Don’t Influence Me: Defeating Third Person Bias in Advertising Research
By Kelsey Howell
Image via Unsplash @davisuko
We all know someone who claims they never pay attention to advertisements but is somehow intimately familiar with popular brands and trends.
This phenomenon is known as third person bias―the false belief that "other people" are more easily influenced by mass media than yourself. It’s a challenging distortion to overcome in marketing research due to the field’s reliance on self-reported media use. Respondents may falsely report low impact from media sources they see as overtly commercial, such as television ads or influencer endorsements.
Thankfully, for every way humans can trick themselves into a thought process, they can also be guided out of it. W5 suggests several tactics to outsmart this common cognitive distortion:
Ask About Past Channel Behaviors, Not Current Channel Preferences
Image via Unsplash @kovah
When asked about their preferred sources of product information, audiences often name self-directed research and word-of-mouth. These are seen as high-brow and unobtrusive sources of information, but don’t necessarily reflect typical sources of awareness, especially in “impulse” categories like food or electronics.
Asking respondents where they have previously learned about products or services keeps these questions within a strictly factual zone. Respondents can preserve their discerning self-image by reporting on advertising as something they are passively exposed to rather than something they seek out.
Question Examples
“How did you learn about the brand(s) of dish detergent you currently use?”
“In the past two years, where have you sourced new lesson plans for your students?”
“In the past 12 months, where have you heard about new fast food products?”
This approach has limitations for a new brand or market product, or ones using an innovative marketing strategy. For these bold and fresh ideas, concept testing may be appropriate.
Use Concept Testing to Predict Attitudinal Impact
Image via Unsplash @pmvch
If the goal is to understand the favorability of a piece of advertising, another way to avoid third-person bias is to present the stimulus during the survey itself rather than ask respondents about their hypothetical willingness to be marketed to. This is known as concept testing and can encompass everything from videos to search engine mockups.
Question Example
Below is a hypothetical search engine advertisement that could appear if you searched “custom printing small business” in your area.
The Print Shop | Your Custom Printing Partner
Two-Day Shipping Nationwide. No Minimum Order. Print On Demand. Affordable And Eco- Friendly Printing For Future-Minded Businesses. Discount For New Customers.
1. On a scale from 1 to 5, with 1 meaning “strongly disagree” and 5 meaning “strongly agree,” please rate your agreement with the following statements about the above advertisement:
This advertisement is clear
This advertisement is appealing
This advertisement is helpful
I would click on this ad or look up the company if I saw it in my search results
2. What claim in the advertisement is most interesting or relevant to you?
Concept testing is excellent for understanding key takeaways audiences may gather from a campaign, such as the qualities associated with the advertised brand or the perceived call to action.
Consider Pre-Test/Post-Test Designs to Measure Behavioral Impact
Finally, the most accurate method to assess behavioral impact is the use of pre-test, post-test design. Consumers are asked a series of brand health questions before and after the introduction of a real-world advertising campaign.
Question Examples
Are you aware of this brand?
Have you previously purchased from this brand?
How recently have you seen an advertisement from this brand?
What do you remember from the advertisement?
This study design allows for a quantitative accounting of the campaign independent of both third person bias and any other predictive biases. Bringing It Together
Many studies can benefit from a combined approach to combatting third person bias.
For example, a miniature pre-test/post-test approach can be used in concept testing by asking about the respondent’s existing brand knowledge before showing the stimulus, and behavioral channel usage questions can be used to supplement either of the experimental study designs.
The right mixture, as always, is based on study goals and intent.
Want to measure your own advertising impact? Reach out to W5!