Wednesday, 6 January 2016

Putting the record straight: when the data you access is erroneous

Businesses, investors, patent searchers and analysts are now highly dependent on the information they can access online. Some of this information is relatively accurate, often coming from official or specialised and verified databases, while other data is drawn from private sector news and comment services, blogs and wikis. In all instances there is a tendency to assume that information once accessed is accurate unless it is plainly erroneous on its face or where it is contradicted by information that is of equal or greater reliability.

Rarely do data users suspect that the data they access may be faulty. Even more rarely do they gain an insight into how (or even whether) data is corrected. This makes last month's Guardian post by Richard Nelsson, "How we correct the record in the Guardian archives as well as on our pages", all the more interesting reading.

The Guardian's reputation for being error-prone
is undeserved: for explanation click here
Not all readers may be familiar with the Guardian's information offerings, since their history is parochial. They originate from the Guardian newspaper -- originally the Manchester Guardian and later a national UK publication. Following the usual pattern of expansion in the internet era, the Guardian now provides a suite of business and technology web pages which are free, easy to access and are targeted by search engines across the world. These include a regular Business section, Business Live, Business Blog, Stock market, Technology page, Economics page, Economics blog and Small business network.  

A Guardian data host
Annually the Guardian receives more than 30,000 communications from readers, of which the majority were either drawing attention to factual errors or complaining about the manner in which presumably accurate information was portrayed. Error correction requires both that notice of the error appears in the same medium as the error and that archived versions of the published item be amended. After explaining how it sets the record straight, this article addresses a further problem:
" ... Guardian content is also to be found on many other databases around the world, so these too have to be amended. Each day the syndication department sends out an email to more than 60 clients such as the newspaper text database hosts LexisNexis and Factiva, containing a summary and link to the corrections page. Again a note is added to the original article ...".
How aware are investment advisers and data analysts that the information they access online may come from a syndicated source and that it may have been amended or corrected at source without their being aware of it? In an era of instant information, when the demand for a swift response is so strongly felt by consultancies and professional services, double-checking one's data is rarely if ever a realistic option.  Bearing this in mind, use of data from truly reliable sources such as the ORoPO not-for-profit database of verified patent ownership records is particularly welcome.

No comments:

Post a Comment