Fig 1.
Methodology steps for developing the survey.
Table 1.
Attributes and levels for the discrete choice experiment and swing-weighting.
Fig 2.
Feedback scores from respondents completing the discrete choice experiment (DCE) and swing-weighting (SW).
White = Ease of understanding the DCE; Light grey = Ease of answering the DCE; Dark grey = Ease of understanding the SW; Black = Ease of answering the SW; Respondents answered on a scale from 1 to 6, 1 being the most difficult and 6 being the easiest; Low health numeracy scored below 9.83 (the mean) on the SNS-3; High health numeracy scored above 9.83 (the mean) on the SNS-3; Health literacy questions are scored 1–5 with the middle question inversed—Low health literacy identified by a score of >3 on any item; High health literacy scored <2 on any item (see Louis et al, 2016 [32]).
Table 2.
Attribute-level estimates for the discrete choice experiment mixed-logit model.
Table 3.
Swing weighting preference weights, calculated through both point allocation and the rank order centroid (ROC) method.
Fig 3.
Proportion of attribute importance relative to sum of all attributes’ importance (DCE and SW calculated through both ROC and PA.
DCE = discrete choice experiment; SW = swing weighting; PA = point allocation; ROC = rank order centroid method.