>> "We tested this hypothesis through a survey comprised of two-paired alternate forced choice questions to identify which image of a cyclist respondents consid

>> "We tested this hypothesis through a survey comprised of two-paired alternate forced choice questions to identify which image of a cyclist respondents consider to be less human," the study's abstract explains. Doesn’t structuring a study like this ensure the outcome / headline? An image of a cyclist was going to come out as “less human” even if none of the respondents would agree with that statement. It’s like if I showed people two images of a politician and asked them “which individual most belongs in a zoo” and then concluded bald politicians are most likely to be perceived as “belonging in a zoo”. Would any of the respondents make or agree with that statement? It seems unlikely based on the methodology of the study.