In my humble opinion

Charles McCabe of the San Francisco Chronicle once said; “Any clod can have the facts, but having opinions is an art.” As a scientist, you won’t be surprised to know I don’t agree and much prefer Albert Einstein who said that; “Few people are capable of expressing with equanimity opinions which differ from the prejudices of their social environment. Most people are even incapable of forming such opinions.” Uncle Albert offers a rather a damning account of the opinions we often hold so dear, and together the quotes illustrate the divide often seen between the media and science. I also like another Uncle Albert quote that; “Common sense is the collection of prejudices acquired by age eighteen” although I would argue that we go on seeing our opinions as improving common sense as we get older, but that we have probably just firmed upon and created new biases.

The idea that opinions are sculptured implicitly from our environment (with all of its biases) echoes my thoughts around the role of the unconscious in forming what we then think are conscious, rational thoughts. It is quite hard to consider, in a world where personal agency is so highly valued, that much of what we think and value is constructed by billions of ancient patterns within the neurons of our ancient brains. When we give an opinion, no matter how well thought out we may feel it is, we automatically employ and expose the ancient software which actually runs our brain. They are patterns our modern living still uses to create new heuristics for managing our complex lives and with it our personal people preferences (biases). To deny the role of the unconscious is to deny the majority of the true explanation for our behaviour.

Depending upon your out of work interests, I find that working with bias is a bit like being an English football fan, a political activist or a celeb watcher; everyone has an opinion and everybody thinks theirs is the right one. Not only that, everybody thinks you should give that opinion so they can then contradict or disagree with it. As soon as people find out that I work as a psychologist with personal biases they want my opinion on the latest news story, or more often want me to confirm their opinions on a particular social group as being particularly stupid, promiscuous, untrustworthy or work shy. They really don’t want me talk about how bias is formed and defended. They certainly don’t want to hear that we are all biased, because our bias blind spot finds that uncomfortable listening. This is certainly not the way to make polite dinner party conversations. We often have to agree to disagree, or at best they leave slightly bewildered that a discussion they thought was about confirming their views on immigration ended up looking at brainwaves and picking around amongst the pages of the Daily Mail for examples of unconscious bias triggers. I think it was Francis Jeffrey who said that; “there is nothing respecting which a man may be so long unconscious as of the extent and strength of his prejudices.” We really cannot see bias in ourselves.

I occasionally get invited to do media interviews around my work on bias. I say occasionally because the offer is soon retracted when I tell them I have no opinions to share. They don’t want the bare facts I have to hand, they want an opinion about the facts with a twist of an agenda or angle thrown in. That is their job and that is their world. But it isn’t mine. That is how a simple press release and radio interview talking about levels of unconscious bias towards non-dominant groups can quickly descend into a ‘compare and contrast’ process like some 1970’s ‘O’ level English literature essay. “So, you’re saying that more people are homophobic now than racist”? No, I didn’t say that, I said levels of implicit and unconscious bias towards Gay and Lesbian people in this study were more prevalent than the levels of unconscious bias toward Black people. They are both at levels where we need to do something quick and smart. I didn’t use the words homophobic or racist. I now never use words like racist and homophobic in describing people because it serves no purpose except to entrench attitudes I want to help people manage. In fact, I have no idea what ‘racist’ or ‘homophobic’ mean to the listening audience except the science tells me it creates the anxiety and anger which reduces our capacity to act in an unbiased way or manage our biases. I won’t be using those words or adding fuel to this already raging inferno of ill informed debate. Exit radio stage door left. No free station mug, no friendly smile and wave, no promise of a call should they need another ‘expert’ in the area. Just a look of disappointment that I didn’t play nicely enough to give them the ill thought through opinion they sought. It is tempting to comply, but to be frank I think there are enough half baked, ill considered opinions out there on bias, without me adding one more based on conjecture.

The fact we struggle to explain or justify our social opinions (we may think we can, but believe me, people are just being polite!) suggests we are opening an interesting window to our unconscious. Although we may rationalise our social opinions as conscious thought, in reality I believe that much of what we see as rational thinking is simply reconciling the outcome of our unconscious processing (or our pre-conscious to be more precise). As William James said; A great many people think they are thinking when they are merely rearranging their prejudices.” We convince ourselves that our view of the world is THE view.
We are instinctively awash with opinion on our individual differences, but corporately we seem to speak as one; everyone is equal, everyone is lovely and valuable. How did we reach that miraculous consensus? Of course we didn’t. Fearful of the outcome of even having the ‘difference’ debate organisations batten down the hatches, produce policy after policy and forbid discussion about our differences except in uber controlled but toxic ‘awareness workshops’ which often make majority and minority groups feel awkward and guilty (and perhaps cry a bit to make the trainer feel valuable). With spectacularly little impact. For example, progress in improving the representation of women on FTSE boards has stalled in the last few years and any increase in the numbers of people from poorer and BaME backgrounds reaching the top universities has to be achieved through something akin to violent hand-to-hand combat across the media. Turkeys just don’t vote for Christmas. Of course the shop fronts have changed, a few new signs and a lick of paint, but behind the frontages the same crumbling buildings and streets are the reality; the supporting structures in society are largely the same. Having a pervasive corporate opinion gives us attractive shop fronts and avoids the cost of a major redevelopment project to change the structure.

US Attorney General Eric Holder said before the election of Obama that the US needed to end the taboo of talking about race, to have an open, honest debate and resolve to deal with the issues instead of ignoring them. This could be said about many of our individual differences in the workplace. This inability to have an informed debate about how we all differ and how we all see other groups and think others view our groups seems to me to be at the heart of dealing with some of the issues of bias at work and in wider society. It is most difficult at work because there is often the ‘right’ corporate opinion and dissenters to that risk career death by a thousand sharp in-takes of breath. And the research evidence is clear that bias suppression creates a rebound effect of making bias worse not better.

The challenge now is how to have that debate, how to understand and express our opinions, how to understand why we think and feel as we do. I like quotes, so here is one to finish from Bernard Baruch: “Only as you do know yourself can your brain serve you as a sharp and efficient tool. Know your own failings, passions, and prejudices so you can separate them from what you see.” Not a bad start?

Non-dominant group networks

This is a brief summary around social networks at work, and around BaME networks which I am looking at in support of a new Diversity Netwroking Analysis tool I am working on.

Any organisation, from small teams to the largest corporation holds within it a social network or networks, largely invisible but none the less influential in the success of the organisation and the careers of those within it. Business restructuring has resulted in organisations with fewer levels of management, and more permeable functional boundaries. In these flatter, more organic and less formal organisations the impact of informal networks of relationships rather than formal reporting structures becomes ever more important to organisational and personal success ( Cross, 2011). This is set against a background of more global marketplaces and labour forces, and of more diverse teams. The notion that minority and majority groups staff members may have different informal networks therefore has the potential to impact careers and business effectiveness if these networks do not promote their own and business needs effectively.

One of the key  problems faced by women and BaME groups  in organizational settings is limited access to or exclusion from informal interaction networks (DiTomaso, Thompson, & Blake, 1988; Fernandez, 1981; O’Leary & Ickovics, 1992). These networks often control resources that are key  for job effectiveness and career advancement as well as  providing benefits such as job and social support (see Tichy, 1981, for a review). Limited network access, therefore, produces many disadvantages, including restricted information around what is going on in their organisation and difficulty in forming alliances.  These are associated with limited mobility and “glass ceiling” effects (DiTomaso et al., 1988; Morrison & Von Glinow, 1990; O’Leary & Ickovics, 1992).

Ibarra (1993) suggested that the organisational context in which social networks are embedded produces constraints on women and racial minorities, causing their networks to differ from those of their white male counterparts in composition and characteristics of their relationships with network members. Context was hypothesized to affect personal networks directly. Ibarra suggested that women and minorities be seen as active agents who make strategic choices among limiting networks. Ibarra (1995) examined differences between the networks of high potential minority group managers when compared to majority (usually dominant) groups High potential managers. BaME  managers had more racially heterogeneous and fewer intimate network relationships. Within the BaME  group, differences in advancement potential were associated with different network arrangemnets: high-potential individuals integrated  same and cross-race contacts; others had networks dominated by ties to dominant group members. High-potential BaME managers also had more contacts outside their groups, fewer high-status ties, and less overlap between their social and functional circles compared to White managers.

Evidence abounds that the people decisions in organisations are subject to the biases of those already within the organisation. There is also evidence that who gets selected, promoted, supported or fired depends not so much on the conscious, reasoned policies of the organisation so much as our unconscious people preferences (biases). This can be one factor leading to an under representation of some groups in senior positions ( Vasista, 2010). For example,  in 2009 only 12.2% of directors on the FTSE 100 boards were women, the number of companies with female executive directors fell to 15 (from 16 in 2008) and the number of boards with multiple women directors fell to 37, from 39 in 2008 ( Sealy et al, 2009). The role that informal networks play in progression cannot be under estimated. Early warning of a vacancy to allow career positioning, support through coaching or advocacy, exposure to the right high profile projects and challenges all impact career progression but much of this operates an at informal level. Vasista (2010) reviewed the evidence that bias had impacted Black and Minority Ethnic (BaME) staff and the outcomes of that bias:

  • In 2007, the business-led National Employment Panel reported that up at least 25% of the ‘ethnic minority employment gap’ (the difference between how many Black and minority ethnic people are employed compared to the general population) is caused by discrimination in employment practices (National Employment Panel 2007).
  • One-third of Asian and 20% of Black managers surveyed say that racial discrimination had been a barrier to succession (Hooker et al., 2008).
  • When subjected to CV testing ( identical CVs submitted with varying dominant or non dominant groups names), private sector employers showed a discrimination rate of 35%

Compared to 4% for the public sector (Wood et al 2009).

  • 70% of 300 professional Black and minority ethnic women surveyed by The Diversity Practice said they had experienced at least some discrimination based on their race, and 65% because of their gender (2007).
  • More than a fifth of individuals surveyed by Race for Opportunity said they had been offended by a racial remark in their place of work, with Chinese respondents the most egregious victims with 35% citing an example, followed by a quarter of Pakistani respondents (Race for Opportunity, 2010).

Although Ibarra (1993, 1995) put forward various theoretical models for how minority group members networks adapted to the constraints, accessing the exact social relationships which exist in a particular team or organisation has traditionally required detailed qualitative study.   How exactly the informal networks which operate within organisations may operate for different groups in a given context and may impact the career experiences of non-dominant groups such as women and BaME staff is usually unknown. Insight to how the social networks are impacted by aspects of diversity may afford the opportunity to identify differential networks experiences and to act. Subtle biases manifest themselves in micro-behaviours around our relationships with others. This may include taking less interest in their work and life outside work, having less social contact and sharing less with people who are unlike us.

Careful now 2 ; the dangers of tackling unconscious bias for the unwary


Unconscious bias (UB) is now high on the corporate agenda of those looking to enhance organisational performance by engaging better with customers and by getting the most from staff. The market for awareness raising, e-learning and other interventions has grown exponentially. At times it seems like an race to be the first to use the latest UB training, testing or coaching tools.

This growth, urgency and the often shallow knowledge in application disguises a risk; UB can be seen as a seductively easy solution to part of a very real problem. The argument goes that if we make good people decisions we will understand our customers better (true), our staff will be happier (probably true) and our business will run better because we will have less conflict (probably true too). UB passes the elevator test in that it can be explained in minutes and its neurological base resonates with senior managers looking to move away from the often restrictive regime of the ‘politically correct’ diversity consultant.   However, what many do not realise is that using poorly thought through unconscious bias interventions may actually trigger or increase bias rather than reduce or control it. Interventions have to be targeted and informed.

It’s about wi-fi

The core problem is that many now trying to us UB management interventions continue to see it as a conscious process, as if we can apply conscious solutions to an unconscious problem. This mind set is at the core of application problems. What the brain experiences as neurons firing together, it quickly wires together neurologically in our pattern matching for future use. Unconscious bias (UB) is really about this automatic pattern matching capability and the human brain has developed a unique capability to respond not just from experience but from what we see and hear from others (including the media). Our unconscious truly is the basking shark of the neurological world, sifting and filtering vast amounts of data for patterns. We should not be surprised that occasionally it absorbs patterns we would rather it didn’t, such as negative associations around some social groups which reinforce existing social stereotypes and which can impact our behaviour. This pattern matching can have unexpected and unwanted effects for those using bias interventions. If we offer our brains stereotypical imagery it will employ its patterns or create new ones reflecting that matching. It makes no assumptions about the intent of the user. Assuming that somehow our clever conscious mind can routinely over-ride this primitive pattern matching is a mistake. It is true that forming a conscious intent can help mitigate bias but if the experiences and materials we use to help the conscious reduce bias contains stereotypes (for example discussions about group stereotypes), then we should expect it to trigger our stereotypes and reinforce the pattern matching.

Bias suppression

Suppression is the anti-bias weapon of choice in most organisations. Just as discussing and priming stereotypes leads to their activation, suppressing stereotypes can lead to subsequent activation. Telling people not to use stereotypes may work in the short term but as soon as the control is removed or attention waivers the stereotypes bounce back, often stronger than had we not suppressed them. It is thought that the act of suppression triggers the stereotype, which we consciously control, but that eventually our resources are depleted and the bias exerts itself. Unless we know accurately what our biases are, so that we can be vigilant with certain groups in particular situations, it is nonsense to expect people to self monitor for every bias and stereotype in all situations.

Toxic sheep dipping

People with low bias towards a group are particularly susceptible to triggering dormant pattern matching. This can be the risk with implementing UB interventions without knowing the bias levels of an individual (which will differ with regards to a range of groups). In a recent study involving a range of tools developed to help reduce bias ( Jones, 2011) it was found that they were highly effective with people who already had strong biases but were as likely to increase as decrease bias in low bias people, it is argued because the materials used triggered groups stereotypes not usually accessed. This highlights the danger of UB interventions without knowing the existing bias levels of an individual, and involving them in discussions and even coaching around biases they don’t have. The power of triggering stereotypes in low bias people (who generally form around 70% of the population) has been repeatedly seen in research and particular care needs to be taken with these people. E-learning can be a welcome addition to the bias management toolkit, but again the content should avoid using specific biases if low bias people are not to have their biases triggered.


If stereotype suppression simply triggers bias, and even discussing bias triggers bias, what are we to do? The obvious answer is not to try and suppress bias, but also try to minimise the impact of our interventions by using methods and tools which allow our natural bias control mechanisms to operate fully. At the core of this is self awareness and awareness of the existing bias levels of individuals as the effects described above are most prevalent in people with low bias . As individuals, having an awareness around our personal biases affords a great advantage. If we know that our biases are more active around some groups of people than others, and in particular situations, we can be more mindful about our behaviour with those groups at those times. The scattergun approaches of some training providers and consultants, in the mistaken belief that knowing more has to be good, is part of the problem and not the solution, which lies to well informed, knowledge led interventions which take account of what research has told us.


Jones, P.C   (2011) Bias reduction at work; first pilots of a battery of bias reduction tools for organisational settings. British Psychological Society DOP Conference Stratford Upon Avon January 12th-14th. 2011

Careful now: Understanding the reality of bias testing


The growth in interest around our unconscious biases at work has led to more training, coaching and in particular bias testing in the area. This is a welcome development for those frustrated by the apparent loss of momentum in achieving real change in organisations. Unconscious bias is now not only on the lips of diversity and inclusions managers but also on the minds of senior managers looking to enhance organisational performance by reducing conflict, engaging better with customers and by getting the most from staff. Whilst we may welcome this greater awareness of our often unintentional and subtle biases there are pitfalls awaiting the unwary. Implementing bias testing requires a different consideration than that normally employed when considering personality or ability testing, in three key areas:

Test Choice

Bias tests come in a range of options in terms of content and delivery. Most bias tests use a variation on the implicit association test (Greenwald et al., 1998) which bypasses any lack of insight or willingness to disclose by tapping the implicit and unconscious associations people may hold about particular social groups. IATs come in many forms but the key distinction is between norm referenced and criterion referenced scoring. Norm scoring will place an individual within the wider population but say nothing about the likelihood of an individual’s unconscious associations becoming behaviour ( Blanton and Jaccard, 2005). Criterion referenced tests are linked to the predilection for both subtle and less subtle behaviours such as bad mouthing, avoiding or joke telling.
Suppliers often offer metrics on the basis of individual differences such as Age, Disability, Ethnic Origin, Faith/Belief, Gender, Nationality, Sexual Orientation and a host of other differences. Prospective test users should be clear about the purpose of testing and whether it is to address particular organisational issues, or as part of a personal development intervention (or both). Testing needs to be both relevant and proportionate. As an organisation, test users might confine testing to areas where they know they may have an issue (e.g. based on staff under-representation, lack of group progression, poor sales to that customer group or particular incidents or complaints). At a personal level test takers might be afforded wider test access to support their personal development (e.g. the full test range). They may also have follow up testing ( e.g. a Black Woman test) once preliminary testing (e.g. a Black Ethnic Origin test and a Gender test) suggests a possible issue for clarification. Knowing that tests are available for such follow up is an important part of test choice.

The engaging nature of IAT tests, the ease of online completion and the apparent simplicity of IAT can lead to test takers rushing into testing without being properly prepared. Proper briefing and test practice is critical if results are to be trusted, as test retest reliability can fall from around .80 to .50 if test takers are learning on the test. Allowing direct individual access to testing and feedback, although bypassing many of the problems listed below around data protection and feedback, also removes the capacity to manage what and how test takers understand and use the test. Unsupervised testing like this can undermine the measurement, credibility and organisational importance of bias testing. It can reduce bias testing to entertainment or,commonly, simply provide confusing materials for posting on social networking sites.

It should be noted that some ‘tests’ bill themselves as measuring our personal biases but use more explicit and conscious sorting tasks similar in appearance to the IAT. They require conscious sorting of stereotypical photographs. Their use can be a triple whammy, because not only do they have no basis in what the psychological research tells us, but they may trigger and reinforce stereotypes. In addition they can alienate prospective users and takers of valid and reliable bias tests who recognise their potential to increase bias through stereotype triggers.

Data Protection

Such is the political sensitivity around bias, created often by well meaning but poor advised internal diversity staff, that many test users find that they have to employ some sort of firewall between the organisation and the personal data. This often requires one or even two levels of external consultants so that there is no question of data being seen by employers. The length of time personalised data is kept is often minimised to between 7 and 28 days.

Feedback and follow up

One of the strongest criticisms of the Greenwald et al IAT is around feedback. Norm based feedback tells us little about how a person may behave, and there is no metric-to metric data in terms of what test scores and changes in test scores mean ( Blanton and Jaccard, 2005). Unsupervised testing where feedback is automatically generated can be misinterpreted by test takes. For example, The Greenwald et al IAT may tell test takers that they have a ‘Strong Preference for White people over Black people’ which test takers often interpret this as meaning they are ‘racist’, which may or may not be the case. This can have profound effects on an individual.

Individual test takers can also be very sensitive to feedback and the nuances of words can become critical. Feedback has the potential to trigger hitherto dormant stereotypes, particularly in those with low bias. Research suggests that even using the work ‘prejudice’ can trigger adverse test taker reactions and increase bias, especially in low bias individuals. Many bias test users prefer to use the word ‘bias’ in feedback rather than prejudice but some have gone further and use terms such as ‘unintentional people preferences’.

Having an idea as to what the organisation and the test taker will do with test results is as critical as making the right test choice. Identifying an issue for the individual or organisation is useful, but the test users must be clear about what support there will be in response to results. Research shows that traditional ‘sheep dipping’ of individuals in bias management training is often ineffective and may even trigger bias in those who begin with low bias. Planning for bias testing therefore must include some thought around subsequent interventions, which may have to be personalised to get the best effect.


Bias testing can be seductive and has the potential to make organisations more effective Implementing testing requires full consideration of all of the issues and implications. Poor thought through or in appropriate bias testing can actually increase organisational bias within the organisation, particularly if test choice, data protection, feedback and intervention are not part of the planning process.