88% of spreadsheets have errors

T

tjscott0

Recycles dryer sheets
Joined
Aug 18, 2009
Messages
116
88% of spreadsheets have errors - MarketWatch

Microsoft Excel makes it easy for anyone to do the kind of number crunching once reserved for accountants and statisticians. But the world’s best-selling spreadsheet software has also contributed to the proliferation of bad math.

Close to 90% of spreadsheet documents contain errors, a 2008 analysis of multiple studies suggests. “Spreadsheets, even after careful development, contain errors in 1% or more of all formula cells,” writes Ray Panko, a professor of IT management at the University of Hawaii and an authority on bad spreadsheet practices. “In large spreadsheets with thousands of formulas, there will be dozens of undetected errors.”Such mistakes not only can lead to miscalculations in family budgets and distorted balance sheets at small businesses, but also might result in questionable rationales for global fiscal policy
Click to expand...
 
GIGO
 
Last edited:
heeyy_joe said:
The error rate was actually found to be 89%, there was an error in his spreadsheet.
Click to expand...

In my lifelong professional experience, I've found that the most difficult task to get people to do is to check their work before submitting it to wherever.
 
I catch errors in my own spreadsheets on occasion. The vast majority of them are in formulas on the summary screen, where I collect numbers from various locations and create totals by asset class. When my individual holdings change, I don't always remember to modify the formulas on the summary page. I'm aware of this tendency and usually do enough cross-checking to find the errors fairly quickly. But it's something I need to be vigilant about, since if my totals by asset class are off, I may be induced to perform some rebalancing that would be unjustified by the correct numbers.
 
The first eight drafts of my spreadsheets have errors. The final draft is always perfect.
 
I hates spreadsheets. :mad:
They remind me of suspense dates (deadlines) to fill in my portions, only to have the "new and improved" one sent back via email, with an even shorter suspense, and have to redo my entries all over again.
Snarl...:mad:
 
I once checked 26 spreadsheets I use to keep track of my assets. Ten had errors when the results were audited by my accountant.

I think my batting average of .615 is pretty high, much higher than every baseball great.
 
heeyy_joe said:
The error rate was actually found to be 89%, there was an error in his spreadsheet.
Click to expand...
Good one. I'm glad I clicked into this thread, hehe.

I put in =if(abs(sumxyx-sumabc)>1,"ERROR","")

Even though I'm an engineer by degree, I did a stint as a financial analyst for none other than I B Megacorp, hehe. Hated that job, but I got to be really good at (don't laugh) Lotus 1-2-3.
 
target2019 said:
I once checked 26 spreadsheets I use to keep track of my assets. Ten had errors when the results were audited by my accountant.

I think my batting average of .615 is pretty high, much higher than every baseball great.
Click to expand...


An error is a fielding mistake and a .615 avg. would mean you would not be playing.... :)
 
braumeister said:
87.6% of all statistics cited as examples are made up.
Click to expand...

:LOL: The perfect comment, IMO.

Sadly, mathematical illiteracy seems widespread and at a shockingly high level these days. Consequently it would be very surprising to me to discover that only 88% of spreadsheets have errors.
 
I can believe it. I'll admit, in the past, I have found errors in some of my spreadsheets. "Now" whenever I change a formula, make a major change or build a new spreadsheet, I'll test it with some simple tasks that I can easily verify with a calculator (or pencil and paper :)) So far that seems to be working for me.
 
errors, what errors - my spreadsheets always give me the answers I want! oh wait maybe they should give me the correct answers instead.
 
If my spreadsheets do not provide the answers I am expecting, I use an FF value. It really comes in handy. I Multiply or Divide the offending result by a FF value and my my results are what I expect. Sometimes this is an iterative process.


(FF = Fudge Factor)
 
One thing that I try to do in important worksheets is to have a check number... IOW, you can usually calculate things in a couple of ways... so I do it both ways and have a cell that should either equal the other one or a cell that equals zero (subtract the two)... this has helped me out a number of times when something gets moved and a formula is not what you think it is... ie, it has a fixed cell when you thought it was dynamic...
 
I find errors in my spreadsheets about 12% of the time, and it's shocking every time! :D

Seriously, I find it hard to believe it's really as high as 88%, but I can't prove it...unless the rest of you are really careless. :D
 
I still have a copy of Multi-Plan and an early version of Lotus 123. - Ah, those were the days:)
 
You must log in or register to reply here.

Latest posts

Back
Top Bottom