Monday, 21 October 2013

Here comes the science bit...why it's so hard to break the Glass Ceiling

I'm slightly obsessed with a book I'm reading, 'Thinking, fast and slow' by Daniel Kahnerman, who won the Nobel Prize in Economics in 2002. The following is inspired by (but in no way authorised by or discussed within the covers of) this illuminating book.

The fundamental premise of the book is that the working of our minds is an uneasy interaction between two 'personalities' (which the author dubs System 1 and System 2). System 1 is in control most of the time, even when we believe ourselves to be 'concentrating' because System 2 is 'lazy' i.e. only engages when forced to.

 I've drawn us all a little picture to the two parts of our minds because I'm not vain enough to think you are bothering to engage your System 2 to read this blog so the easier I make this for your impatient System 1 the better, methinks.


We are all familiar with ideas such as 'the first 30 seconds of a job interview determine the outcome' but this book goes a lot (and really scarily) further than that. It summarises many long years' work which shows that not only are most of the decisions we make most of the time  (by System 1) based on insufficient data, but that, even when invoked, the more deliberate System 2 will often find reasons to back up this instantaneous decision through 'laziness'. This means that even in a 'nice' firm, where the decision makers are all consciously non-discriminatory, I may face the difficulties posed by the System 1 parts of my assessor’s mind, (I call him ‘boss-judge’ below), which include:
1. The 'Availability' heuristic (rule): We make judgements about the probability of events based on how easy it is for us to think of similar examples. Our brains operate on the basis that if we can recall something, it must be important.

Application to the 'Glass Ceiling at a nice firm': In general, given that women are rarer than men at the 'top', the average 'boss-judge' who is assessing me vs. a male competitor will find it far easier to recall successful men than successful women. In addition if, for instance, the position demands an intense level of commitment, he will likely find it far easier to recall reasons (e.g. from his own home life) why I would struggle to meet these demands than he would for a male competitor, particularly if I am within a childbearing/child rearing age range (say 25-50) i.e. exactly when I would likely most want to be progressing my career.

It must also be true, however, that should an example of a woman who is successful at the top is readily available, say at the same firm, this is likely to go some way to helping me. Similarly, if there has been a recent / recallable incident where a woman 'gave up' or 'failed' at a position citing reasons which appear at first reading to be applicable to me, this is likely to bias him further against me.

2. The 'Representativeness' heuristic (rule): We tend to estimate the likelihood of an event through comparing it to an existing prototype (stereotype) in our heads. Interestingly however, if we are immediately given information which contradicts the stereotype we use it well, significantly decreasing our reliance on the representative prototype.

For instance, if asked who is likely to be more aggressive, 'Paul' or 'Samantha?', most people would pick 'Paul', presumably due to a gender stereotype. If they are told, however, that both Paul's and Samantha's mothers commute to work in a bank, their reaction changes and Paul and Samantha are judged more equivalently i.e. this additional up-front information which upset the stereotype (not even of the characters but relating to their mothers!), made people engage their System 2s and think more carefully.

Application to the 'Glass Ceiling at a nice firm': This isn't hard. I know the stereotypes I am battling against (and poor boss-judge man won't even know he is applying them). What is interesting given all the kerfuffle and criticism against women who 'act/dress like men to get ahead', doing so is probably not unhelpful as it 'upsets' the stereotype (although there are probably a myriad other ways I can achieve a similar effect without losing my keen fashion sense).

 3. Judging based on limited evidence:
·  What You See Is All There Is: Most of the time, we all find it difficult to see beyond the evidence at hand even if (and this is the crux of it) we could very easily generate the other side / alternatives. The following is a quote from the book:
Consider the following: "Will Mindik be a good leader? She is intelligent and strong". An answer came to your mind and it was yes. You picked the best answer based on the very limited information available, but you jumped the gun. What if the next two adjectives were corrupt and cruel? Take note of what you did not do as you briefly thought of Mindik as a leader. You did not start by asking "What would I need to know before I formed an opinion about the quality of someone's leadership?"
·  Substitution: System 1 is prone to substituting a simple question for a more difficult one e.g. in answering the complicated question to which you don't know the answer e.g. 'How far will this woman go in politics' your System 1 will substitute an easier question to which you do know the answer e.g. 'does this woman look like a winner?'

Application to the 'Glass Ceiling in a nice firm': Anyone judging anyone is doing so on incomplete evidence. Managing the evidence being used in assessing me is even harder in the day-do-day, informal judgements which occur without anyone being conscious of them. If What People See about me is All There Is, I need to make sure I am projecting the good /useful stuff alongside the stuff I project because I can't help it (e.g. being a woman of childbearing age, being quieter or less overtly aggressive than male colleagues etc.).

When it comes to answering the complicated question to he does not know the answer e.g. 'Will she be the right person for the job', boss-judge will substitute easier questions e.g. 'Do I like her?, 'Does she appear more ambitious than the average woman?' (stereotype unambitious) or 'Can I think of any reason not to hire her?'.

Even harder to control are substitute questions like 'Is she similar to her potential new peer group?' or 'Does she remind of me of someone already succeeding?' (See the Availability heuristic above).

4. Causality and 'The illusion of Understanding': We constantly make sense of the world, attributing reasons (made up or real) and stories as background 'Her husband left her, I guess he finds her annoying, I certainly do' is an example of this.

We LOVE 'good' stories i.e. simple, coherent one which make sense. Even if we are told that the story we are hearing is from a bad source (e.g. someone who openly admits hating the person they are talking about) we cannot help using the information. What You See Is All There Is.

We also massively overestimate talent and skill and massively underestimate luck (good and bad) because our brains simply cannot deal with that level of uncertainty in our lives. 'Experts' in a subject are far from immune.

Application to the 'Glass Ceiling in a nice firm': Basically anyone 'nice' who decides against me because of an imperfect System 1 is unlikely to think he did so out of any sort of discrimination /stereotyping. An extremely plausible reason for decisions being made will always be immediately offered by the boss-judge's System 1 and is very likely to be accepted by his System 2. Boss-judge is therefore likely to be genuinely hurt, amazed and bewildered if accused of bias. Ensuring that I communicate my story in a simple to understand, coherent way (even if this means leaving stuff out or simplifying) is likely to influence boss-judge's understanding by ‘making sense’ of me as a person outside a stereotype.  

5. Inferring the general from the specific: The book shows powerfully that whatever we are told about 'general' behaviour each of us quietly exempts ourselves from it. In the book for example, when psychology students were told about a certain study and the outcome, and then told to apply that knowledge in making a judgement about specific individuals, they had learned 'almost nothing' from the teaching. However, when they later learned, to their surprise, of their mistake and understood that the specific person complied with the overall finding, they generalised incredibly well i.e. 'Subjects' unwillingness to deduce the particular from the general was matched only by their willingness to infer the general from the particular'.

Application to the 'Glass Ceiling in a nice firm': If the boss-judge has to hand a specific example of an individual similar to us that has surprised him, he is far more likely to generally believe that this is possible in the future (hopefully with me).

We can make more sense of all this by drawing it together in a very simple formula/model for predicting the relative likelihood of a person getting a particular ‘top job’ as:

    Score = perceived skill x perceived desire x perceived fit x luck
If each factor can be assigned a number between zero and 1 then the best score a person can get is where:

Score = 1  x  1  x  1  x  1  = 1 (i.e. this is a shoe in).

For each term of skill, desire, fit and luck, we should note that the score assigned has to take into account anything which detracts from its value in the eyes of boss-judge. So let’s say:

Perceived skill = true skill   ± a bility to communicate skill

Perceived desire = true desire  ±   other hurdles to be overcome with our ‘desire energy’

Perceived fit = true fit  ±  discrimination / bias

In considering what may prevent the score for each factor (skill, desire and fit) simply equalling the ‘true’ score,  we could argue, based on intuition and the heuristic arguments above, that unacknowledged gender bias can easily mean that to equal a man in similar circumstances, I would need more of everything (or a disproportionate amount of luck).

In the case of desire, for instance, a working mother’s amount of perceived desire will only equal what’s left over after all the effort that has gone into making family arrangements and overcoming spousal or other disapproval.
Of course, gender is not the only thing which can reduce our perceived scores to below their 'true' value, but is likely to be contributing factor in many circumstances on top of other, specific detractions we may face e.g. our age, race or attractiveness.

Although we cannot help stereotypes people hold about us, understanding how boss-judge’s System 1 is working for or against us can help us; for instance it must be true that aiming high in an organisation where a woman to whom we bear some resemblance has recently and perhaps surprisingly made it 'big' would help this process.  Success breeds success and not just because we have a role model to follow.

Targeting boss-judge’s System 1, we can provide what it needs to make sense of the world or, where necessary, shock System 1 enough to awaken and engage System 2 in our favour (obviously assuming that the rational choice is to employ / promote us).

It won’t be easy but we're worth it.