Soft laddering was conducted by two trained interviewers (authors AB and IF). Participants in the soft laddering group were presented with the same a priori concrete attributes list as presented to the hard laddering participants (four different randomisations, which were allocated in block fashion to mothers). In the same manner as the hard laddering formats, mothers were first asked to choose between 1 and 3 items from the attribute list which best described what they considered to be the most important attribute if they were to choosea breakfast for their child. The interviewer went through each attribute chosen, asking the mother ‘‘and why is that important for your child?’’. The answer to this question was followed by further similar probes. Where more than one reason was nominated as to why an attribute or consequence was important, each reason was probed by the interviewer, thus allowing forking’ of answers. The procedure continued until the respondent could provide no further answers. The same procedure was repeated for all chosen attributes. le.