I am not a fan of Excel.

Reasons not to use Excel

My main reason is that it does things without telling me. Anyone who ever tried to import an English formatted table into a German speaking program knows what I mean – it drops “leading” zeros (as in 0.2), because in Germany we use commas instead (0,2). Excel unapologetically changes the 0.2 into a 2 without telling me and at the same time OVERWRITES the document. I hate you Excel, I really do, because if the document contains a few thousand entries and not all of them start with a zero, chances are that you don’t notice. I don’t really need to explain the implications, do I? And yes. It happened to me. And yes, for ages I did not know what was happening to my data… and yes, I cursed not just quietly after finding out…

Also, if I tell Excel to save something to *.csv (comma-separated-values, a very simple text-based table data format, which usually every program can read), it doesn’t do it correctly (forgets to put text-entries with commas in quotation marks), which can lead to horrible complications.

Also, the graphics created by Excel are ugly.

Also, there is actually a scientific article on how statistical procedures in Excel are not accurate ().

But what to use instead? I cannot expect everyone to invest the time and effort needed to learn the statistical programming language R, which I prefer, just to draw a little pie chart (google “don’t use pie charts” before doing it!).

Reasons to use LibreCalc instead

I tell people to install LibreOffice and to use their table calculation program “Calc”. It looks in essence a bit like an old Excel, but

  • It is free
  • It does not do things without telling me (it simply asks me, seriously Excel, that’s not too much to ask, is it?!)
  • It imports and exports csv perfectly and readily, offering different separators and all that
  • It creates much nicer graphics, that are – I think – more easily adapted
  • I don’t think I’ve heard anyone complain about its statistical calculations
  • I like it. And I’m not the only one.

So, if you don’t need to do a complicated statistical analysis, you don’t need complicated statistical software, of course. A spreadsheet-program might be what you need, absolutely.

Just don’t use Excel.

Please.

Bibliography

McCullough, B.D., and David A. Heiser. 2008. “On the Accuracy of Statistical Procedures in Microsoft Excel 2007.” Computational Statistics & Data Analysis 52 (10): 4570–78. https://doi.org/10.1016/j.csda.2008.03.004.

Sophie Schmidt

Founder & Editor

About the Author

My name is Sophie, I am a prehistoric and computational archaeologist and have been research associate at the Universities of Bonn and Cologne, as well as for the NFDI4Objects project at the German Archaeological Institute. I teach statistics for archaeologists, work on new methods in settlement archaeology (GIS, geostatistics in R and stuff) and am interested in archaeogaming. Now I started my PhD-project on the 5th mill. BC in Brandenburg (that's North-East Germany).

View Articles