Table 2.

Provisional scaled item response values

Response Level	Q1	Q2	Q3	Q4	Q5	Q6	Q7	Q8	Q9	Q10	Q11	Q12
0	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000
1	0.582	0.021	0.462	0.000	0.000	0.449	0.000	0.704	0.000	0.422	0.733	0.698
2	1.493	0.021	0.880	0.745	0.653	1.091	0.000	0.704	1.423	1.072	1.099	0.852
3	1.493	0.206	0.880	0.745	1.187	1.268	1.009	1.492	1.423	1.727	1.632	1.695
4	2.079	1.143	2.199	1.421	1.720	1.509	1.446	1.657	1.423	1.727	1.710	2.141
5	2.354	1.356	2.333	2.324	1.890	2.302	1.206	1.749	1.423	2.110	1.926	2.789

Short Item Key: 1 Q1 - Draining ear; Q2 - Smelly ear; Q3 - Hearing at home; Q4 - Hearing in noise; Q5 - Discomfort/pain; Q6 – Dizziness; Q7 – Tinnitus; Q8 - Activity restriction; Q9 – Need to limit exposure to water; Q10 - GP visits; Q11 - Taking medicines; Q12 - Overall QoL impact of hearing problems

Entries are the adopted item response values for the 5 levels for each of the 12 questions when regressing them against a preliminary total score. Here that dependent variable was the total 1st PC total (out of 5 X 12 = 60, unscaled, but scored continuously as integers), for the average of 1st two visits so in effect a grain of 1/120 in the dependent variable. The exemplified independent variable category means estimated are the set from visit. The 12 scaling regressions shown in Table 2 are done individually for the chosen dataset eg Visit 1, 2 combined), taking the responses of 0–5 as category labels for their means in the raw total to be estimated from data by regression mapping, not as integer arithmetical values