Ethnicity was also directly self-reported by study participants, rather than inferred from other information, and we have been explicit in our categorisation of ethnic groups.
In contrast, patients with locally advanced T4 tumours have a much poorer outcome and some histological studies have also identified extramural invasion greater than 5 mm (T3c/d) as a poor prognostic feature (Merkel et al, 2001) and we have recognised this in our categorisation of CT-predicted T-stage.
To achieve categorisation in our study, three diabetes specialists independently classified CGM profiles into groups of very good, good, satisfactory, fair, and poor metabolic control.
In our previous work on the categorisation of Supreme Court cases [18], we identified several reasons for this.
The assessments utilised in our survey involved the systematic categorisation of cognitive deficits into five subgroups.
In our search for the best categorisation of PMN-E, we have used isotonic regression analysis (Barlow et al, 1972; Foekens et al, 1994b) using the rate of overall response to tamoxifen treatment as end point.
In our search for the best categorisation of hK6 and hK10, we have used isotonic regression analysis (Barlow et al, 1972; Foekens et al, 1994b) using the overall response rate as endpoint in those patients who received tamoxifen therapy for advanced disease.
In a sense, our current categorisation of reward size already combines rewards that are up to and including the average reward size across the three reward steps under the "small" reward condition, and rewards that are larger than the average in the "large" reward condition (in Figure 2 and also in Methods, in the subsection "Reward Schedule").
This classification can serve as a reference, both in our study and in further epidemiological research, for categorisation of surgical procedures in terms of risk level of Creutzfeldt-Jakob Disease acquisition.
Poor categorisation of age groups in our study might have been responsible for the inability to detect any significant relationship.
In our data, the effect elicited by the items appeared to be more negative in the Categorisation groups than in either of the Recognition groups; more negative in the un-exposed Control groups than in the Pre-exposed groups and more negative for List 1 items than for List 2 items.
