Millisecond Forums

Hi!

I would like to know if you can help me with the literature's reference(s) regarding the values used in the IAT's Performance Summary (if ever published; relevant portion of the script below)? I couldn't find that information anywhere.

If never published, how have you arrived at these values?  

And also... Are they considered "universal" irrespective of the type of population (general, professional groups...), nationality,... etc.?


Thanks!

Ana


_________________________________________________________________________________________________________


***********************************************************************
Performance summary
***********************************************************************

<trial summary>
/ stimulustimes = [0=summary]
/ validresponse = (" ")
/ recorddata = false
/ ontrialbegin = [values.magnitude = "pequena a nenhuma"]
/ ontrialbegin = [if( abs(expressions.d) > 0.15 ) values.magnitude = "leve"]
/ ontrialbegin = [if( abs(expressions.d) > 0.35 ) values.magnitude = "moderada"]
/ ontrialbegin = [if( abs(expressions.d) >= 0.65 ) values.magnitude = "forte"]
/ ontrialbegin = [if (expressions.d >= 0.0) expressions.preferred = item.targetALabel.1]
/ ontrialbegin = [if (expressions.d < 0.0) expressions.preferred = item.targetBLabel.1]
/ ontrialbegin = [if (expressions.d < 0.0) expressions.notpreferred= item.targetALabel.1]
/ ontrialbegin = [if (expressions.d >= 0.0) expressions.notpreferred= item.targetBLabel.1]
</trial>

<text summary>
/ items = ("A sua pontuação TAI (D) foi de <% expressions.d %>, o que sugere uma <% values.magnitude %> preferência automática por <% expressions.preferred %> em relação a <% expressions.notpreferred %>.~n~n~nCarregue na BARRA DE ESPAÇO para completar esta sessão.") 
/ fontstyle = ("Arial", -32, true, false, false, false, 5, 0)
/ size = (60%, 60%)
/ hjustify = center
 </text>

The classification of D-scores into broad categories ("slight", "moderate", "strong") is a matter of convention. It is based on large-scale studies and intended to be analogous to common effect-size classifications. See e.g. https://implicit.harvard.edu/implicit/demo/background/raceinfo.html and https://implicit.harvard.edu/implicit/demo/background/faqs.html#faq8