Sheriff Courts - remand and bail outcomes: occasional paper

Methodology Annex

SCTS Hearings data for sheriff summary and sheriff solemn courts were gathered and subject to usual initial preparation processes (e.g. data cleaning, converting to formats required for analysis).

Data cleaning

Some entries were identified where the appearance date predated the case registration date, and these were removed from the analysis. Hearings where there was no identifier for the accused recorded (<1% of all hearings) or where there was no court type recorded (e.g. Custody Court) (<1% of all hearings) were removed from the analysis.

The analysis was restricted to cases heard on summary complaint or solemn petition (i.e., not solemn indictment) and then to “New” or “After Warrant” diet types, as in the vast majority (97%) of summary complaint or solemn petition cases the first diet on the case is a New or After Warrant diet. Appearances after an initiating warrant are excluded.

The analysis is concerned with the first bail decision points of summary complaints or solemn petitions in sheriff courts and the outcomes for the accused in terms of their continuation status (e.g. remand/bail). The first bail decision point was simply identified as the court record with the earliest diet date for a particular case-accused (e.g. case number 50, accused number 1).

There are occurrences in the data in which the same accused in the same case has two diets on the same date. We have excluded these occurrences from the analysis, rather than trying to identify which is the first. Such occurrences make up less than one in 2,000 first bail decision points.

In all, analysis has been restricted to:

• the first meaningful calling of case-accused,
• on a New or After Warrant diet,
• in sheriff court,
• on complaint or petition

After these restrictions, the data set contained a relatively high proportion of hearings where no Continuation Status was recorded (15%). Without further processing this would be an unacceptably high level for meaningful analysis. It was observed that in many of these instances the next hearing type for the accused on that case was an “After Warrant” diet (which could mean that the accused had failed to appear at the original diet and a warrant issued for their arrest). In these cases, the Continuation Status was recorded as “Warrant”. For other instances where no Continuation Status was recorded but a) the hearing was the final hearing for that case-accused, b) the hearing date matched the recorded verdict date, and c) the case was no longer recorded as “active” in the database, this was recoded as “Final Diet” for that case-accused. After this process, the proportion of hearings with no recorded Continuation Status was reduced to an acceptable level for analysis (4%). Finally, these cases with no Continuation Status were grouped with the “Other” continuation status outcome into a new “Other / NA” grouping.

The dataset contains data on information about all cases registered from 2015-16. To ensure no issues with lags between case registration date and the hearing date, hearings from financial year 2015-16 were excluded and results reported for hearings from 2016-17 through to the latest full financial year 2020-21. (Financial year here taken as 1st April to 31st March).

Aggregation

The accused “Continuation Status” at the end of each hearing (e.g. Remand, Bail) were grouped into high level groupings (Remand, Bail, Ordained, Other/NA). The “Other/NA” group includes non-prison remand (e.g. Child Remand, Hospital Remands) as well as Reported cases. These groupings are shown in table 1.

The appearance type of the accused was taken from the “Court type” data, with the court types grouped into high level groupings (Custody court, Undertakings court, Initiating Warrant, Other), as shown in table 2.

Analysis

Data on counts and frequencies (or “likelihoods”) of particular continuation status outcome were then analysed by financial year, by prosecution forum (e.g. summary or solemn) and by appearance court type (e.g. Custody court or Undertakings court). Lastly, for solemn petition appearances the counts and likelihoods of outcomes were assessed by financial year and whether the accused had a bail aggravator on their case.