94% of the 88 spreadsheets audited in 7 studies have contained errors.
Panko (2008)
Spreadsheets are easy to use and very hard to check.
Chen & Chan (2000)
Errors in spreadsheets... result in incorrect decisions being made and significant losses incurred.
Beaman, et al (2005)
A significant proportion of spreadsheets have severe quality problems.
Ayalew (2007)
Even obvious, elementary errors in very simple, clearly documented spreadsheets are... difficult to find.
Galletta, et al (1993)
Programmers exhibit unwarranted confidence in the correctness of their spreadsheets.
Krishna, et al (2001)
The software that end users are creating... is riddled with errors.
Burnett & Myers (2014)
People tend to believe their spreadsheets are more accurate than they really are.
Caulkins, Morrison, & Weidemann (2006)
Most executives do not really check or verify the accuracy or validity of [their] spreadsheets...
Teo & Tan (1999)
Spreadsheets have a notoriously high number of faults.
Rust, et al (2006)
Every study that has looked for errors has found them... in considerable abundance.
Panko & Halverson (1996)
Spreadsheets are commonly used and commonly flawed.
Caulkins, Morrison, & Weidemann (2008)
Despite being staggeringly error prone, spreadsheets are a highly flexible programming environment.
Abreu, et al (2015)
Spreadsheets are more fault-prone than other software.
Kulesz & Ostberg (2013)
Spreadsheet errors have resulted in huge financial losses.
Abraham & Erwig (2007)
Most large spreadsheets have dozens or even hundreds of errors.
Panko & Ordway (2005)
...few incidents of spreadsheet errors are made public and these are usually not revealed by choice.
Kruck & Sheetz (2001)
1% of all formulas in operational spreadsheets are in error.
Powell, Baker, & Lawson (2009)
The untested spreadsheet is as dangerous and untrustworthy as an untested program.
Price (2006)
Spreadsheets are dangerous to their authors and others.
Durusau & Hunting (2015)
The issue is not whether there is an error but how many errors there are and how serious they are.
Panko (2007)
Spreadsheets can be viewed as a highly flexible programming environment for end users.
Abreu, et al (2015)
Spreadsheets are the most popular live programming environments, but they are also notoriously fault-prone.
Hermans & van der Storm (2015)
A lot of decisions are being made on the basis of some bad numbers.
Ross (1996)
Spreadsheet errors are still the rule rather than the exception.
Nixon & O'Hara (2010)
Spreadsheets are notoriously error-prone.
Cunha, et al (2011)
Every study, without exception, has found error rates much higher than organizations would wish to tolerate.
Panko (1999)
Spreadsheet shortcomings can significantly hamper an organization's business operation.
Reschenhofer & Matthes (2015)
Spreadsheet errors are pervasive, stubborn, ubiquitous and complex.
Irons (2003)
Developing an error-free spreadsheet has been a problem since the beginning of end-user computing.
Mireault (2015)
Spreadsheet development must embrace extensive testing in order to be taken seriously as a profession.
Bock (2016)
Overconfidence is one of the most substantial causes of spreadsheet errors.
Sakal, et al (2015)
Spreadsheets... pose a greater threat to your business than almost anything you can imagine.
Howard (2005)
The results given by spreadsheets are often just wrong.
Sajaniemi (1998)
It is now widely accepted that errors in spreadsheets are both common and potentially dangerous.
Nixon & O'Hara (2010)
Studies have shown that there is a high incidence of errors in spreadsheets.
Csernoch & Biro (2013)
Spreadsheets are extraordinarily and unacceptably prone to error.
Dunn (2010)
The quality and reliability of spreadsheets is known to be poor.
Bishop & McDaid (2007)
Your spreadsheets may be disasters in the making.
Caulkins, Morrison, & Weidemann (2006)
Never assume a spreadsheet is right, even your own.
Raffensperger (2001)
Spreadsheet errors... a great, often unrecognised, risk to corporate decision making & financial integrity.
Chadwick (2002)
Despite overwhelming and unanimous evidence... companies have continued to ignore spreadsheet error risks.
Panko (2014)
Research on spreadsheet errors is substantial, compelling, and unanimous.
Panko (2015)
Spreadsheets are alarmingly error-prone to write.
Paine (2001)
It is irrational to expect large error-free spreadsheets.
Panko (2013)
Spreadsheets are often hard, if not impossible, to understand.
Mireault & Gresham (2015)
Spreadsheets contain errors at an alarmingly high rate.
Abraham, et al (2005)
60% of large companies feel 'Spreadsheet Hell' describes their reliance on spreadsheets.
Murphy (2007)
Untested spreadsheets are riddled with errors.
Miller (2005)
Errors in spreadsheets are as ubiquitous as spreadsheets themselves.
Colbenz (2005)

Spreadsheet bibliography

Title An empirical study of end-user behaviour in spreadsheet error detection & correction
Authors Brian Bishop & Kevin McDaid
Year 2007
Type Proceedings
Publication EuSpRIG
Series  
Abstract

Very little is known about the process by which end-user developers detect and correct spreadsheet errors. Any research pertaining to the development of spreadsheet testing methodologies or auditing tools would benefit from information on how end-users perform the debugging process in practice.

Thirteen industry-based professionals and thirty-four accounting & finance students took part in a current ongoing experiment designed to record and analyse end-user behaviour in spreadsheet error detection and correction. Professionals significantly outperformed students in correcting certain error types.

Time-based cell activity analysis showed that a strong correlation exists between the percentage of cells inspected and the number of errors corrected. The cell activity data was gathered through a purpose written VBA Excel plug-in that records the time and detail of all cell selection and cell change actions of individuals.

Full version Available
Sample
Errors corrected over cells inspected
Errors corrected over cells inspected

An important research goal was to determine if there was a correlation between the number of cells inspected and error detection/correction performance.

A scatterplot for errors corrected versus coverage shows a moderate-strong linear relationship.

That is, the more cells that are inspected, the more errors are detected.

Go to top