Skip to main content

Table 5 Partial log-likelihood (revealed choices)

From: Do patients’ preferences prevail in hospital selection?: a comparison between discrete choice experiments and revealed hospital choice

Attributes excluded

LL

Difference in LL (relative to full model)

% of total difference

Cumulative %

Relative order

Breast cancer

 None (full model)

− 613.866

    

 Travel distance

−855.918

− 242.052

85.5

85.5

1

 Clinical outcome indicator (tumor-positive resection margin)

− 632.560

−18.694

6.6

92.1

2

 Patient experiences

− 625.541

−11.675

4.1

96.2

3

 Waiting time

− 624.841

−10.975

3.8

100.0

4

Cataract

 None (full model)

− 1145.788

    

 Travel distance

− 1501.405

− 355.617

98.5

98.5

1

 Patient experiences

− 1149.633

−3.846

1.1

99.6

2

 Waiting time

− 1146.522

−0.735

0.2

99.8

3

 Clinical outcome indicator (per-operatively performed vitrectomy)

−1146.474

−0.687

0.2

100.0

4

  1. As suggested by Lancsar et al. [29], we used weights to ensure the weighted sample size was equal across all blocks of the experimental design
  2. LL Log-Likelihood