Understanding and Mitigating Social Desirability Bias

By Charles Roberts and Matthew Gadsden

Click To Watch

Introduction

In our last Practical Behavioral Scientist segment, we discussed the Say-Do Gap, which is the discrepancy between what we say and what we actually do. In that segment, we focused on how situational factors influence what we do and how we can best measure behavior. In this segment, we will focus on what people say and the potential biases that arise when individuals answer questions.

We recently sent out a survey about recycling to our Hotspex team members. They were presented with the same list of items and asked to indicate either which ones they would recycle (a direct question) or which ones other employees in the organization would recycle (an indirect question). The results indicated a higher number of recycled items when questions were asked directly versus indirectly. This is because recycling is a socially desirable activity which leads individuals to select more items from the list when asked about their own behavior to represent themselves as someone who has strong recycling habits.

Definition and Examples

Social Desirability Bias is a form of response bias in which individuals construct their responses based on consideration of other peoples’ interpretation of them rather than answering accurately and truthfully. Individuals do this to create a more flattering perception of themselves, and therefore, their response is biased towards what they perceive as socially desirable or “correct.” Social Desirability Bias leads to the overreporting of socially desirable behaviors or attitudes and underreporting of socially undesirable behaviors or attitudes. This bias can be uncovered utilizing different methods of investigation, including direct and indirect lines of questioning as we mentioned above.

A study by Fisher (1993) had students evaluate a Walkman model that had social implications due to its visibility, from either their own perspective (direct) or the perspective of a typical college student (indirect). The product's perceived objective benefits like “These wireless headphones provide better freedom of movement while working out” were less influenced by Social Desirability Bias than normative outcomes such as “I will be perceived favorably by others if I get this Walkman.” Social Desirability Bias affected evaluations of normative outcomes when asked directly but this was mitigated through indirect questions. Therefore, a combination of direct and indirect questions gave a complete picture of human behavior and the social context.

Hotspex: How Can We Help?

At Hotspex, we can help you understand the full picture and mitigate the Social Desirability Bias through our behavioral science-driven methods. We can help you answer the following questions:

1. What associations do consumers have with my brand/category?

2. What are the drivers (motivations) or barriers to engaging with my brand/category?

3. How will my assets, packaging, etc., stretch brand associations?

For example, we worked with a leading grooming brand to understand the impact of using visible minority talent on purchase behavior and brand associations. We tested two visuals, one using Caucasian talent and another with minority talent. While there was no apparent difference in the purchase behavior and brand perceptions, we discovered a difference in brand equity in our proprietary Implicit exercise. This exercise suggested that there was potentially a compensatory response to the unconscious racial bias. Using a combination of indirect (i.e., implicit psychometrics) and direct (i.e., virtual shopping environment, stated preferences) methods of measurement, we were able to successfully eliminate the effects of Social Desirability to confirm an unconscious bias existed and detect the impact on behavior.

In another example, we explored how to persuade truck intenders to consider more environmentally-friendly vehicle options. We found that although customers viewed environment-friendliness and fuel efficiency as an important factor for purchasing a vehicle, it was not translated to purchase behavior. Using both implicit and explicit methods, we uncovered true motivations behind vehicle purchasing behavior and identified ways to nudge this behavior.

Conclusion

Social Desirability Bias is the tendency of people to respond in a manner that creates a more flattering image of themselves rather than answering truthfully. This bias should be kept in mind when designing research to capture your customers’ attitudes, motivations, and behavior. Using a combination of direct and indirect methods can allow you to understand the social context in which responses occur, mitigate the effects of Social Desirability Bias, and get a full picture of individuals' attitudes and behaviors, all of which Hotspex can help you with.

What are some behaviors that may be susceptible to Social Desirability Bias for your brand?

Sources: Fisher, R. J. (1993). Social desirability bias and the validity of indirect questioning. Journal of consumer research, 20(2), 303-315.