Follow

Reminder about Excel and research 

Never use Excel for important research

It is malware!

> Gene name errors are widespread in the scientific literature

genomebiology.biomedcentral.co

> The spreadsheet software Microsoft Excel, when used with default settings, is known to convert gene names to dates and floating-point numbers. A programmatic scan of leading genomics journals reveals that approximately one-fifth of papers with supplementary Excel gene lists contain erroneous gene name conversions.

Reminder about Excel and research 

Microsoft Excel: It's fine to use!

As long as you watch it every single second to be sure that it didn't convert something into something else inappropriately

And you're okay with every single cell having potentially extensive code inside it, where you can't really tell that it's code and not a number until you click on it

And you're okay with a lot of the functions it performs not being reproducible unless you write down the steps for what you clicked on

Show thread

Reminder about Excel and research 

And you're okay with the file being corrupted sometimes and you just sort of lose all your data

And you're okay with Excel files not even being compatible with Excel on other platforms in some cases

And you're okay with it sometimes including meta-data that you don't mean to include

Show thread

Reminder about Excel and research 

So like, if we're talking about you weighing pros and cons of different printer models that you want to buy, or making a monthly budget that's for your own reference, probably okay

But if you're doing research, a CSV and an R script will get you where you want to go and with a lot fewer surprises

Show thread

Reminder about Excel and research 

@bgcarlisle Holy shit.

Reminder about Excel and research 

@bgcarlisle People don't talk nearly enough about That Time An Excel Error Caused Europe To Think Austerity Might Work.
Even now they wonder again whether austerity will return and not a mention of how the stupid economists behind the concept had just completely fucked up in Excel and caused mass impoverishment.
Excel: not even once

Reminder about Excel and research 

@cathal @bgcarlisle like this: theguardian.com/politics/2013/

Experts on the job! Amazing.

Lol @ talking about libreoffice and google calc as alternatives that might work.

Tbh i am not sure what they should use.. Python Pandas maybe, but they suck a bit too.

Reminder about Excel and research 

@jasper @bgcarlisle
The errors they made could be made in any spreadsheet, yes. Something nice about code, I think, is that the flow of logic _can_ be made more expressive, so if you refer to a data-range by name, it always refers to the same thing, while a slice over that range makes it clear that you're only referring to a portion. That could have avoided the mistakes they made.
But, code, lol, who will volunteer their economy for my JS-based economics project?

re: Reminder about Excel and research 

@cathal @bgcarlisle What's austerity in this context?

re: Reminder about Excel and research 

@Felthry @bgcarlisle
It's difficult to answer the question as you have framed it. But, this article may offer the needed context:
theguardian.com/politics/2013/

re: Reminder about Excel and research 

@cathal @bgcarlisle hmm. we don't know anything about financial politics other than "money is a thing that we would be better off without" so i don't really understand much of that article, honestly.

Impression I get is "paper said thing was bad because other thing, politician tried to change other thing to make thing good, then it turned out thing was actually perfectly fine all along and changing other thing made thing worse"?

re: Reminder about Excel and research 

@Felthry @bgcarlisle That's the official version of events. It's widely felt that "Politicians did self serving thing, to everyone else's detriment, using heterodox economics paper as excuse, continued doing it when paper was found to be fallacious, because paper's conclusions supported their otherwise unsupportable social notions".
Either way the paper was trash and the actions taken on its basis were, at best, unhelpful.

Reminder about Excel and research 

@bgcarlisle I learned my lesson with Excel very early on, when I took home an Excel 2000 file from school to open on my family's home PC, only to have it refuse to open due to incompatibility between, somehow, the version of Excel 2000 at home.

Reminder about Excel and research 

@bgcarlisle Also, doing anything involving the RNG is completely non-reproducible. Since it re-generates values every time you do *anything*

Reminder about Excel and research 

@bgcarlisle

Remember Clippy?

"It looks like you're trying to enter a number in scientific notation. Shall I convert it to floating point for you?"

It was so annoying that MSFT made Clippy disappear...except they DIDN'T really.

The spirit of Clippy lives on! It still haunts MS office...a invisible spectre that not only tries to predict your intentions, it also assumes it is always right and does what it thinks it should without even asking anymore!

Reminder about Excel and research 

@bgcarlisle I heard many horror stories from researchers when I was a research data librarian.

Reminder about Excel and research 

@bgcarlisle so r we like genetically modifying crops with excel errors

Reminder about Excel and research 

@bgcarlisle we are defs imposing calculations on labouring bodies with excel errors, so...

Sign in to participate in the conversation
Scholar Social

The social network of the future: No ads, no corporate surveillance, ethical design, and decentralization! Own your data with Mastodon!