's talk? I didn't realize it was the same Duke study until a couple minutes in and I was amazed that they sent that many letters without being taken seriously or suspending the trials to review the study. In complete honesty, I am appalled they didn't check their work before continuing and that no one listened to Dr. Baggerly and others.
 {{Template:Aporras1Links}}

← Older revision		Revision as of 06:38, 24 October 2017
Line 108:		Line 108:

	[[User:Johnllopez616\|Johnllopez616]] ([[User talk:Johnllopez616\|talk]]) 23:26, 23 October 2017 (PDT)		[[User:Johnllopez616\|Johnllopez616]] ([[User talk:Johnllopez616\|talk]]) 23:26, 23 October 2017 (PDT)
		+	== Arash Lari's Responses ==
		+	#
		+	#
		+	#
		+	#
		+	[[User:ArashLari\|ArashLari]] ([[User talk:ArashLari\|talk]]) 23:38, 23 October 2017 (PDT)
		+	{{template:ArashLari}}

@@ Line 92: / Line 92: @@
 #*Dr Baggerly recommends a more strict and regimented  protocol, which include: (1) Following conventions for labeling data, (2) Heavy documentation of all processes, and (3) Most importantly having a reproducible workflow. DataOne reinforces these practices, especially in reference to reproducibility of all experiments coupled with proper documentation.
 #What best practices did you perform for this week's assignment?
-#*
+#*For this assignment, I made sure to save frequently and make clear labels for all headers so that it was there was no ambiguity about what each column's information contained. In addition, I also wrote each step down on a note pad with pen and paper so that I would be able to modify the assignment steps specifically to my groups assignment.
 #Do you have any further reaction to this case after viewing Dr. Baggerly's talk?
-#*
+#*I'm not sure if my reaction has changed, I both was and still am in shock about what happened, especially at the negligence that took place to allow for it. I'm not one for invading others space but when this type of autonomy can be just given... too much power can be shifted into the wrong hands, and no one will know until we see horrors like these, until it is too late.<br>
 ==John Lopez Responses==

← Older revision		Revision as of 06:26, 24 October 2017
Line 95:		Line 95:
	#Do you have any further reaction to this case after viewing Dr. Baggerly's talk?		#Do you have any further reaction to this case after viewing Dr. Baggerly's talk?
	#*		#*
		+
		+	==John Lopez Responses==
		+
		+	# Perhaps the biggest issues with the data and analysis involved the fact that the data could not easily be reproduced. There was no clear way to initially show how the results were obtained. Furthermore, the lack of good, correct labels, in order to reproduce the data was also absent. These are two of the best practices violated by DataONE. However, improper labeling would be the more common of the errors presented.
		+	# Labeling properly is generally what Dr. Baggerly recommends for reproducible research, as well as making sure the research is in fact reproducible. The labeling ties heavily with the DataONE recommendation since it allows for clear interpretation of what is presented as well as good to be reproduced.
		+	# I performed several of the best practices for this week's assignment, including using consistent column names, creating descriptive column names without spaces/special characters, using descriptive file names, and leaving missing data blank.
		+	# After viewing Dr. Baggerly's talk, combined with the realizing the easy potential for errors in my own assignment, is that data is extremely tricky to work with and it is still so easy to manipulate to achieve undesirable results. Although the presentation made it clear that the data should have easily been determined as incorrect, it's still difficult to determine exactly how far down the research things went wrong.
		+
		+	{{Template:johnllopez616}}
		+
		+	[[User:Johnllopez616\|Johnllopez616]] ([[User talk:Johnllopez616\|talk]]) 23:26, 23 October 2017 (PDT)

← Older revision		Revision as of 06:20, 24 October 2017
Line 85:		Line 85:

	{{Template:Aporras1Links}}		{{Template:Aporras1Links}}
		+
		+	==Simon Wroblewski's Reflections==
		+	#What were the main issues with the data and analysis identified by Baggerly and Coombs? What best practices enumerated by DataONE were violated? Which of these did Dr. Baggerly claim were common issues?
		+	#*The main issues identified by Baggerly and Coombs has to be the fact the the data was not reproducible and therefore incapable of being verified. Many practices from DataONE were violated, such as mislabeling, indexing, and manipulating data.
		+	#What recommendations does Dr. Baggerly recommend for reproducible research? How do these correspond to what DataONE recommends?
		+	#*Dr Baggerly recommends a more strict and regimented protocol, which include: (1) Following conventions for labeling data, (2) Heavy documentation of all processes, and (3) Most importantly having a reproducible workflow. DataOne reinforces these practices, especially in reference to reproducibility of all experiments coupled with proper documentation.
		+	#What best practices did you perform for this week's assignment?
		+	#*
		+	#Do you have any further reaction to this case after viewing Dr. Baggerly's talk?
		+	#*