Anybody who says “don’t use Excel for data analysis” needs to realize how simple the needed analysis is at millions of businesses, how many people like using Excel, and how little they dislike being told to change how they’re comfortable working.
OP is citing a post about genomics research data sharing, where they have identifiers that are exceptionally vulnerable to Excel’s date inference and conversion algorithm.
OP’s post is probably still a good idea for researchers, where something like SPSS offers better protection for datasets, albeit with a higher learning curve.
For the rest of us, so long as we pay attention, Excel is an incredibly underrated analysis approach.
Alternatively, do use Excel for data analysis, but instead of opening CSV files directly in Excel, create a new blank workbook, then go to the Data tab, select From Text/CSV, and follow the prompts, which will allow you to preview your data and select proper data types for each column.
This will avoid all the issues alluded to here. And yes, I know that it's stupidly simple to go the wrong route, but I also think by now people would be wise to all of this, since the proper functionality has only been there for a decade or so. But, hey, apparently total helplessness keeps getting clicks.
6 comments
[ 2.7 ms ] story [ 19.3 ms ] threadThe problem is so bad in the discipline they changed how they name things: https://www.theverge.com/2020/8/6/21355674/human-genes-renam...
OP’s post is probably still a good idea for researchers, where something like SPSS offers better protection for datasets, albeit with a higher learning curve.
For the rest of us, so long as we pay attention, Excel is an incredibly underrated analysis approach.
This will avoid all the issues alluded to here. And yes, I know that it's stupidly simple to go the wrong route, but I also think by now people would be wise to all of this, since the proper functionality has only been there for a decade or so. But, hey, apparently total helplessness keeps getting clicks.