Problems with understanding and engaging with task from the 2033 respondents completing all five TTO tasks.

*indicates final criteria used to determine respondents failing to engage or understand with the task. Further information available from the author.
Notes:
i. The logical test appears only in the first TTO task where respondents are asked if they would prefer 10 years in full health or 10 years in a health state worse than full health.
ii. Given the experimental design, the five scenarios given to each respondent included a mixture of mild and severe health states. If the respondent considers all five health states to have the same value then they were deemed to have not understood the task.
iii. For health states better than dead, the iteration procedure begins at 5 years (value of 0.5). The further away from 5 years the point of indifference is found, the more choices are required (and therefore more time). If the respondent was not engaged, the quickest way to complete the task is to answer at 5 years. Of course their true preference might be at 5 years, and so the number of consecutive values at 0.5 from the last TTO task are counted, as this might indicate whether they lost interest during the exercise (e.g. if values were 0.7, 0.3, 0.5, 0.5, 0.5 then this would be counted as 3, but 0.5, 0.5, 0.7, 0.8, 0.3 would be counted as zero).
iv. The first question in the task determines if the respondent considers the health state better or worse than dead. A number of health states considered worse than dead was considered to indicate an unengaged respondent.
v. A pairwise logical inconsistency was considered where the state with a less severe problem on a particular dimension, compared to another state, given its problems on the other dimensions are no more severe – e.g., 11121 versus 11131 and 32211 versus 32313 – is given a lower value.