Much of our knowledge of good questionnaire design is based on decades of research in social psychology on attention, memory and information processing. Over time, we have learned how to interpret the results of global studies in a way that takes into account other psychological findings such as acquiescence bias or extreme response bias, but what about the underlying concepts themselves? With 96% of the world’s psychology research being conducted on 12% of the world’s population (and most of it in the Western world), it’s possible some of our most fundamental concepts are going uncontested.

Social psychologist and fellow of the Society of Consumer Psychology Sharon Shavitt from University of Illinois, Urbana-Champaign, recently suggested a new theory of attitudes that highlights how differently they function in Western and non-Western cultural contexts. Reading the article, I was struck by the fundamental challenge this new theory is posing to us as an industry, so I wanted to ask her more questions.

You recently wrote an article (Preferences don’t have to be personal: expanding attitude theorising with cross-cultural perspective in Psychological Review) that quite radically challenges most of what market researchers see as conventional wisdom about attitudes and how we should measure them.  How would you describe what our attitude has been towards attitudes? What are our biggest misconceptions about attitudes and their measurement?

Shavitt: In our cultural context, our attitude toward attitudes is very positive. We like attitudes. We cherish our attitudes. They define who we are as individuals.

This positive attitude toward attitudes is no less true in the research literature. In 1935, one of the giants in social psychology, Gordon Allport, wrote that “attitude is probably the most distinctive and indispensable concept in contemporary social psychology” and today multi-volume handbooks are devoted to this topic. Western theorists have viewed attitudes as stable and consistent properties of individuals. Attitudes also define who we are – that is, they are self or identity expressive.  

In your article you talk about how, in non-Western contexts, attitudes can be context dependent and change depending on the situation, whereas Western research methods implicitly assume that attitudes and preferences are more or less constant over time. And that additionally, people in non-Western contexts can often hold attitudes that are more inconsistent, even within one person. What causes these differences and do they apply to all product categories in a similar way?

Shavitt: Yes, you would expect that attitudes in non-Western contexts would be less stable and internally consistent. In Western contexts, personal preferences are central to attitudes (“I like it, because it makes me feel good”). But, in non-Western contexts, there is a stronger emphasis on connections with others and within groups, which brings with it an emphasis on behaving appropriately and fitting in with others’ expectations. You wouldn’t want rigid and unchanging attitudes in that cultural environment. Such attitudes would get in the way of fitting in and being socially appropriate. Instead, you would want malleable attitudes that are responsive to others’ view (“I like it because others I am connected to like it”).

Related to this, it’s also worth mentioning that contradictory opinions are perceived as more acceptable in non-Western cultural contexts. So, people in non-Western contexts may feel more comfortable expressing different attitudes over time and across situations. In the West, doing that can be considered hypocritical or, at best, flaky. 

We think these points apply broadly across types of products. Research has shown cultural differences for a wide range of products and topics of attitudes. One could speculate, however, that these differences will be stronger the more socially visible the product or more likely it is to be consumed with others. For such products, norms may play a stronger role in non-Western contexts.

You also mention that the differences in Western and non-Western ways of thinking result in different patterns of survey responding.  While researchers are aware of things like acquiescence bias, what other things should we take into account when designing surveys that span multiple markets and cultures? Under what circumstances can marketers and researchers trust the findings from non-Western contexts? And how can we, as market researchers, make things better?

Shavitt: I think that if we use methods of surveying that are designed around Western ways of thinking, we shouldn’t be surprised when those methods provide unreliable evidence about non-Western consumers.

One recommendation would be to measure and map attitude variability among non-Western consumers. Their attitudes toward an object are expected to differ across social contexts. For example, attitudes toward having soup for dinner with a colleague can differ substantially from attitudes toward having soup for dinner with a family member. In our model, such differences are not “problems” – they are of focal interest. Mapping them can help marketing researchers to identify the use occasions that resonate for the target audience, and learn which contexts to feature in marketing communications.

Elina Halonen is communications officer at the Society for Consumer Psychology and a partner at the Irrational Agency.