For those using association, for example, one must consider that association and response latency procedures tell us little about whether stereotype activation is the result of an automatic or controlled process. Simply because a participant responds quickly to an attitudinal enquiry, this is not necessarily proof of a stereotype’s automatic activation. It may also have been retrieved quickly through the use of an efficient but controlled process. Therefore, this undermines the validity of association-based explicit measures by suggesting that speed of association may not effectively illustrate the automaticity of …show more content…
Namely, correlations between explicit and implicit measures are positive, but extremely low (Greenwald, McGhee and Schwartz, 1998; Hofmann et al, 2003), suggesting that, worryingly, they may not measure the exact same concept. Nonetheless, one explanation for this is that they predict contrasting types of overt behaviour as part of stereotypical attitudes (Dovidio et al, 2002). This paper proposed two forms of stereotypes, aptly named and differentiated by their “Implicit” and “Explicit” qualities. Explicit behaviours encapsulate verbal statements affected by stereotypes, which are relatively controllable, whereas implicit behaviours are less controllable, and include non-verbal friendliness towards other social groups. Interestingly, each type of stereotype-rooted behaviour is more highly correlated with its respective measurement type (implicit behaviours are more effectively measured by implicit techniques, and vice-versa). Although this somewhat explains the lack of strong positive correlation between explicit and implicit measures overall, it also presents implications for the applicability of both types. If implicit and explicit procedures measure different concepts, this limits their overall applicability to stereotypes, showing that it is more specific than originally