Come backstage with our bass playing Questionnaire Design 201 Instructor Greg Timpany
Meet Greg Timpany. Greg’s career arc has delved into the world of marketing, analytics and strategy for over 20 years. His expertise bridges the space between the structured world of IT and the creative, customer-centric needs of marketing, sales, product development, and customer service. He is a contributing author to several market research blogs, and has served as an instructor for Research Rockstar and Meredith College. He is a co-founder Off2Market Ltd. and Anova Market Research. His past market research, competitive intelligence, and strategy engagements include Global Knowledge, The Los Angeles Times, Guitar Center, Wilkin Guge Marketing, and Omnitrans. Greg holds undergraduate and graduate degrees from California State University, San Bernardino. He is a distinguished member of Alpha Kappa Psi, a Professional Business Fraternity. He can be followed on Twitter @DataDudeGreg. Greg was recently interviewed by Melissa Donovan.
MD: Hi Greg, thank you for taking the time to speak with me. What is the most common questionnaire design mistake you see?
GT: The most common mistake I see is a failure to realize the respondent’s most precious resource…their time. Many survey researchers fail to keep their surveys to a reasonable time limit. This is especially critical if we are asking more complex questions. These question types place a higher cognitive load on the respondent. Keeping surveys short and focused is the best method for obtaining reliable data. I have asked my clients if they would be willing, or capable, of responding accurately to a 20-30 minute survey. Sadly, most could not. Establishing a stream of research is a better alternative to a single catch-all survey.
MD: Are there any common misperceptions about scales that you would like to correct? Pick just one!
GT: To keep it to one is a challenge! But as a start, misperceptions about when to use Likert scales comes to mind.
- They are useful—especially if the goal is to measure multi-item constructs.
- They are prone to acquiescence bias. There are many ways to re-word both the questions and the response categories to minimize this form of bias. First, we need to give respondents permission to use all ends of the scale. This is a simple process and tells the respondent in the instructions that it’s acceptable to use all points of the scale.
- As for response categories, I am in favor of moving away from the “strongly agree” to “strongly disagree” format to one that is specific for the question, such as “not satisfied” to “very satisfied” for satisfaction questions.
MD: What is something interesting about you that many people may not know?
GT: When I am not crunching data, I am channeling my creative energy into my music. I am a bass player and developing songwriter. This current health crises has given me more time to focus on creating music. This has helped to keep my head in a positive place and not focus on things outside of my control.
Interested in taking Greg’s upcoming Questionnaire Design 201: Scale Selection? Current Backstage Pass and Certificate members can attend live or on-demand (24/7) from the Training Portal. The next real-time session starts Sep 15 and runs 4 consecutive Tuesdays (9/15, 9/22, 9/29, & 10/6) from 11am-12:30pm. Not yet a member? Check out our options.
About this course:
Questionnaire Design 201: Scale Selection. In this 4-part program, Instructor Greg Timpany covers various ranking and rating scales, including Likert, semantic differential, constant sum, and more. The class also delves into decisions related to the choice of even versus odd scales, number of points (5? 7? 10? 11?), and related topics.
Upon course completion, you will be able to:
- Understand the criteria for selecting scales
- Choose the best scale type and labeling for different measurement needs
- Avoid common questionnaire design errors
- Identify alternatives for cases where standard scales may be inadequate
