Timeline for I have egregiously sloppy (possibly falsified) data that I need to correct. How should I go about this?
Current License: CC BY-SA 4.0
25 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 1, 2021 at 19:35 | comment | added | Dan Is Fiddling By Firelight | While the only consequences to me were just a trip to the ER and a permanent scar, I managed to bury the fact that a childhood accident whose circumstances made no sense was that rather than pushing back on bad how to instructions (to swing a hammer with my right hand despite being left handed) I did something that I thought would just result in an "OWWW!" after which my point for "that's the wrong way to do it" would have been made for about 25 years. | |
| Mar 1, 2021 at 19:27 | comment | added | Dan Is Fiddling By Firelight | @BryanKrause one way I could see it happening would be to intermingle test data with real data. If the OP was using modified copies of the real data to test various cases within their analysis path and was sloppy about keeping track of which files were which when copying data around. Similar to Andrew Savinykh; after 6 years I could see myself wondering if maybe I wasn't just way sloppier than I thought possible but in a moment of stupidity/insanity did something worse and then dumped all recollection down a self sealing memory hole. | |
| Feb 25, 2021 at 14:48 | comment | added | PLL | @user541686: As Andrew writes, it may be difficult for people with good memories to “get” how this kind of bad memory can work. In my adult professional life I’ve never committed such misconduct; but in high school, I did (if I remember right) fabricate data on a couple of science projects/assignments, and in undergrad, I remember at least being tempted to do it again on one project. I don’t think I did in the end, but I don’t remember for sure. It’s hard to distinguish in memory between the feeling of serious temptation/considering it, and actually doing it. | |
| Feb 25, 2021 at 4:08 | comment | added | user541686 | @AndrewSavinykh: Did you ever have any situations where you didn't remember if you intended to engage in misconduct? It's one thing to not know whether you are doing it (even in the present you might not know whether something constitutes misconduct...), it's another thing to not know whether you intended to engage in misconduct. | |
| Feb 24, 2021 at 22:12 | comment | added | Robbie Goodwin | Anything like "I majorly screwed up the analysis and made a lot of data-entry errors…" will stand only to your shame. "That version was corrupted. Here's the corrected version…" should work. None of that changes the fact that your exposition might look detailed but in fact, it's merely wordy… | |
| Feb 24, 2021 at 16:30 | comment | added | RBarryYoung | Using or not using Excel vs retaining the original raw data have nothing to do with each other. It has to do with using proper methodology, not what tool you are using. And also, Excel formulas are code (and I say that as a professional programmer). | |
| Feb 24, 2021 at 4:41 | comment | added | Numeri | @AndrewSavinykh I agree completely. I am capable of forgetting not only what day and month it is, but what I did that morning. This is not an unbelievable story. | |
| Feb 23, 2021 at 21:32 | comment | added | Andrew Savinykh | People who has perfect memory cannot get this. I totally can. Many times in my life including when I was young, I had situations where I could not remember some pieces of information, other people would deem "impossible to forget". One of the situation was very similar, I was accused of a misconduct, and I could not remember the exact chain of events, that would clear my name, I could not confidently say I did not do it. I suffered the punishment, and then a witness that was unwilling to come forward during the process, approached me and informed me that I in fact did not do it. It does happen | |
| S Feb 23, 2021 at 19:16 | history | suggested | JonH | CC BY-SA 4.0 |
No need for cussing.
|
| Feb 23, 2021 at 19:04 | review | Suggested edits | |||
| S Feb 23, 2021 at 19:16 | |||||
| Feb 23, 2021 at 17:45 | answer | added | emily | timeline score: 9 | |
| Feb 23, 2021 at 15:00 | history | tweeted | twitter.com/StackAcademia/status/1364228537503404035 | ||
| Feb 23, 2021 at 9:47 | comment | added | user21820 | @8263xiao: Hmm. I don't have a problem remembering whether I was deliberately dishonest or not, even a decade ago. From your last comment, it sounds like you do not have a habit of being dishonest, and you do not remember intentionally being dishonest. If that is so, take it at face value, because carelessness can cause a huge amount of errors. In particular, a mouse-slip might cause an entire chunk of text to be dragged from one point to another, and if middle-click does a paste operation then an accidental middle-click can overwrite data with data from somewhere else without you knowing. | |
| Feb 23, 2021 at 9:41 | comment | added | 8263xiao | @user21820 I did this research more than half a decade ago. I had to email my former PI just to have the data since I no longer had it. When I saw it, I couldn't believe the inaccuracies in my spreadsheets - while I did not alter any of the original raw data, I often miscopied the data into the spreadsheets to such an extent that I found it difficult to believe that it was ordinary carelessness. The graphs also seemed a bit nicer with the inaccurate data entry/analysis. | |
| Feb 23, 2021 at 9:25 | comment | added | user21820 | I find it hard to believe your implication that you do not know whether you deliberately mishandled the data in the past. Do you have any reason for not knowing for sure? | |
| Feb 23, 2021 at 8:24 | history | became hot network question | |||
| Feb 23, 2021 at 8:19 | comment | added | 8263xiao | I never modified the original raw data. I did egregiously miscopy that raw data into spreadsheets where I do my calculations. | |
| Feb 23, 2021 at 5:18 | comment | added | Captain Emacs | @Jeff This. Data should be always available in the original raw form as it was created by the measurement device or at the time of the experiment. That version of data goes into Fort Knox and is never ever modified. | |
| Feb 23, 2021 at 3:54 | comment | added | Jeff | A bit of a tangential point, but this is also why you should not be doing data work in spreadsheets. Data work should be done through code, without altering the originals, so that it can be reproduced and audited. | |
| Feb 23, 2021 at 2:01 | answer | added | Andreas Blass | timeline score: 55 | |
| Feb 23, 2021 at 0:56 | comment | added | Bryan Krause♦ | How is it that your own data dwell in some Schrödinger-esque contraption in which they are either falsified or merely erroneous? Do you mean to say you've actually falsified these data but are trying to be coy and test whether it can go unnoticed as you correct them? Or looking for a reason to keep it in the altered form? | |
| Feb 23, 2021 at 0:43 | answer | added | Buffy | timeline score: 17 | |
| Feb 23, 2021 at 0:35 | history | edited | 8263xiao | CC BY-SA 4.0 |
added 10 characters in body
|
| Feb 23, 2021 at 0:27 | history | edited | 8263xiao | CC BY-SA 4.0 |
added 297 characters in body
|
| Feb 23, 2021 at 0:21 | history | asked | 8263xiao | CC BY-SA 4.0 |