Bias-Trainings: What can they really do?

By our guest author Nina Locher, student in the HSG Master Program "Management, Organization and Culture".
December 20, 2019

"Think manager, think male" is still very common
The Swiss top management is made up of 80% men and women are promoted much less frequently (Advance & HSG, 2018). Pay gaps are also still common. The higher the occupational position, the lower the proportion of women and the greater the pay gap between men and women (Federal Statistical Office, 2018). No wonder that stereotypical notions of "good managers" are still widespread. While women are regarded as kind-hearted, caring and sensitive, men are characterized by performance orientation, willingness to take responsibility and rationality (Heilman, 2012). Stereotypical, yes, but quite effective, as we associate this with manager’s characteristics that are more likely to be assigned to men: "think manager think male". Our idea of the good manager is dominated by stereotypical masculine characteristics. Women in management positions find themselves in a dilemma: If they show stereotypical feminine qualities , they are quickly perceived as less competent for a management position. However, if they show the qualities required for a manager, such as assertiveness or ambition, they are perceived as less feminine, which in turn results in a more negative assessment(Eagly & Karau, 2002). No matter what they do, they cannot win, and the prevailing stereotypes prevent them from advancing their careers.

Bias trainings and what they (can) change
More and more companies around the world have recognized this problem and taken action. “Bias training" is currently a widespread form of intervention. Its aim is to free employees from their conscious and unconscious prejudices and thus ensure equal opportunities in career advancement. 
But how do these trainings work and what can we expect from them? In my master thesis I investigated the question of the effectiveness of bias trainings. After initial research into the current state of research, it became clear that effectiveness can only be assessed if two fundamentally different training objectives are distinguished: On the one hand, the change of attitudes, which includes stereotypical ideas of women in management positions as well as their evaluation, on the other hand the change of behaviour, which can then be expressed in decisions made differently. Here it becomes immediately clear that changing individual attitudes alone will not make a decisive difference in decisions for or against the advancement of women as leaders (Noon, 2017). The people who decide on the staffing of top positions must also change their behaviour if something is to change. In my master thesis I therefore differentiated between the following two questions: 1. "Are bias trainings resulting in attitude change?" and 2. "Can bias trainings effect changes in behaviour?".

What does the current state of research tell us?
Fundamentally, the state of research shows that bias trainings are very able to create awareness of diversity and discrimination (Carnes et al., 2015; Majumdar et al., 2004; Moss-Racusin et al., 2016), they contribute to an increase of knowledge (Bezrukova et al., 2016) and, at least in the short term, to change implicit attitudes (Jackson, Hillard & Schneider, 2014; Girod et al., 2016). The implicit attitudes are those that are not explicitly addressed in the situation in question. In contrast to the explicit attitudes, they represent a particularly large research interest. The work of Jackson, Hillard and Schneider (2014), for example, is interesting in this respect. The researchers examined the influence of bias training on attitudes towards women working in typical male disciplines at various universities. In their methodologically exemplary study, they came to the conclusion that bias training is perfectly capable of changing implicit attitudes at short notice. Similarly, Girod, Fassiotto, Grewal, Ku, Sriram, Nosek and Valantine (2016) examined prejudices against women in leadership positions. Their study also shows that bias training can influence implicit attitudes in the short term. The extent to which bias training can be expected to have long-term effects, however, is still unclear.

While the effect of bias training on short-term change in implicit attitudes seems to be clearly demonstrated (Lai et al., 2016), this is less clear for the effects with regards to behaviour. However, this is due in particular to the very poor state of research. Hardly any study exclusively investigates this relationship. Often, certain behavioural adaptations are measured only incidentally in a study that actually focuses on the analysis of attitude changes. For example, the study by Forscher, Mitamura, Dix, Cox and Devine (2017). This study shows that people who have undergone bias training are more willing to speak out publicly against a discriminatory contribution than they would have done without the training. The findings of my master thesis clearly show that further research would be necessary for providing evidence about long-term effects and behavioural adjustments resulting from bias training. Furthermore, as far as I know, no study so far has particularly investigated the effect of bias training in the very context of recruitment and promotion decisions.

If bias trainings, then do it right!
Despite these important research gaps  it can be said that bias training has positive effects. For the practical implementation of these trainings it is of course particularly interesting, which concrete training elements contribute to a training success in a company. Many studies list such beneficial elements for bias interventions. Central to this are all elements that go beyond the change of individual attitudes and address action competences - i.e. behaviour - or are integrated into larger strategic or structural measures such as the management development of a company (Dobbin & Kalev, 2016). In the following I present the most important findings based on my analysis: 

A longer duration of the training can have a positive effect on its effectiveness (e.g. Bezrukova et al., 2016).

• Training should also include the learning of specific action competences and the thematizing of concrete strategies and should not be limited to raising awareness of discrimination and changing attitudes (e.g. Bezrukova et al. 2016; Roberson et al., 2009). 
• It is recommended to combine different mediation methods (presentations, discussions etc.) and to make the workshops lively and emotional (e.g. Lai et al., 2016).
• The effectiveness of diversity and bias trainings can be supported by additional structural measures, e.g. the integration of the equality goals into the company goals, or the integration of the workshops into everyday work outside the training context (e.g. Bezrukova et al. 2016; Williamson & Foley, 2018).
• In order to facilitate the transfer of training into everyday working life, competence-based, action-oriented learning should be dealt with in depth in the interventions (e.g. Carnes et al. 2015; Girod et al., 2016; Devine et al., 2012).
• Support from superiors and colleagues is important when conducting training. These have an important influence on the application of transfer strategies (Roberson et al., 2009).

With regard to bias training, the studies reviewed provide encouraging results that give us some hope for a future in which equal opportunities for women and men in career advancement becomes a reality. However, in order to clarify the question whether bias trainings are actually able to contribute to the permeability of the well-known glass ceiling, additional long-term studies and studies on behavioural adaptations with validated measurement methods would be needed. These will be able to answer the question of whether people's attitudes can only be influenced in a short term range or if these changes can be sustained over a longer period of time and whether this will actually change behaviour in the decision-making situation.