While Class-1 prefers innovations targeting adults the most, followed by young people relative to elderly, Class-2 and Class-3 hold different preferences: Class-2 does not prefer innovations targeting elderly and is indifferent between other age groups, Class-3 prefers spending on elderly, but not on young people.
Three scale class solutions were unstable for some starting value seeds (and were considered to be unlikely given the number of modes in Figure 1), as were two-scale three- and four-preference class solutions; hence, a two-scale two-preference class solution was preferred.
princĭpātus, ūs, m. [princeps], the first place, pre-eminence, preference (class.).
Respondents' characteristics that affect the probability of belonging to a latent preference class are environmental behaviour and attitude, actual living environment, age class, education level, and organisation membership.
These naturally account for preference class membership and 'net out' differences in scale.
Of these, only health status and education level were found to be associated with preference class in secondary latent class analyses (see online supplementary material).
Multivariable logistic regression produced no associations between any of the sociodemographic variables and preference class at the 5% level, and only income was associated with scale class (results not shown).
Notably, respondents who were not overweight or obese, who had no experience of weight loss surgery, or with better overall health were more likely to belong to a preference class for whom lifestyle commitment was considered to be particularly important.
While respondent age was not observed to affect priority choices, BMI was perhaps unsurprisingly associated with preference class membership, reaffirming the need to give careful consideration to whose preferences are sought to inform priority decisions the public or individuals with some direct or indirect experience of the condition.
The estimated SALC models used the five sums of squares variables (one for each attribute) from the cluster analyses in Appendix 2 (Supporting Information) as predictors of preference class membership and the Empirical Scale Parameter as a predictor of scale class.
Finally, the heterogeneity-adjusted population level tariff was calculated in the following way: The average values across all respondents were calculated, by taking the mean of the sets of preference class estimates, weighted by the average mixing coefficients across the sample.
