Is it the real truth or social desirability talking?

Social desirability bias isn't often discussed in depth in the context of market research, but it deserves more of our attention. By 'our' I mean those who design and conduct research and also those who examine and use it. Failing to consider and account for this type of bias places a huge question mark over the validity and reliability of market research and the decisions made from it.

What is social desirability bias?
Social desirability bias reflects people’s efforts to present themselves to others in a positive manner. 

---

It turns up in the ​form of over-reported attitudes and behaviours that are perceived as socially acceptable. For example, someone might over report how often they make donations to charity or say that they recycle household waste when they actually don’t. It also shows up as under-reported attitudes and behaviours that are socially undesirable, such as not admitting to infidelity in marriage or under reporting the amount of illicit drug use they have been engaged in.

Psychology researchers have found that social desirability bias is fuelled by two components.

1. Impression management. This is where we deliberately alter our responses to boost our image. We might say we earn $100k per year when we actually earn $60k, because it will make us look more successful to other people. Or we might say we donated the equivalent of a week’s pay check to help the victims of a natural disaster when we actually didn’t donate at all, because if we told the truth we fear we would appear shallow.
2. Self-deception positivity. This is where people present an overly favourable, optimistic view of themselves that they believe is true, but it may not pan out in reality. For example, a person might believe their calorie intake for the day is less than 1,200, yet when an independent, objective means of measurement is employed, they are found to be consuming closer to 2,000.  

​Are they ethical consumers, or socially desirable consumers?
Social desirability bias has some important implications for the research on ethical consumption and conscious consumerism. According to a surge of market research over the past couple of years, the majority of us in the western world have said we are eager to consume more ethically produced goods and services from more socially responsible brands. Not only that, we are also willing to pay more for the privilege. No doubt these findings have fuelled many a green venture headed by social entrepreneurs hoping to replicate the success of brands like Wholefoods.

The problem is that the majority of this research is based on consumer self-reporting likely biased by social desirability. The level of support for these goods and services has been over inflated because consumers want to make themselves look more virtuous (or less shallow) by saying they would buy them. They may also have fallen victim to self deception positivity by being overly optimistic about carrying through with these behaviours at the checkout. Either way, these markets may not be as lucrative or sizable as reports suggest.
 
Why social desirability bias goes unchecked and what can be done?
Although social desirability bias is well known and accounted for in psychological research it can often go unchecked in the world of market research. One of the biggest problems market researchers face is that the methods developed to deal with social desirability bias are difficult for commercial researchers to employ. For example, there are a variety of social desirability bias scales that can be incorporated into questionnaire designs that measure and control for its effect. Unfortunately these self report scales contain between 30 and 40 items. This can literally double or triple the fieldwork costs of quantitative survey projects, which invariably rules out their use when researchers and clients are placed under increasing pressure to do more with shrinking budgets.

Having consumers answer additional questions on top of those related to the subject of investigation also takes more of their time. This leads to increased participant refusal and drop-out rates. Dealing with increased refusal and drop-outs means over recruiting which once again leads to higher fieldwork expenses.

Even if some researchers and clients wanted to account for social desirability bias with these scales, they still may choose not to since experts have suggested that some of these scale items are potentially offensive e.g. “If I could get into a movie without paying and be sure I was not seen, I would probably do it,” and “I have sometimes doubted my ability as a lover,” or “I sometimes tell lies if I have to.” Clients are unlikely to support the use of these scales if they cause distress to consumers or tarnish their brand - no matter how much the validity of results might improve.

It is assumed that the guarantee of confidentiality is enough to ensure people don’t engage in impression management in research. Supposedly if a person knows their responses are going to be kept confidential then they have nothing to gain by reporting false information, so they don’t. Unfortunately the promise of confidentiality won’t change the amount of self deception positivity inherent in responses - where the person is blind to their own positivity. Impression management is also very difficult to control in qualitative research where there is the added pressure to present oneself favourably, not only in front of the researcher but also in front of other participants.

Some have claimed that digital research methods like mobile ethnography and online focus groups can minimise social desirability bias because the researcher has less impact on consumers and consumers enjoy more anonymity. But once again they are really referring to impression management. Digital research methods are unlikely to help minimise occurrences of self deception positivity.
 
What does work?
Indirect questioning has been found to limit the occurrence of social desirability in self reports. Asking someone to report from the perspective of another allows them to project themselves out of any limits and bounds placed on them by who they are. Instead of asking “What is the likelihood that you would pay more for goods and services that have been ethically produced?“, ask “Do you think that John/Jane, who is a typical grocery shopper at your supermarket, will pay more for goods and services that have been ethically produced?”

Ultimately though, the best method of dealing with social desirability bias is to measure actual behaviour rather than rely on self reports. Do accompanied shops, ask participants to save and share their supermarket dockets, look at sales data. Once you have a clear idea of the behavioural landscape, then you are in a much better position to ask targeted questions that will help you understand attitudes, motivations and desires. You are also in a much better position to challenge consumers when their reported attitudes and behaviour aren’t in line with actual behaviour.