Homicide in Scotland 2017-2018: statistics

Commentary

Homicide cases

Table 1; Charts 1, 2

• The number of homicide cases recorded by the police in Scotland decreased by 5% between 2016-17 and 2017-18, from 62 to 59. This is the joint lowest annual number of homicide cases since 1976, the first year for which comparable data are available. Since 2012-13 the figures have been more stable, ranging between 59 and 63 cases each year. This follows a generally downward trend since 2004-05 (Chart 1).

Chart 1: Cases recorded as homicide by the police, Scotland, 1996-97 to 2017-18

• Homicide cases involving more than one victim are relatively rare, in 2017-18 there were no cases with more than one victim (Table 1). Of the 59 homicide cases recorded, 13 involved more than one accused person.

Table 1: Summary table of homicide cases, victims and accused persons, Scotland, 2008-09 to 2017-18

	2008-09	2009-10	2010-11	2011-12	2012-13	2013-14	2014-15	2015-16	2016-17	2017-18
Total number of cases	97	80	98	91	63	61	62	59	62	59
Cases by number of victims:
1	95	78	97	90	63	60	62	59	59	59
2+	2	2	1	1	-	1	-	-	3	-
Total number of victims	99	82	100	93	63	62	62	59	65	59
Median age of victims	37	41	33	35	38	42	42	37	40	39
Cases by number of accused:
1	67	53	68	66	45	47	49	48	48	45
2	19	20	19	13	11	8	9	9	5	8
3+	8	6	10	10	6	6	4	1	6	5
unsolved	3	1	1	2	1	-	-	1	3	1
Total number of accused	132	121	142	127	85	90	80	69	78	81
Median age of accused	28	27	28	29	30	30	31	29	30	35

The unsolved case in 2017-18 was re-classified as a homicide after the initial data checks were completed and was identified through the enhanced quality assurance procedures that were put in place this year. For more information see the Data Quality Statement.

• In 2017-18, Glasgow City had the highest number of homicide cases (11), representing 19% of the Scottish total (Table 2). Glasgow City has nearly double its share of homicides compared to its population (with 11% of the Scottish population in 2017^[1]). It is important to note however that victims may not necessarily reside in the local authority where the homicide took place. Glasgow City has also witnessed a large fall in homicides of 54% since 2008-09, accounting for over one third (34%) of the overall national decrease.

Chart 2: Location of homicide cases, where known, Scotland, 2008-09 to 2017-18

• In 2017-18, 69% of homicides occurred within a residential location (dwelling and other residential), 22% in outdoor public places (street, footpath and open outdoor area) and 3% in indoor public places (Chart 2). In 2017-18, one case (2%) occurred within a prison.

Victims of homicide

Table 5; Charts 3, 4

• Between 2016-17 and 2017-18, the number of homicide victims in Scotland decreased by 9%, from 65 to 59 (Chart 3).

Chart 3: Total number of victims and victims by gender, Scotland, 1996-97 to 2017-18

• In 2017-18, there were 45 male victims, representing 76% of all homicide victims. Males are more likely to be victims compared to females, with an overall rate for males of 17 victims per million population, more than three times the rate for females (five victims per million population).
• Chart 4 shows the victimisation rate by age group and gender for homicides since 2008-09 (i.e. for the past 10 years – representing 744 victims). The victimisation rate is either the same or higher for males than for females for all age groups. The rate for males peaks in the 21 to 30 age group.
• The median age of a victim of homicide in 2017-18 was 39 years old, similar to 2016-17 when the median age of a victim was 40 years old. Over the past ten years the median age of a victim has ranged between 33 and 42 years old (Table 1).

Chart 4: Age profile of homicide victimisation rate by gender, Scotland, 2008-09 to 2017-18

Persons accused of homicide

Table 6; Charts 5, 6

• The number of people accused of homicide has shown an overall downward trend since 2004-05 (Chart 5). In 2017-18, there were 81 persons accused of homicide, three more than in 2016-17 and the fourth lowest number since 1976, the first year for which comparable data are available.
• The vast majority of persons accused of homicide since 1996-97 have been male, making up 95% of all accused in 2017-18.
• Of the 59 cases of homicide recorded for 2017-18, 58 were solved and one is currently unsolved.

Chart 5: Total number of accused and accused by gender, Scotland, 1996-97 to 2017-18

• In 2017-18, the total number of individuals accused of homicide equated to 15 per million population. This is the joint third lowest rate recorded in the last ten years.
• In 2017-18, the rates for those accused of homicide were highest for the 16 to 20 age group, at 33 per million population.
• Chart 6 shows the rate for accused by age group and gender for homicides since 2008-09 (i.e. for the past 10 years – representing 1,005 accused). The rates for males accused of homicide per million population were considerably higher across all age groups than for females.
• In 2017-18, the median age of a person accused of homicide was 35 years old, which is older than in 2016-17 when the median age was 30 years old. Between 2008-09 and 2016-17, the median age of a person accused of homicide was fairly stable before rising to 35 in 2017-18 (Table 1).

Chart 6: Age and gender profile of persons accused of homicide per million population, Scotland, 2008-09 to 2017-18

Method

Table 7; Chart 7

• Chart 7 shows that the most common main method of killing in each of the last ten years was with a sharp instrument. This includes 58% (34) of homicide victims in 2017-18 of which all but one involved a knife. The next most common main method was hitting and kicking, accounting for 14% (8) of homicide victims in 2017-18.
• Homicide by shooting is relatively rare in Scotland. In 2017-18 there were two victims recorded and since 2008-09 there has only been one year where more than two victims were recorded (2011-12).

Chart 7: Victims of homicide by main method of killing, 2008-09 to 2017-18

Relationship of main accused to victim

Tables 8-10; Charts 8-10

• In the majority of cases, the victim and main accused were known to each other, representing 67% of homicide cases solved in 2017-18. The accused was unknown to the victim in 16 cases (28%). This latter group includes 12 cases where the victim was not known to the accused and four cases where the victim was known to the accused. Of those cases where the victim and accused were known to each other, 67% were acquaintances, 23% were partners or ex-partners and 10% were relatives. The relationship to the main accused was unknown in the remaining 5% of cases.
• Chart 8 and Chart 9 show the percentage of homicides for male and female victims by their relationship to the main accused since 2008-09. Male victims were consistently most likely to be killed by an acquaintance. Female victims are generally more likely to be killed by a partner or ex-partner. However, in recent years as the number of female victims has decreased (14 in 2017-18 compared with 28 in 2008-09) there has been greater year-to-year fluctuation in the proportions of the accused's relationship to the victim.
• Table 9 shows that between 2008-09 and 2017-18, a total of 36 children under the age of 16 years were victims of homicide (where there was an accused person). Of these, 83% were killed by one of their parents. Since 2008-09 there have been 15 victims aged under one year old and in all but two cases (where there was an accused person) the main accused was a parent.

Chart 8: Male victims of homicide by relationship to main accused, where relationship known, Scotland, 2008-09 to 2017-18

Chart 9: Female victims of homicide by relationship to main accused, where relationship known, Scotland, 2008-09 to 2017-18

• Table 9 shows that for all homicides recorded in the last ten years, just under half (49%) of the female victims aged between 16 and 70 years were killed by their partner or ex-partner, 29% were killed by an acquaintance and 8% were killed by a stranger.
• Chart 10 shows that two-thirds of all homicide cases (66%) recorded between 2008-09 and 2017-18 involved males killing males. Cases where the main accused and main victim were both female accounted for just 4% of the total number.

Chart 10: Relationship between main accused and victim by gender, where relationship known, Scotland, 2008-09 to 2017-18

Main motive

Tables 11 - 14

• The most common reasons recorded for committing homicide in the ten year period between 2008-09 and 2017-18, were fight or quarrel, and rage or fury, with half (50%) of all victims in solved cases killed in such circumstances. In 2017-18, this figure was 36% (21 victims), with a much smaller proportion for female (7%) compared with male (45%) victims (Table 11 and 12).
• In the period 2008-09 to 2017-18, 81% of all female victims were killed in dwellings, compared with 55% of males. There is significant variation surrounding the location and motives of homicides over the past ten years. For females, the most common set of circumstances in which they become victims of homicide are in a dwelling, in a rage or fight with a partner or ex-partner, with this accounting for one fifth (20%) of victims. Location is less of a factor for male victims, where killings most commonly occur as a result of a rage or fight with an acquaintance either in a dwelling (19% of male victims) or not in a dwelling (15% of male victims) (Table 12).
• Twenty-eight (48%) victims were reported to have been killed in drug-related homicide cases in 2017-18, of which 25 were male (Table 13). The number of drug-related homicide cases has increased over the past two years, though Police Scotland advise that this may be due at least in part to an improvement in recording practice (i.e. better identification of where motives can include a drug-related element). Given this, Police Scotland will be reviewing cases from preceding years to determine if any of the figures on motives need to be revised. We will update users as to the outcome of this exercise.
• None of the homicides recorded in 2017-18 were reported to have had either a homophobic motivation or a racial motivation. In all homicide cases in the ten years leading up to 2017-18, 2% had a homophobic or racist motivation (Table 14).

Use of alcohol and drugs

Tables 15-18; Chart 11

• Of the 81 persons accused in homicide cases in 2017-18, 30 (37%) were reported to have been under the influence of alcohol, drugs or a combination of both at the time of the homicide. This 37% was broken down into 15 (19%) who were under the influence of alcohol, 11 (14%) who were under the influence of both alcohol and drugs and four (5%) who were under the influence of drugs alone (Table 15).
• In 2017-18, the alcohol and drug status of the accused was unknown for 49 persons, 60% of all accused. Two accused persons (2%) were reported to have been neither under the influence of alcohol nor drugs at the time of the homicide (Table 15).
• In the ten year period between 2008-09 and 2017-18, around half (46%) of all accused were reported to have been under the influence of alcohol and/or drugs at the time of the homicide. This is slightly higher for males (47%) than females (41%) (Table 16).
• Chart 11 shows that over the last ten years, 24% (96) homicides, where the accused was suspected of being under the influence of alcohol, occurred on a Saturday. Furthermore, nearly three fifths of homicides where the accused was under the influence of alcohol (60% or 240) occurred during the weekend (i.e. Friday, Saturday and Sunday).

Chart 11: Distribution of the accused of homicide under the influence of alcohol^[2] by day of the week 2008-09 to 2017-18

Comparator statistics

Table 4; Chart 12

• This bulletin has reported a decreasing trend in homicides over the longer term. Chart 12 shows this trend and compares it to the trend in attempted murder and serious assaults. This latter information is taken from the Recorded Crime in Scotland, 2017-18 bulletin. Over the ten year period between 2008-09 and 2017-18, the reduction in levels of homicide has broadly followed a similar fall in levels of attempted murder and serious assault. The number of homicide cases has fallen by 39% over the last ten years, compared with a 65% fall in attempted murder and a 31% fall in serious assault.

Chart 12: Indices in selected crimes of violence, Scotland, 2008-09 to 2017-18 (Index 2008-09 = 100)

• In a similar position to Scotland, England & Wales also produce National Statistics on Homicide, which is extracted from a dedicated database - the Home Office Homicide Index. The latest data published was for the 2016-17 financial year^[3].
• Direct comparisons between Scotland and England & Wales should be treated with some caution due to uncertainty around different counting and classification conventions.
• Further to this, users should note that in England & Wales, as in Scotland, it is possible for homicides that occurred sometime in the past to be recorded in more recent years. For example whilst the data for 2017-18 are not yet available, the Office for National Statistics (ONS) highlighted that the figures for 2016-17 included the 96 victims of Hillsborough which occurred in 1989, representing 14% of all victims in 2016-17.
• One area where some general comparisons can be made are the broad trends of change over time between Scotland and England & Wales, though still allowing for the caveats outlined above.
• In Scotland, homicides have fallen over the longer term but have been more stable in recent years. The 2017-18 figures for England & Wales will be published in early 2019. The latest publication for 2016-17 reported that homicide in England & Wales has also fallen over the longer term but has increased more recently (excluding the 96 victims of Hillsborough).