Does a Double-Entry Model of Data Entry Work in Reducing Errors?

As any scientist, college professor, or an executive can tell you, proper data entry is the lynchpin of any research project, thesis, or enterprise. The importance of accurate data entry cannot be overstated, since even small mistakes that are repeated can lead to massive catastrophic cascade failures that can wipe out years of work or research, while simultaneously sullying the names of the people leading the project. Ask Martin Fleischmann and Stanley Pons, the inventors of ”œcold fusion”�, about the importance of accurate data entry.

Typically, ensuring accuracy in data entry falls to the person doing the work. The most common method of error checking is referred to as ”œvisual checking”�, and essentially tasks the person performing data entry duties with stopping occasionally and going back through their own work to verify that entries are accurate. The second most common method of data entry is called ”œsingle entry”�, and doesn’t actually involve error checking at all ”“ only a list of directives for the person entering the data, encouraging them to be as accurate as possible. While some offices and laboratories incorporate a second tier of data reviewers ”“ a group of individuals who review entered data and ensure accuracy ”“ this is far from a universal practice.

If this seems outlandish, that’s because it is. Accurate data entry is crucial, often the difference between a successful experiment and a waste of resources, or a profitable corporation and a bankruptcy. But even with everything on the line, most organization are willing to let data entry accuracy be judged almost entirely by an entry level student or day worker with little or no experience in the more sophisticated aspects of the information they are channeling.

Thankfully, a new process has been developed that allows individuals to perform a self-check process effectively, and that has been shown to reduce errors in a real-world data entry environment by over 95%.

In a recent study out of the University of Las Vegas, double entry data recording was tested against single entry and visual checking methods, and showed massive increases in accuracy across the board across multiple data entry projects. Interestingly, the difference in accuracy between single entry and visual checking was almost non-existent, which proves that instructing data entry specialists to check their own work is a waste of time. While double entry takes longer to perform than single entry or visual checking, it is certainly far more accurate.

– The Data Czar @ DEO