The article explores different ways to analyze data from clinical trials where subjective outcomes are measured. Researchers compared different models to the common one used and found that the Exemplary Dataset method accurately estimated power for these models, even with small sample sizes. They also found that the proportional odds test statistic performed well, even when the assumption of proportional odds was not met.