from an article By Jan Lee
Statistics don’t lie if you choose the right ones. The sign of a good writer is often reflected in the depth of her research. Statistics impress readers. Dates, times, and other anecdotal information can paint vivid pictures and often help to convey the writer’s point.
But anyone who has done research on the Internet knows that it can sometimes be difficult to ensure that the information presented is absolutely accurate.
Maryan Pelland’s guest blog for Problogger.net sums it up when it comes to the importance of using primary sources for quotes and anecdotal information. That irresistible quote you found on the internet, those stories that cinch the point your blog is making, should all come from a primary source, says Pelland.
“As a writer, you must understand primary and secondary sources. A primary source is the clichéd horse’s mouth. It’s the woman who pontificated the idea; the man who discovered the discovery. You’re obligated to find their phone number and dial them up. Or send an email. You ask direct questions and receive direct answers which you can quote, without making any alterations, or you can paraphrase if you indicate the paraphrasing.“
The same caution must be used when referring to statistics. You know, that government data you came across that a writer has nebulously quoted from another source that you can’t easily verify.
The case for good research
A case in point relates to an article I was researching some months ago. While trying to locate Census Bureau data on the number of private businesses in the United States that franchise to private investors, I discovered that the statistics were already widely published by newspapers, blogs, and business firms. We say statistics don’t lie, but it seems that one writer had published an inaccurate summary of a government report and generously rounded up statistics from the Census Bureau. Dozens of publications then followed suit, sometimes inflating the number more.
Business attorneys, real estate firms, newspapers, you name it, had all cited varying versions of the statistic they had read about without tracing its origin. In some cases, the authors cited the website they used, but they never made an attempt to confirm the data had been accurately cited by their source.
So how do you ensure that the data you are using in your blog post, article, or book is accurate? Are there criteria that can help writers evaluate their sources and make sure they aren’t falling prey to generalizations and rumors? Are there steps researchers can follow to help them assess their sources?
Assess your sources and make sure your statistics don’t lie
Writer and retired college instructor Robert A. Harris, who has written several textbooks on English composition and ways to avoid plagiarism, says there are steps that will help you negotiate these pitfalls. In his article Evaluating Internet Resources, he highlights what he calls CARS: Credibility, Accuracy, Reasonableness, and Support, a list of standardized questions that are routinely used to evaluate both print and Internet sources. Not surprisingly, Harris’ checklist is used by several colleges in the United States as a reference guide for English composition students.
In a nutshell, Harris proposes that there are four key points to consider when evaluating whether the website you pulled last year’s statistics from is trustworthy as a reference. His article gives a lengthy explanation of how you can test out each of these criteria. I have listed some of the examples he offers below:
Credibility (This determines the site and author’s authority as a reference):
- What are the author’s credentials?
- Is the website a trustworthy source?
- Whether the site is a respected and known source?
Accuracy (Confirms the truthfulness and precision of the source you are using):
- Is what you are citing accurate, up-to-date, and the latest published statistic or report?
- Is the data comprehensive enough?
- Is it relevant to your audience?
Reasonableness (Assures that the source is balanced, fair, and avoids bias):
- Does your source use objective criteria for its argument?
- Does it strive for balance?
- Is it devoid of conflict of interest issues?
Support (Confirms its argument through supportive research):
- Does it support its data with triangulation (at least two independent sources you can confirm)?
- Are the sources it uses identifiable and reliable?
- Is there supporting documentation?
In order for the statistics or data we use to be believable, they must come from a well-respected credible source. They must be up-to-date and broad enough to argue your point. They must not be able to be shot down by oversight or accusations of bias, and they must be able to be corroborated by your reader’s own diligent research. And many readers take the homework of fact-checking seriously.
Even small articles can be well-sourced
As bloggers, journalists, and authors, we want our public to trust the information they find on the Internet because we want them to trust us as credible sources. It’s up to us to prove that our work is fair, accurate, credible, and well-supported by using good research tools and ensuring that our sources demonstrate the same authority we want our readers to attribute to our polished works.
