|Home||Background||Business Case||Barriers||IAT Explained|
The IAT is a computerised method for indirectly measuring the strength of the association via a double-categorisation task. It relies on the assumption that people are able to categorise strongly associated concepts more quickly than concepts that are weakly associated (Greenwald et al., 1998). The method for the IAT involves the categorisation of stimuli presented as words or pictures, typically across a series of trials. In practice the IAT effect is usually interpreted as a measure of attitude or preference.
As explained earlier, a gender and leadership IAT is being developed to measure implicit attitudes. Based on the research papers of Greenwald et al. (2009), it was decided to use ‘picture’ stimuli in the target category man versus woman. The pictures were selected with care: the different age categories are equally represented, all men and women are ‘Caucasian’ and have a similar appearance and pose. Deciding on ‘word’ stimuli items for the attribute category (Leader versus Follower), was more complicated. As shown in literature review, there are numerous leadership theories and there is no agreement about the definition of leadership. Also, the organisation Philips is urged, not only, to adopt a leadership theory that is in line with contemporary research but also fits the present organisational practices and culture. Furthermore, according to the article about conceptual considerations of the IAT (Nosek et al., 2007) an attribute stimuli need to be understandable, free of bias, short and unambiguous.
Extensive analysis showed that the transactional, transformational and laissez-faire leadership styles, discussed earlier in the literature review, would be the best theoretical framework to adopt. It was decided to use the categorisation as proposed by Bass (1990), who divided these 3 leadership styles into 9 categories. The same categories are frequently cited by Eagly and Johanessen-Schmidt (2001) in their studies to female leadership. Furthermore, a practical linkage is made by the consulting firm McKinsey & Company (2009), who uses these leadership categorisations in their studies to corporate performance. In their research ‘Women Matter’ (McKinsy & Company, 2008) these nine leadership categorisations are used to analyse the impact of women on corporate performance and are linked to their model of organisational excellence. To meet the IAT-design criteria, the nine categorisations were slightly reframed (see appendix). The words that represent the attribute category leadership are: ethical leader, stimulate growth, inspirer, motivator, mentor, drive performance, thought leader, decisive and chairman. The words that represent the attribute category follower are: follower, helper, assistant, supporter, co-worker, subordinate.
Participants are presented with a target category (man versus woman) that is paired with an attribute category (leader versus follower). The IAT consists out of 7 trials, also referred to as blocks. In the first trial, the participant is presented with a ‘picture target’ stimuli on the computer screen and is instructed to sort it into one of two possible target categories (man or woman). This is done by pressing the designated response key, the ‘E’ for left and the ‘I’ for right, as fast as possible. The second trial is also a form of a two-category task, this time the participant is asked to classify a ‘word attribute’ stimuli into the correct category (leadership or follower). In the third and fourth trial the categories of the former two tasks are combined; they can be congruent (man & leadership versus woman & follower) or incongruent (woman & leadership versus man & follower). The choice for congruent and incongruent is based on the theoretical framework presented in the literature review (‘think leader, think male’). In the fifth trial, the participants practice a reverse version of the response assignment for the two target categories. Finally, the fifth and sixth trails combines the reversed target categories from step five with the unchanged attribute categories used in step three and four. An illustration of this procedure projected in Table 1. Only the latency times of block 3, 4, 6 and 7 are used for data analysis; the other blocks are practice trials. The difference in response latencies between the two concepts for the congruent and incongruent pairing provides the basis for the IAT measure (Greenwald, 2003).
Table 1 Sequence of trial blocks in the gender and leadership IAT.
Note. For half the subjects, the positions of Blocks 1, 3, and 4 are switched with those of Blocks 5, 6, and 7, to test for order effects in the IAT design.