Categorization into five groups of patients (details in Additional file 2) and into six groups of peaks (details in Additional file 3) shows that, for the first and the second groups of patients, high values of the fifth and sixth groups of peaks are characteristic, while the first group reveals also higher values in the first cluster of peaks.
The reclassification table presents the categorization into risk groups according to the initial and updated risk models.
Categorization into exposure groups was done on all samples; however, size/group varied somewhat in the regression analyses because of missing values for other covariates.
Categorization into diagnostic groups was made on a clinical basis at consensus conferences including neurologists, neuropsychologists, a neuropsychiatrist and study coordinators.
A decision was taken to reduce this categorization into two groups: likely (very likely, quite likely) and unlikely/uncertain (unlikely, very unlikely or unable to say).
Qualitative variables like sex, Apgar score (categorization into two groups – <6 at 5 minutes and ≥ 6 at 5 minutes), maternal and postnatal problems (again categorized-present or absent) were compared among the three groups by Fisher's exact test.
However, if one uses semantic similarity measures for the categorization of genes into groups of interest, one has to consider that categorization is not about predicting how close two terms are but assessing how well two terms go together.
Social categorization is the process of mentally categorizing individuals into groups based on characteristics such as age, sex, and ethnicity.
Table 3 shows the number of confirmed term pairs after the categorization stage divided into groups by the number of words in the terms.
The categorization of units into groups representing units with either a more or less established conservative management pathway was relatively crude and is a key limitation of the study.
Because data were derived from 3 hospitals that undertook reviews for different purposes, the slight differences in inclusion criteria and death categorizations necessitated subanalyses for the death within 30 days comparison and the collapsing of the categorizations into 3 groups.
