Prison Statistics Scotland, 1998

14. Annex : Notes on statistics used in the bulletin

Note : In producing the detailed breakdowns by age and gender for this bulletin, a number of figures in the historical receptions series have been revised. The revisions are :

Table 14 ‘Relationship between reception and existing custodial sentence for sentenced receptions’, 1997-1998 — Figures of ‘No existing custodial sentence’ (9,538) and ‘Concurrent or part concurrent with existing sentence’ (1,552) were previously published. These have been revised to ‘No existing custodial sentence’ (9,537) and ‘Concurrent or part concurrent with existing sentence’ (1,553).

Tables 24 to 24A — ‘Adult direct receptions (of males) to penal establishments’. Figures of ‘Vehicle defect offences’ (4) and ‘other’ (0) were previously published. These have been revised to ‘Vehicle defect offences’ (0) and ‘other’ (4).

Table 29 ‘Young offender fine default receptions to penal establishments by length of sentence’, 1989-1998.

(This table was previously published as table number 30). The average sentence imposed (days) for 1992 was previously published as 24. This has been revised to 22.

Table 30 ‘Young offender fine default receptions to penal establishments by fine outstanding’, 1989-1998. (This table was previously published as table number 33). The receptions for 1996 relating to categories ‘£2,500-less than £5,000’ and ‘£5,000 or more’ were previously published as 3 and 0 respectively. These have been revised to 1 and 2 accordingly.

Table 32 ‘Adult fine default receptions to penal establishments by length of sentence’, 1989-1998. (This table was previously published as table number 31). The receptions for 1996 relating to categories ’30 days/1 month’ and ‘31-59 days’ were previously published as 29 and 67 respectively. These have been revised to 10 and 86 accordingly.

Coverage and data collection

1. The statistics in this bulletin cover all penal establishments in Scotland and also include Legalised Police Cells.

2. Prior to 1 April 1996 population and reception information was collected from the penal establishments via manual statistical returns submitted on a monthly basis. With effect from 1 April 1996 the source of all the statistical information provided by the penal establishments changed (with the exception of Legalised Police Cells). The new computerised information system, the Prisoner Records System (PRS), introduced by the Scottish Prison Service, enables on-line recording of information in respect of each prisoner. It is an operational system but also provides all the statistical information previously collected manually from Scottish Prison Service establishments. The change in data collection from manual submission to electronic receipt of data from Prisoner Records System is partly responsible for some of the changes evident in the population and reception time series between 1995 and 1996. There are particular points to note:

Population

2.1 Historically, a number of variations in recording practices have occurred between establishments and over time. It has been found, for example, that establishments have differed in their method of recording population in cases where prisoners have multiple warrants; in particular those instances in which existing prisoners receive subsequent sentences. The recording of the overall sentence length category for each prisoner under the manual system was an onerous task and the arrival of further warrants for prisoners already serving custodial sentences had to be carefully recorded and the impact of these warrants reflected in the prisoner’s overall sentence category. For example, a second warrant of 3 months duration arrives for a prisoner already serving a 6 month sentence. It is to be served consecutively to the existing sentence. From the date of the arrival of the second warrant the prisoner’s overall sentence category should be recorded as 9 months. It is possible that such changes were not always accurately recorded under the manual system; hence the changes evident in some population categories between 1995 and 1996.

Receptions

2.2 During 1995 establishments were managing two systems: continuation of the submission of the manual statistical returns and the implementation of the Prisoner Records System. This undoubtedly resulted in some under-recording of receptions under the manual system during 1995 as effort was concentrated on the new Prisoner Records System roll-out. Following publication of the 1995 Prison Statistics bulletin, court proceedings data for 1995 became available and was published in March 1997. An investigation of the 1995 prisons data was carried out using the court proceedings data for comparison purposes. This provided an indication of the extent of possible under-recording of receptions in 1995. The table below has been updated to reflect the most recently available data and details the direct sentenced prison receptions recorded in each of the years 1990 - 1998 compared with court proceedings-derived ‘reception’ figures. The court-derived ‘reception’ figures do not take account of the backdating issue referred to in Annex Note 5; hence the reason why, in most years, the proportion of prison : court ‘receptions’ is less than 100 per cent.

Comparison of direct sentenced prison receptions to court derived ‘receptions’,

1990 — 1998, Scotland

The figure of 89 per cent for 1995 is low when viewed in the context of the figures for the period 1990 - 1998. If, for example, the figure of 94 per cent (the 1994 prison : court proportion) is assumed for 1995 then the estimated number of direct sentenced receptions in 1995 would be 12,200. Thus, an estimate of the actual increase in the number of direct sentenced receptions between 1995 and 1996 is likely to be of the order of 8 per cent rather than the 14 per cent which the raw data suggests.

If the same level of under-recording is assumed for the fine default receptions in 1995 then the estimated number for 1995 would be 8,000. Thus, the increase in the number of fine default receptions between 1995 and 1996 is likely to be of the order of 12 per cent rather than the 20 per cent which the raw data suggests.

The 1997 and 1998 figures for prison receptions : court-derived ‘receptions’ (98 per cent and 100 per cent respectively) look reasonable in the context of the time period 1990 - 1998. It is likely that the changes to the data collection procedures during 1996 resulted in some increase in the number of receptions recorded; perhaps 1 — 2 per cent.

2.3 The Prisoner Records System (PRS) has undoubtedly resulted in improved recording of the details of receptions (both direct and fine default). For example, automatic generation of the overall sentence length for each reception should reduce the errors previously incurred in this calculation. It has also resulted in some changes in the main crime/offence classification since this is now system generated rather than produced manually. The main crime/offence reflects the crime for which the longest sentence was awarded by the court. There are inevitably going to be differences when this code is generated automatically as opposed to manually. In addition, it has come to the attention of the statistics unit that historically some establishments incorrectly recorded the main crime/offence. For example, crimes of ‘handling an offensive weapon’ were recorded in the category ‘breach of the peace’ rather than the specific category for such crimes and there were historical differences in distinguishing between serious and petty assault. It is also believed that the historical variations in recording practices led to an undercount of long term sentences for under 21’s in 1988 and 1989.

2.4 Some reception information previously collected on the monthly returns is no longer available via PRS. For example, prior to 1995 prisoners provided information at the point of reception as to whether they had previously served a custodial sentence at some time in the past. This information is no longer asked of prisoners. However, such information will be available from PRS in future but the record of previous periods of imprisonment held within PRS which will be used to provide this information will take a number of years to mature.

Missing data

2.5 Some fine default information regarding ‘amount of fine outstanding’ was unavailable at the time of publication. The amount of fine outstanding for 65 receptions during 1998 has been estimated using an average of the appropriate categories. Where a reception category displayed missing or singular values for the fine amount outstanding, an average was taken using the average fine outstanding of the nearest appropriate category.

2.6 During 1998 there were at least 20 weekdays for which information was unavailable. Where this occurred the data was imputed by copying the previous day’s information and applying it to the missing day to generate the population figures. Any warrant information entered on these days could be lost. This would mean that receptions of prisoners serving only a day or two such as fine defaulters, may not be counted. Because of this, we estimate that the figure of 8,400 fine default receptions is under counted by at least 250 receptions. However, similar problems would have occurred in 1997, therefore comparisons between 1997 and 1998 remain unaffected.

Legislative changes

3. There have been several legislative changes which have impacted on prison population and receptions. These are :

Population

3.1 On 1 October 1993 the Prisoners and Criminal Proceedings (Scotland) Act 1993 (‘the 1993 Act’) came into force. This had two main implications in relation to prisoner numbers. The 1993 Act changed the early release rules for prisoners serving less than four years. Instead of being released automatically after serving two-thirds of their sentence, prisoners are now released after serving one-half of their sentence (effective in respect of those sentenced on or after 1 October 1993). In addition, for those prisoners serving sentences greater than four years, release on parole may be possible after one-half sentence rather than one-third. Secondly, the transitional arrangements for early release set out in the 1993 Act resulted in a decrease of 600 in the number of short-term prisoners on 1 October 1993. The population decreased immediately on 1 October 1993 from 5,800 to 5,200. Direct comparisons of the figures for 1993 and subsequent years with that for any previous year must therefore be treated with some caution.

Receptions

3.2 In 1992 a legislative change was made to the table of tariffs which revised the relationships between the amount of outstanding fine and corresponding prison sentence imposed for fine default (i.e. tariff), effective from 1 October 1992, e.g. an outstanding fine of £500 would previously have attracted a period of imprisonment of 60 days: the change in tariffs reduced this to 14 days.

Counting conventions

4. ‘Receptions’ are not equivalent to ‘persons received’. Where a person has several sentences imposed on him by one court in one day this is counted as one reception. However, where custodial sentences are imposed on the same person by 2 or more courts in one day, 2 or more receptions are counted. Where a person is reconvicted while serving a sentence and a further custodial sentence is imposed on him this too is regarded as a further reception.

5. Direct sentenced prison receptions in this bulletin (i.e. excluding fine default receptions) are counted differently from custodial court disposals published in The Scottish Executive Justice Department annual statistical bulletin ‘Criminal Proceedings in Scottish Courts’. There are 2 main reasons for this. Firstly, in the case of backdated sentences if, after backdating, it is found that the custodial sentence has expired, neither the warrant nor the person sentenced will be taken to prison thus a reception will not be counted though the sentence will be included in the court statistics. Secondly, if a person is given one or more custodial sentences on each of 2 separate sets of charges from the same court on the same day, this will be counted as 2 custodial sentences in the court statistics but only one direct sentence reception. The table below illustrates both direct sentence prison receptions and custodial court disposals.

6. Only the main crime or offence is recorded for persons convicted of more than one criminal act. The main crime or offence is the most serious, assumed to be that for which the longest period in custody was imposed. Where a person is to serve consecutive sentences, the length of the sentence given is the sum of the sentences; for concurrent sentences, the period of the longest sentence is given. These sentences represent the period which would be served if no remission or parole were granted to the prisoner.

7. A small number of persons aged 21 or over are received into young offenders institutions. In addition, prisoners up to the age of 23 may be kept at young offenders institutions.

8. With effect from 1996, receptions for compensation order default are included in the fine default figures as these receptions are not separately identifiable from the data received from the Prisoner Records System (PRS).

Historical recording practices

9. The number of persons received on remand (Tables 15 to 15B) was collected annually up to 1988 and monthly until March 1996. With the exception of children on unruly certificates (see Note 10 below) only a breakdown by age group (under 21 and over 21), gender and type of reception is available until March 1996. No further details e.g. details of the alleged crime, are available historically.

10. Prior to 1 April 1996, a return was made to the statistics unit for each child received into the penal system prior to trial or sentence on an ‘unruly certificate’. The return provided details of the age and gender of the child, date of receptions, date of release and alleged crime(s). An audit of the unruly information resulted in the introduction of stricter procedures in 1994 to ensure that a return was submitted for every child received into the penal system on an unruly certificate. Direct comparisons of the figures for 1994 and later years with that for any earlier year must therefore be treated with caution as there may be some undercounting in earlier years. With effect from 1 April 1996 unruly certificate information is obtained directly from the Prisoner Records System (PRS). Tables 16 to 18 provide details of children received on unruly certificates.

Classification

11. The classification of crimes and offences used by The Scottish Executive Justice Department for criminal statistics contains approximately 350 codes. The detailed offence information contained on the warrants on the Prisoner Records System is mapped to the relevant SEJD classification code. The groupings used in the bulletin are :

Symbols

12. The following symbols are used throughout the tables in this bulletin.

Contact

13. Further analyses of prison statistics can be supplied on request. In certain cases a fee will be charged. For details of what can be provided please contact Neil White on (0131) 244 2147 or e-mail your queries to neil.white@scotland.gov.uk or neil.white@sps.gov.uk