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

RIVA COURSE 303: ADVANCED MODERATING
October 27-29, 2014
RIVA Training Institute will hold a course, themed 'Advanced Moderating,' on October 27-29 in Rockville, Md.
MOBILE SHOPPING 2014
October 28-30, 2014
Worldwide Business Research will hold its mobile shopping conference on October 28-30 at The Wigwam, Litchfield Park, in Phoenix.

View more Related Events...

Related Articles

There are 1041 articles in our archive related to this topic. Below are 5 selected at random and available to all users of the site.

Questionnaire design - some general considerations
As with any research method, questionnaires have advantages and disadvantages. The following article is an excerpt from the paper “Designing and Using Questionnaires” that addresses the advantages and disadvantages of using a questionnaire and the importance of having well-defined goals for the research.
Why people come before data every time
A director of data visualization offers her strategies for shaping and presenting research information.
Trade Talk: Cheap PCs: Is the revolution finally here?
An overview of the current and future state of the PC industry.
Data Use: Bayes nets: Understanding the best newest thing - Part II
In the second part of a two-part series on Bayes nets, Steven Struhl explores their capabilities and outlines what to look for in a Bayes software program.
Trade Talk: Wanted: marketing scientists
Marketing research is short on qualified data analysts. Linda Burtch says that marketing researchers with a strong knowledge of multivariate analysis and other statistical areas - which she calls "marketing scientists" - are much in demand.

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