| February 27, 2008 |
David's IT Tips |
Opening Multiple Windows in OutlookWhen you switch from viewing your Inbox to your Calendar, or some other type of item in Outlook, you lose sight of information you may need from the previous window. The easy, but often overlooked, solution is to open your Outlook folders in their own windows. To do this, right-click on an icon in the Outlook bar or on a folder in the folder list and chose Open in New Window. You can then size and position the windows so you can see what you need to.
|
| UPCOMING EVENTS |
|
Mar10-13, 2008, St. Pete Beach, FL Marketing WorkshopThis unique event for marketing professionals, offered by AMA, gives you the opportunity to design a customized program that most benefits you! With over 20 two- and four-hour presentations to choose from, including tracks specializing in the latest in strategic marketing, CRM, branding and customer satisfaction/value/loyalty you can create your own schedule to meet your personal objectives. March 28, 2008, Chicago, IL Beyond Marketing 2.0: Harnessing the Power of Social Media for Marketing Campaign ResultsThis first-ever AMA Hot Topic on Social Media is just the place to learn the what, why and how of Social Media and how your organization can benefit from the latest marketing tools and techniques. April 3-4, 2008, Charlotte, NC AMA Marketing Boot CampIn today's ever-changing marketplace, businesses need their marketing professionals to understand the basics of marketing now more than ever. Business is rapidly changing and needs have increased competition faster than ever imagined. To help reinforce and focus your teams or your own marketing skills, the AMA introduces this highly interactive program facilitated by an experienced AMA instructor. |
|
Look for MR perspectives again next month to keep up to date with Marketing Research issues, opportunities and challenges. And please check out our new and improved website at www.polarismr.com for articles, tools and tips that will help you make the most of your marketing research! |
Which is more difficult to control -- the weather or bias in a research survey? That question would be a real stumper to many in the research field. Some types of survey bias are easier to know and manage than others, but as a whole, the topic is a difficult one that affects all survey modes.
If we're talking about question/response bias, then there are plenty of techniques that can be used to mitigate this bias. Some frequently used methods are:
1. Write questions that are clear, precise, and relatively short
2. Do not use “loaded” or “leading” questions
3. Avoid double-barreled questions
4. Avoid double negatives
5. Use both mutually exclusive and exhaustive response categories for closed-ended questions
6. Reverse the wording in some of the questions to help prevent response set bias
7. Rotate the responses
These are very commonly and successfully used to avoid response bias, but what about nonresponse bias? Polaris Marketing Research defines nonresponse bias as "Error that results from a systematic difference between those who do and do not respond to the measurement instrument." (Polaris Marketing Research - Glossary of Terms)
There are many schools of thought and opinions on the degree of impact of nonresponse bias on survey results. On the University of Florida's website, Glenn Israel reports that, "As nonresponse increases, the potential for a biased sample increases. This means that the obtained responses of a probability sample may no longer be representative of the larger population." Nonresponse impacts surveys in overall bias, item bias and in making the data collection period longer (which can introduce more bias).
Short of throwing the data away, here are some options to understanding and managing nonresponse bias when the data has been collected.
1. Generalize to the respondents only.
2. Assume there is no response bias and generalize to the population.
3. Use a different methodology to re-contact nonrespondents to get their information.
4. Compare data in hand on respondents and nonrespondents (e.g., demographics, customer information).
5. Compare characteristics of early respondents with late respondents (who are more likely to be similar to nonrespondents).
6. Increase mailings or contact efforts.
In most cases, researchers don’t have much information about nonrespondents, so the best way to address nonresponse bias is to encourage response in the design and management of the survey. Some ways of doing this are to:
1. Use cash incentives.
2. Use multiple follow-up contacts.
3. Use short, easily understood surveys when possible.
Overall, the response and nonresponse biases are aspects that must be addressed in every survey in order for the results to be meaningful. Question/response bias can be more easily controlled, while nonresponse bias requires forethought, great care and intuition, and even then can be a source of uncertainty.
____________________________________
Jeremy Lee is a senior data analyst in the analytics department of Polaris Marketing Research Inc., where he is responsible for handling data manipulation tasks involved in survey research. Lee has a bachelor's degree in computer science from The University of Tennessee.
