1 / 45

What’s all the noise about?

What’s all the noise about?. Steven Gittelman, Ph.D. Our data shows that the frequency of problem responders (professionals, speeders, etc.) is far lower in Europe than in the United States: we have acculturated our respondent pool.

Télécharger la présentation

What’s all the noise about?

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. What’s all the noise about? Steven Gittelman, Ph.D.

  2. Our data shows that the frequency of problem responders (professionals, speeders, etc.) is far lower in Europe than in the United States: we have acculturated our respondent pool. Problem respondents may be defined as those who bias market research data so that it misrepresents information needed for business decisions. Thus, speeders, professionals, satisficers, etc. matter little unless their purchasing behavior is different. So are the variations noise or signal? Is it noise or signal?

  3. Are the differences cultural or just plain old noise?

  4. We need to measure it. To measure it we need to agree upon metrics. What are the important parameters--- - Psychographics? - Demographics? - Purchasing Behaviors? - Or should we hang our hats on “problem respondents?” - What about duplication technologies? What can we do about it?

  5. Once upon a time we lived in a world of probability samples. There was a standard, consistency in our lives. Now we are in a non-probability universe online. But somewhere there is a platform—a theoretical population online. We need a new basis for comparison, a new standard. We cannot find ourselves in typologies of problem respondents. We have tried post survey to anchor ourselves in a collective standard-it is retrogressive. But we need a progressive standard that changes with each new sample population that we add. To build such a model we need more data that provides a broader anchor. Brave new world

  6. Respondent Types Professional Respondents fall into four categories: (1) Self report taking on-line Surveys “practically every day” (25% of Total). (2) Self report (open ended) taking over 30 online surveys “in the past month” ( 15% of Total). (3) Multiple panel membership >panels. (4) Panel members of long vs. short duration. Inconsistency: Brand vs. Price, Price vs. Brand, Happy with standard of living vs. unhappy. Failure to follow instructions: Instructed to enter a predetermined answer, also known as a trap question. Speeders: survey times in the bottom 10 percent. Straight liners: low variance in answers to grid questions. Duplication technology

  7. I can see clearly now! Compared survey results from 17 sets of US Consumer Panels, 1 UK and CATI. 400 completes per source. Dec. 2007- Dec. 2008. Selected demographic quotas (age, income, gender, ethnicity) were used to simulate census. Median length was 13 minutes. Questions covered: Technology and the media, Participation in market research, Buyer Behavior, Values and lifestyle, Demographics, Questionnaire Satisfaction.

  8. Number of Surveys Taken per month

  9. Average Number of Panel Memberships M1 and M2 were not asked the number of panels.

  10. Professional RespondentsBy Sample Source Panel M1 and M2 were not asked number of panels.

  11. Professional RespondentsSample sources grouped by type.

  12. Impactof tenure on Panel

  13. Impact of Max Panel Age----Just when you thought it was safe to go into the water.

  14. Max age on panel by panel (Eliminated M1-M5)

  15. Distribution of Survey Flaws, Inconsistencies (Brand over Price)

  16. Distribution of Survey Flaws, Failure to follow instructions

  17. Distribution of Survey Completing Time

  18. Cumulative Elapsed Time as a Log Normal Distribution

  19. Distribution of SpeedersShortest 10%

  20. Straight-Liner Distribution for RDD Phone Sample

  21. Straight Line Responses as Standard Errors (Variance) -Online

  22. Gender, Age, Income, and Ethnicity set by Quota Compare Distributions by: Education Having Children under 18 Employment Compare Across Panels Telephone---what do we do about cord cutters? Census Demographics

  23. Education Distribution

  24. Education Distribution by Panel –No Quotas Set No Quotas Set Point System UK

  25. Having Children Under 18No Quotas Set Professionals

  26. Having Children Under 18 No Quotas Set

  27. Employment DistributionNo Quotas Set

  28. Employment Distribution No Quotas Set Social Network Point System Access Panels River UK

  29. Let’s go to the videotape….We’ve examined panels in the usual format, demographics and respondent typologies.But where does that leave us?------Do we have a model for the future as yet? So we will give it another try---segmentation and cluster analysis bring variables together.

  30. Social opinions and behavior can be expected to drive purchasing behavior or at least provide a basis for segmenting the market. Consistency of these measurement may likewise be critical. Variables Groups Internet Use. Taking Surveys Having a Passport Social Characteristics Measures: Driving Variables Variation in Opinions

  31. Key Social Variables • Selected key variables determining Social Clusters. • Social Behavior, Unconventional, Passports, Time over Money, Risk Avoidance

  32. Measuring buyer behavior is the objective of most marketing research. And therefore, consistency of those measurement are critical. Variables Number of High Tech Items Purchased. Internet Purchase behavior Purchasing Opinions Measures: Clusters (Segments) Driving Variables Variation in Buyer Behavior

  33. Buyer Behavior Segment Description

  34. Buyer Behavior Segments by Panel Social Network Point System Access Panels River UK

  35. Statistical Panel Profiles Against Buyer Segments

  36. Buyer Behavior Segments by Respondent Type

  37. MDS (Multi Dimensional Scaling)Position based on Buyer Behavior

  38. Principal Buyer Behavior vs. Professionals (>30 Surveys/Month) M1-5 (US) Deleted M15 (UK) Deleted

  39. Principal Buyer Behavior vs. Speeders (<10 Percentile of Completion Speed) M1-5 (US) Deleted M15 (UK) Deleted

  40. Are we there yet? Are we there yet?Our discussions have been retrogressive, looking back at cumulative data collected. For us to provide workable tools they must be progressive, not retrogressive. After all, we don’t want to go through this every time we add a new panel. Do we? We don’t want to re-segment each time there is a new panel on the block.

  41. In Progress….Currently collecting data in 40 Global Markets.

  42. When we look at global panels we have a whole new world. Developing markets show us the universe in creation. Ageing panels help us forecast the future. As we create our model we can add within country comparisons just as if they were within mode comparisons here in the states. If our model is truly progressive we will add each new market and compare it to the whole. All we need is the data………… It’s like looking at the stars…..

  43. Distribution of Buyer Behavior Segments Homogeneity, Stability, Predictability and Reliability

  44. Comparison of Buyer Behavior Segments Among Panels by Country

  45. THANK YOU! 200 Carleton Ave. East Islip, New York 11730 631-277-7000 800-645-9850

More Related