The experimental measurement of social preferences has led to somewhat equivocal results. The experimental Equality Equivalence Test proposed by Kerschbamer (2015) is a promising, simple, model-free and comprehensive tool for eliciting distributional social preferences. We here assess the validity of this method by modifying it so that we can test its key assumption: that the strength of the concern for the inactive player depends only on whether her payoff is above or below that of the decision-maker. In general, we find that this assumption holds. Moreover, the prevalence of types of social preferences that we observe is similar to that in the original paper, with selfish and quasi-maximin (Charness and Rabin 2002) being the most common.