Skip to: Main Content / Navigation

  • Facebook
  • Twitter
  • LinkedIn
  • Add This

The trouble with convenience sampling



Article ID:
20130526-3
Published:
May 2013
Author:
Noel Roos

Article Abstract

Customer feedback is often emotionally-charged. The author reminds researchers that relying on non-representative data in favor of convenient data from established customers can lead to expensive mistakes and poor decision-making.

Editor's note: Noel Roos is vice president at Diedrich RPM, a Burnsville, Minn., research company. She can be reached at 320-760-8223 or at noel@diedrichrpm.com. Roos is located in Burnsville and Alexandria, Minn. This article appeared in the May 20, 2013, edition of Quirk's e-newsletter.     

 

Recently, a student completing his master's degree called to ask a question regarding convenience versus random sampling - and the perils of making decisions based purely on a convenience sample. 

 

"Is it poor business to make decisions on convenience data?" asked the student.


To the great frustration of the student, I said, "It depends."

  

Many companies today have continuous feedback surveys - and it's a good practice, especially if there is an incentive and a company can capture those who are not motivated by extreme happiness or anger. But the bottom line is that while continuous feedback surveys can reveal trends, companies should compare these results with a random sample before making expensive product, service or resource decisions.

  

For example, a retail company routinely gathers a convenience sample of its shoppers via a link on its Web site. Why does a shopper look for and click on this link? Usually because they are motivated by a particular emotion. This same company finds that the convenience sample reveals that it needs to offer lower prices on a certain product or risk losing customers. Should the company react and lower its prices (and possibly product quality to maintain an acceptable profitability level)?

  

No.

  

At least not yet. This trend warrants further research before changes and a new price/value strategy are created. We would recommend this retailer take the feedback seriously but caution that it's possible that only the emotionally-motivated have filled out the survey, missing the bulk of its shoppers.

  

A survey based on a random sample should be implemented to understand the true impact of price (and possibly quality) to understand if there is an issue. A change in price strategy before understanding how the full shopping population feels could result in keeping the small segment who complained while leaving money on the table from customers who did not complain - or worse yet, losing the rest of the population because the price/value equation was altered.

  

The squeaky wheel has many channels to voice displeasure in today's world - including social media, e-mail and yes, feedback surveys. Meanwhile, the happily satisfied keep quietly moving along. Given how loud the world can be, it's more important than ever to get the full picture through sound research practices before making expensive decisions that could negatively impact the bottom line.

Comment on this article

comments powered by Disqus

Related Glossary Terms

Search for more...

Related Events

IRI SUMMIT 2015
April 20-22, 2015
IRI will hold its annual summit on April 20-22 at the J.W. Marriott in Austin, Texas.
MX 2015 MANAGING EXPERIENCE
April 20-21, 2015
Adaptive Path will hold its MX2015 Managing Experience Conference on April 20-21 at the Parc 55 Hotel in San Francisco.

View more Related Events...

Related Articles

There are 1108 articles in our archive related to this topic. Below are 5 selected at random and available to all users of the site.

Why the most important part of exhibiting comes after the conference
Steve Henke provides a three-step process for creating and executing post-conference follow-up.
Propensity score analysis: A tool for MR
Propensity score analysis is used in non-experimental and quasi-experimental research to adjust for group differences that existed before a treatment or intervention. The author shares how variations of this concept can be used in MR.
Your checklist for picking the right research software
The author highlights areas to examine when searching for a software tool that is right for your current and future research needs.
Software Review: WinCross 14
Freelance data processor Eric Ramon outlines why he’s still sold on his favorite tab package.
Discover more with Quirk’s Webinar archive
This article provides three recommendations for those looking to dive into Quirk’s Webinar archive.

See more articles on this topic

Related Suppliers: Research Companies from the SourceBook

Click on a category below to see firms that specialize in the following areas of research and/or industries

Specialties

Conduct a detailed search of the entire Researcher SourceBook directory

Related Discussion Topics

Market research report
08/20/2013 by Aarkstore Store
Most commonly used research techniques
07/28/2010 by Curtis J. Fedder
yes, I have experience with those ethics issues
06/23/2010 by Michael R. Hollon
Ethics
06/11/2010 by Jim Santilli
Research and analysis to foresight your business perspective
02/19/2010 by Emmanuel M. Mendy

View More