Dana Sparber of NBCUniversal unveiled the results of an online survey investigating inflation and the multicultural shopper. It was conducted using a Numerator panel in Q4 of 2022. Among the 7,000 surveyed were a mix of Asian, black and Hispanic shoppers, all adults +18, half men and half women. These consumers had all bought items in the categories of grocery, health and wellness, beauty, household products, non-alcoholic beverages and electronics. The study not only looked at online shopping but in-store and bodega sales—an important aspect often missed. Multicultural consumers across the board and regardless of household income had a much sunnier view of the state of inflation and were far more likely to say they would continue their normal spending habits. All projected the common theme of resiliency.
Steve Keller (SXM Media) looks at opportunities for sonic interventions: What are things in the world, in our culture, where sound can provide a positive solution? Sonic intervention, for the purposes of this research, starts with the concept of the color line, first addressed by W.E.B. Dubois in his 1903 collection of essays, “Soul of Black Folks.” For Dubois, the color line was the dividing line between Black and White individuals. A century later, SUNY Binghamton professor Jennifer Lynn Stoever defined the sonic color line as the hierarchical division between the whiteness and blackness of sounds that have been created and perpetuated by a dominant culture on the listening ear. Indeed, in the 1920s radio show Amos ‘n’ Andy with 40 million listeners, two white personalities performed racialized sonic tropes. Black radio performers were forced to play these stereotypes. Fast-forward to today, the sonic color line still cuts through our technology, our smart speakers, and often in the studio Black voiceover actors are asked to sound more “urban”. Yet the error rate in recognizing prompts from Black speakers is significantly higher than the rate for White speakers.
This ARF Pharma Council event followed up on the Council’s podcast episode on “Inclusive Futures of Humancare,” focusing on the importance of inclusiveness in pharma research and marketing with respect to both demographic characteristics and health conditions. Four speakers delivered brief presentations, followed by a discussion moderated by Pharma Council Co-Chair Marjorie Reedy of Merck.
Steve Keller – Sonic Strategy Director, Studio Resonate, SXM Media
Steve Keller (SXM Media) looks at opportunities for sonic interventions: What are things in the world, in our culture, where sound can provide a positive solution? Sonic intervention, for the purposes of this research, starts with the concept of the color line, first addressed by W.E.B. Dubois in his 1903 collection of essays, “Soul of Black Folks.” For Dubois, the color line was the dividing line between Black and White individuals. A century later, SUNY Binghamton professor Jennifer Lynn Stoever defined the sonic color line as the hierarchical division between the whiteness and blackness of sounds that have been created and perpetuated by a dominant culture on the listening ear. Indeed, in the 1920s radio show Amos ‘n’ Andy with 40 million listeners, two white personalities performed racialized sonic tropes. Black radio performers were forced to play these stereotypes. Fast-forward to today, the sonic color line still cuts through our technology, our smart speakers, and often in the studio Black voiceover actors are asked to sound more “urban”. Yet the error rate in recognizing prompts from Black speakers is significantly higher than the rate for White speakers.
All of the above is the background for Steve’s research in which he builds a business case for sonic diversity in radio, podcasts and other audio platforms. To the best of his knowledge, this is the first large-scale sonic diversity study of its kind: the impact of racialized listening on advertisement favorability and effectiveness. In two experiments (one a large field experiment, the other a quantitative experiment with panelists) researchers used three podcasts—Sway’s Interviews (interviews with black entertainers), Song Exploder (about deconstructing songs, a varied audience) and This American Life (predominantly white program and audience). The study used 20 voice actors divided evenly black/white/male/female, who were not coached, and who were paid as if they were working on a regular ad to provide a conversational read from the advertising scripts, for 60 sonic stimuli total. Partnering with Veritonic the researchers asked participants (4,000-plus sample size) questions about favorability and attributes in the ad, and intent to listen, and whether they thought the voiceover was “definitely,” “probably” or “not sure” either Caucasian or Black.
A second experiment analyzed advertisement effectiveness: Would there be a rise in podcast listening as a result of the ads?
Key Takeaways
Whereas White actors were consistently perceived as White, there was a lot of misattribution for Black voices. This is probably due to “sonic markers,” i.e., what we listen for. And the race of the listener influenced the perception of the race: Black listeners were much better able to correctly identify the race of the voice actor, although not always.
Context matters: Misattribution occurred when there were other potential contextual cues, for example, Black voices were more often correctly identified when advertising for Sway’s Interviews than for This American Life.
Black voice actors received higher aggregate scores, even from White panelists, even breaking down into categories of likeability, empowering, trustworthy or intent to listen to the podcast. Whereas Black respondents tend to rate ads higher when the voice actor is perceived to be Black, with White respondents there was little to no difference.
Neither race nor gender had an impact of ad effectiveness, although in the second experiment there was a rise in podcast listening because of the ads. So, if we’re hearing more Black voices, we’re creating the ads that have the potential to be judged more favorably by Black audiences without necessarily any negative impact on the effectiveness for any consumer group.
On November 1, 2022, the ARF Cultural Effectiveness Council hosted a discussion on ways that brands have gained an edge by focusing their insights and subsequent marketing on traditionally under-represented communities. Speakers from General Mills, Vevo, and muliti-cultural agency Alma shed light on the kinds of research they find helpful in uncovering valuable inclusive insights and the potential rewards of this strategy. They cited examples of those insights and discussed the societal trends that underlie them. The session was moderated by Council Co-Chair Janelle James of Ipsos.
On October 25, 2022, media and measurement experts shared the latest research-based insights about today’s media landscape at OTT 2022 held on the Warner Bros. Discovery Studio Lot in Burbank, CA. Attendees learned more about what viewers really want when it comes to FAST, AVOD and SVOD, and whether there are viable solutions for cross-platform measurement.
Ads that highlight under-represented groups and their experiences can have broad appeal and be highly effective. However, they can backfire – even among the group the ad is focused on – if they fall into the “sadness trap”.
May 5, 2022 — Episode #4: Contending with Algorithmic Bias, a conversation with, Janelle James, SVP, Ipsos IUU (Council Co-Chair), Kalinda Ukanwa, Assistant Professor of Marketing at USC’s Marshall School of Business. Amanda Bower, Machine Learning Researcher at Twitter. Hui Wang, VP/Director of Global Data Intelligence – Analytics Service, at Publicis Media. Kathy Sheehan, Senior Vice President at Cassandra. Ilinca Barsan, Data Science Director at Wunderman Thompson.
We’re increasingly dependent on algorithms in a variety of areas. This has led to problems with algorithmic bias, in which statistical and econometric models or a programmed set of instructions systematically treats members of some groups differently than others. This can be due to the unconscious biases of engineers who build the models, biases in the data they are trained on, biases inherent in the models themselves or algorithms’ treatment of human attributes as single, isolated components rather than intersectional identities. We look at the ways in which models, programs, and algorithms in the media, marketing, and advertising industries can unintentionally favor majority populations and ignore or even discriminate against minority segments and how to overcome this problem.
June 17, 2022 — Episode #5: Contending with Algorithmic Bias pt 2 , A conversation with, Kalinda Ukanwa, an Assistant Professor of Marketing at USC’s Marshall School of Business. Amanda Bower, a machine learning researcher at Twitter.
Technology is an essential part of our everyday life. We rely on it to help us navigate our finance, employment, housing, health care and purchases, among other things. But, perhaps unknowingly, we may be impacted by algorithmic bias, in which systematic repeatable errors in AI treat members of some groups differently than others. In part two of our look at algorithmic bias, we delve further into what consumers and companies can do about it.
Many companies have employed prosocial messaging and more diverse portrayals in advertising. New research explored if and how those strategies can improve consumers’ perceptions of brands and increase sales. Read more »
