Bias. Who, me?

No method to study thought, attitude and behaviour is ever free from bias. So how do we address it in our research?

Our social environment powerfully shapes behaviour. We are constantly influenced by social cues that tell us what we can and cannot express and how we should behave in a given situation.

This is the case when doing research, just as much in everyday life. Although an awareness of social bias is nothing new, all too often it is forgotten in practice. When a researcher asks a sensitive question (“How often do you drive while being intoxicated?”), for instance, many respondents under-report undesirable activity. Self-report methods are especially likely to be skewed by this phenomenon of social desirability bias.

People self-edit their responses, both consciously and unconsciously. And as researchers, we are just as much to blame in how we choose to interpret those responses.

Techniques have been developed to bypass or uncover mental correction processes, such as efforts created to decrease systematic error in self-reports. One example is the bogus pipeline method. It’s a devious trick: the researcher hooks the participant up to some electrodes and tells the participant they are attached to a polygraph machine: a lie detector. In reality, it’s just an empty box. It turns out that if you think you will be caught telling a lie, you are far more likely to tell the truth. Researchers apply this accountability hack in a variety of ways, but it’s not very practical every time you have to ask someone sensitive questions.

Other more subtle approaches exist. Social psychology and behavioural economics tell us that attitudes (preferences) are relative, social and contextual - not absolute. For example, social norms sign appropriate (and not so appropriate) behaviour. When people were told that “80% of your neighbours consume less electricity than you”, the majority subsequently brought down their energy consumption (Cialdini & Schultz, 2004). Or, when someone expresses a dislike of gender stereotyping, you probably won’t tell her that stereotypical joke.

Psychologists test people’s explicit (what you say) and implicit (what you think before controlling it) response. It boils down to our fundamental desire to maintain a positive view of ourselves. According to the Theory of Grounded Cognition (Barsalou, 2010), cognition is thus not purely rational and ‘unlocatable’, but emerges from the interplay between brain, body and world.

So when out in the field, how do we, as researchers, study both the internal and external landscapes of human experience with – or indeed without -social bias in mind? With 7 out of 10 people in the UK between 16 and 64 owning a smartphone, mobile ethnography is becoming a popular approach to observe cognition and behaviour in their natural surroundings, and as an alternative to the strongly biased self-report measures. It allows for the observation of a wider set of behaviours than, say, the lab or focus group.

But still, mobile ethnography is not free from bias either. Because really, no method to study thought, attitude and behaviour is ever free from bias. People self-edit their responses, both consciously and unconsciously. And as researchers, we are just as much to blame in how we choose to interpret those responses. The task, then, is to reflect on what the individual’s intent is and in what context the responses were obtained, all the while doing so with an awareness of our own backgrounds, assumptions, and personal experiences. The challenge is to unpick the self-guided narrative through this, along with all of its contradictions, gaps and inventions.

Because environment shapes our cognition and behaviour, and because this applies to both people out in the field and researchers themselves, it calls for a mindful approach to research. All research is eventually context-dependent: it should be interpreted as a social constructive process rather than clean data collection. Although the social sciences have debated and made great strides to acknowledge this, the questions remain: how do we observe individuals without inducing a strong sense of self-awareness? How do we motivate people not to self-edit their responses? And once even those questions are answered – if that’s ever fully possible – how do we as researchers interpret what we experience in the field with the neutrality and sensitivity that it demands?

If you’d like to be a part of this conversation, please get in touch with us by email or on Twitter.

Mindset reminders:

  • Both bodily cues (including emotions) and the social environment powerfully shape behaviour. Mind, body and world interact and influence our perception, action, memory, knowledge, language, and thought, resulting in imperfect rationality (or in other words: bias).
  • Even with an acute awareness, subtle biases sneak into the way researchers and participants engage in research.
  • These biases call for a mindful approach to research, asking us to innovate approaches that help to reduce and acknowledge bias throughout our research. 


Barsalou, L. W. (2010) "Grounded cognition: past, present, and future". Topics in Cognitive Science, 2 (4), 716-724.
Beyer, H., & Liebe, U. (2015) "Three experimental approaches to measure the social context dependence of prejudice communication and discriminatory behaviuor". Social science research, 49, 343-355.
Cialdini, R., & Schultz, W. (2004) "Understanding and motivating energy conservation via social norms". William and Flora Hewlett Foundation.
Lee, P (2014) "Deloitte, Mobile Consumer 2014 Report"
Neeley, S. M., & Cronley, M. L. (2004) "When research participants don’t tell it like it is: pinpointing the effects of social desirability bias using self vs. indirect-questioning". Advances in consumer research, 31, 432-433.

Image courtesy of Bibi Groot.

Bibi Groot