Exploring Differences in Criminogenic Risk Factors and Criminal Behavior Between Young Adult Violent Offenders With and Without Mild to Borderline Intellectual Disability

Abstract

The relation between mild to borderline intellectual disability (MBID) and violent offense behavior was studied among a group of former juvenile delinquents currently in a diversion program for persistent young adult violent offenders from Amsterdam (N = 146). Offenders were considered MBID if they had received juvenile probation from the local youth care agency specialized in intellectual disability (21%). A file study was used to estimate prevalence rates of criminogenic risk factors. Police data were used to depict recent criminal behavior. Nearly all offenders grew up in large and unstable multi-problem households and had psychosocial problems. More MBID offenders displayed externalizing behavior before the age of 12, were susceptible to peer pressure, and had low social-relational skills. MBID offenders committed more violent property crimes than offenders without MBID. Youth care interventions for MBID offenders should focus on the acquisition of social-relational skills and on the pedagogical skills of parents.

Keywords

borderline intellectual functioning mild intellectual disability violent crime young adulthood persistence

Introduction

Over recent years, evidence has been built up that juveniles with a (mild to borderline) intellectual disability (MBID) are overrepresented in the criminal justice system (e.g., Frize, Kenny, & Lennings, 2008; Kroll et al., 2002; Van der Put, Asscher, Stams, & Moonen, 2014). Although prevalence estimates of intellectual disability (ID) among juvenile criminal offenders are sparse, the available literature suggests that this overrepresentation accounts for the Dutch situation as well (e.g., Kaal, 2010; Teeuwen, 2012; Van Marle, 2004). The (composite) MBID definition is applied to groups with MBID. The definition comprises those having a mild ID (MID; 55 < IQ < 70) or those with borderline intellectual functioning (BIF; 70 < IQ < 85 with accompanying problems in adaptive functioning). Both levels of intellectual functioning are considered a disability only when they originate before the age of 18 (Schalock et al., 2010; Van Nieuwenhuijzen, Vriens, Scheepmaker, Smit, & Porton, 2011).

Juvenile offenders in general have received much attention from disciplines such as criminology, forensic psychiatry, and orthopedagogy. Juvenile offenders with ID however, have not sufficiently been considered a subgroup that merits scrutiny regarding the prevention of criminal behavior and recidivism, especially concerning violent crime. Also, compared with juveniles and adults, criminological research on young adults is still in its infancy (Blokland, Palmen, & Van San, 2012).

Overview of the Literature

With respect to population studies, literature reviews strongly indicate a relation between ID and delinquency (e.g., Farrington, 1995; Loeber, 1990; Loeber & Farrington, 1998; Moffit, Gabrielli, Mednick, & Schulsinger, 1981; Murray & Farrington, 2010), even while controlling for socioeconomic status (Taylor & Lindsay, 2010). Fairly recently, a study that set out to overcome common methodological limitations in empirical criminological research provided strong evidence for the association between IQ and delinquency (Beaver et al., 2013). There is no evidence for a causal relationship however. Although the possibility exists that ID in itself contributes to the development of delinquent behavior, a clear view on the exact relation between ID and criminal behavior is often polluted by the phenomenon that offenders with ID are more prone to be caught and arrested, prosecuted, and sentenced than offenders without ID (Boertjes & Lever, 2007). Some explanations for this phenomenon have been documented. Offenders with ID are found to be less able to manage anger and frustrations (Boertjes & Lever, 2007; Novaco & Taylor, 2004), to adopt a disrespectful or belligerent attitude toward police officers more often (Yun & Lee, 2013), and to be more prone to make false confessions while in police custody (Clare & Gudjonsson, 1993) than offenders without ID.

Research evidence points toward an inverse relation between IQ and offending rates (e.g., Goodman, Simonoff, & Stevenson, 1995; Hirschi & Hindelang, 1977; West & Farrington, 1973). This relation is not apparent for those with an IQ < 70 (e.g., McCord, McCord, & Zola, 1959; Taylor & Lindsay, 2010). Also, compared with offenders without ID, offenders with ID are more prone to commit property crimes and less prone to commit violent crimes (Van Marle, 2004), to engage in delinquent behavior at a younger age, to be more criminally active, and more likely to commit sexual offenses and arson (Barron, Hassiotis, & Banes, 2004).

Asscher, Van der Put, and Stams (2012) found no differences in the number of committed offenses between juvenile offenders with and without ID. They did find, however, that more juvenile offenders with ID had committed a crime against a person and had problems concerning general attitude, aggression, and social-relational skills. Compared with offenders without ID, substance use problems were less common.

Koolhof, Loeber, Wei, Pardini, and d’Escury (2007) found that different prevalence rates of criminogenic risk factors could largely be attributed to environmental characteristics that were more frequently problematic among juvenile offenders with ID (e.g., number of delinquent/antisocial peers, diluted pro-social network, bad neighborhood, detrimental housing) and less to personal characteristics. Also, accumulation of risks was higher among offenders with ID.

Last, Kaal, Brand, and Van Nieuwenhuijzen (2012) studied juveniles with a PIJ order (Placement in an Institution for Juveniles for mandatory treatment), the most severe measure in the Dutch juvenile justice system (Stevens & Van Marle, 2003; Van der Linden, Ten Siethoff, & Zeijlstra-Rijpstra, 2001). They found more similarities than differences between three different IQ groups—a finding that was largely attributed to the fact that among PIJ juveniles in general, problems are so widespread that IQ’s distinctive capacity is smaller than in other forensic groups (Kaal et al., 2012). More important is that the differences they did find related strongly to social (relational) skills, with the lowest IQ group having the most (severe) problems (e.g., low ego-strength, susceptibility to peer pressure, lack of problem awareness and insight).

In the present study, we elaborate on differences and similarities between offenders with and without ID. To be specific, the forensic group under focus consists of a sample of young adult violent offenders from Amsterdam with a history of juvenile probation. From this group, the presence of criminogenic risk factors during childhood and committed offenses during young adulthood are investigated. Our research questions are as follows:

Research Question 1: Which differences in prevalence rates and severity of criminogenic risk factors, present during childhood, can be identified between young adult violent offenders with and without MBID?

Research Question 2: What are the differences in type and number of committed offenses during young adulthood between young adult violent offenders with and without MBID?

Method

Sample

The study was centered on a cohort of former juvenile delinquents. This cohort consisted of men, born between 1985 and 1994, with a known history of juvenile probation in Amsterdam. From this cohort, only young adults included in a local diversion program for violent repeat offenders were drafted. This program started in 2011 and maintained the inclusion criteria that one had to have been an arrested suspect at least three times in the preceding 5 years (from 2007 to 2012) for a violent crime (attempts included) of which at least one had to have been committed (or attempted) in the last 2 years. Also, one had to have been arraigned to the examining magistrate in the same period. Offenders with only one or two of such violent crimes but with at least 33 additional arrests also met the program’s inclusion criteria. Examples of crimes considered by the program as violent are violent theft and burglary (residential home), manslaughter/homicide, public/aggravated assault, street robbery, and armed invasion/robbery.

MBID

Presence of MBID is notoriously difficult to determine. Consequently, a lack of solid information about MBID was anticipated due to either non-availability or uncertainty of IQ scores. Important to consider is that previous research has shown that IQ scores tend to be less reliable among ethnic minorities (e.g., De Jong & Colijn, 2010; Resing & Hessels, 2001).

Instead, the differentiation between violent offenders with and without MBID was based on the youth care agency that had executed case management during the periods the offenders received juvenile probation measures. In Amsterdam, there are two agencies that coordinate all voluntary and mandatory youth care programs, interventions, or treatments that a child receives. Jeugdbescherming Regio Amsterdam (JBRA) is a youth care agency responsible for child protection and juvenile probation and, in addition, offers preventive assistance to juveniles and their family members within a voluntary setting. The William Schrikker Groep (WSG) operates similarly but is a youth care agency specialized in the guidance of children and their family members with intellectual disabilities or chronic illnesses.

In retrospect, the general referral process of children who received juvenile probation in Amsterdam was as follows: Children were commonly referred to JBRA. Children were delegated to WSG when they were known to have (MB)ID and subjected to complex problems, before or during the period they were already under guidance. Also, if a child’s family or family member was already known by the WSG, the child in question was referred to WSG.

In practice, referral of a child to WSG because a family member was already known by WSG mostly occurred in the realm of child protection. Regarding juvenile probation, referral to WSG was almost always based on the level of intellectual functioning of the child. Professionals from both youth care agencies endorsed the current choice to retrospectively determine ID based on the guiding youth care agency. This classification is more likely to resemble a clinical assessment of ID than a bare IQ score (if available at all). So, all offenders having received guidance from WSG were considered to have MBID. A sensitivity analysis will be presented in the section “Discussion” in which we ratify this choice more thoroughly.

Criminogenic Risk Factors During Childhood

A file study was performed. Relevant information on criminogenic risk factors was extracted from archived youth care files. All violent offenders in our sample received youth care in Amsterdam between 1990 and 2011, most in 2004, and two thirds of all offenders received their first guidance before 2006.

Case management as executed by JBRA and WSG was similar in nature. Subsequently, youth care files from both organizations were similarly built up during the periods in which children had been under their active guidance. The content and extensiveness of these files vary, mainly according to the total time children spent under guidance and to the nature and severity of their problems. Valuable to the study were psychiatric and psychological reports and inquiries of the council for child protection. These have common ground and share language with instruments with which to quantify file information. Other documents of interest were police reports, orders from juvenile courts, and documentation on the progress of youth care programs.

Offense Behavior During Young Adulthood

Registration data on criminal offenses, committed in the period January 2007 to August 2012, of all offenders in our sample were provided by the police department of Amsterdam–Amstelland, with permission of the Dutch Council of Attorneys General from the Dutch Public Prosecutor’s Office. The following types and numbers of committed offenses could be recognized: violent theft/burglary (residential house), manslaughter/homicide, public assault against a person, aggravated assault, street robbery, (armed) invasion/robbery, property crime without violence, violent property crime, violent offense, drug-related offense, weapon-related offense, sexual offense, traffic offense, arson and general offenses in the public domain (e.g., public intoxication, nuisance, vandalism).

Instruments

File information was quantified with the Juvenile Forensic Profile (FPJ; Brand & Van Heerde, 2004, 2010), originally developed to retrieve data for forensic research from files from juveniles with a PIJ order. The FPJ consists of around 70 criminogenic risk factors and is based upon other validated measurement questionnaires, such as the Child Behavior Checklist (CBCL), the Youth Self-Report Questionnaire (YSR), and the Structured Assessment of Violence Risk in Youth (SAVRY).

All FPJ items are ordered into seven domains, namely, (1) history of criminal behavior, (2) upbringing & environment, (3) offense situation & drugs, (4) psychological factors, (5) psychopathology, (6) social & relational, and (7) behavior during stay in the institution (for an extensive overview of all risk factors included in the FPJ, see Brand & Van Heerde, 2010). The FPJ allows raters to choose per risk factor, with a few exceptions, from three answer options: 0 = no problem, 1 = moderate problem, or 2 = severe problem.

The FPJ has been shown to have good psychometric properties (Brand, 2005a, 2005b; Van Heerde & Mulder, 2005). It has an overall inter-rater reliability of r = .73 and Kappa = 0.61, which ranges between 0.45 (domain social & relational) and 0.85 (domain offense situation & drugs), and high convergent validity with the SAVRY was demonstrated (Kappa = 0.61; Van Heerde, Brand, Van ’t Hoff, & Mulder, 2004). Also, the predictive validity of the instrument was tested (see Brand, 2005a) with an area under receiver operating characteristic curve (AUC) of 0.803 for the first five domains.

Risk factors from the last domain behavior during stay in the institution were omitted. Although 76% of our sample had been detained during childhood, reports on their functioning in juvenile detention centers were only sparsely available. A few items were added to indicate general information about the family situation (e.g., dominant parent in upbringing, parental divorce/parents living separately, presence of a stepparent, number of siblings). From the existing domain history of criminal behavior, the items “number of non-violent offenses” and “number of violent offenses” were removed due to insufficient information. For the same reason, the items “number of offenses perpetrated in a group/solo” were replaced for “offenses predominantly committed solo/in a group.” Last, the item “somatic problems parents” was added to the domain upbringing & environment.

Procedure

Data collection started in October 2011 with a set of practice files that were “multiple” scored by four educated and trained raters independently. This process was continued until the threshold of 80% consensus of all administered scores was reached. Then, files were allocated to single raters who always had the opportunity to consult each other. Every 10th file was double-scored by at least two raters, but preferably all, independently.

Inter-Rater Reliability

Statistical measures were computed that offer a good indication of the attained inter-rater reliability (IRR): the raw correlation coefficient (r), the intra-class correlation coefficient (ICC), the proportion of raw agreement, and Kappa. Kappas were interpreted according to the Landis and Koch classification (Landis & Koch, 1977). All IRR statistics, with the exception of the ICC, were computed for each pair of raters separately and averaged into a mean consensus score. The ICC was computed for all raters simultaneously and interpreted according to Cicchetti (Cicchetti, 1994). IRR statistics were computed for the complete set of criminogenic risk factors and for each separate domain, except history of criminal behavior, which consisted of items with different levels of measurement (nominal, ordinal, and scale). These items are also not represented in the overall IRR statistics.

In total, 146 unique youth care files were quantified, of which 127 were scored by one rater only, and 19 files were multiple scored. This yielded a supply of 68 files from which IRR statistics could be computed. The IRR was r = 0.75 (p < .001), ICC = 0.78 (p < .001), 95% confidence interval (CI) = [0.75, 0.80], and Kappa = 0.62 (p < .001). The proportion of raw agreement was 77.3%. The proportion of raw disagreement in administered scores by one point (0 vs. 1 or 1 vs. 2) was 20.0% and by two points (0 vs. 2) was 2.7%. Domain specific analyses revealed IRR to be highest for the domains psychopathology and upbringing & environment and lowest for the domains social & relational and psychological factors. For these latter two domains, the proportion of disagreement by one point (e.g., no problem vs. moderate problem) was 41.1% and 33.5%, respectively, and the proportion of maximum disagreement (i.e., no problem vs. severe problem) was 3.3% and 2.9%, respectively (see Table 1).

Table 1.

Inter-Rater Reliability.

	r ^a	ICC^a	% agree^b (0-0/1-1/2-2)	% disagree (0-1/1-2)	% disagree (0-2)	Kappa^c
Total	.75	.78	77.3	20.0	2.7	0.62
Upbringing & environment	.75	.77	76.5	19.7	3.9	0.62
Offense situation & drugs	.56	.62	88.3	10.3	1.5	0.62
Psychological factors	.48	.52	63.6	33.5	2.9	0.37
Psychopathology	.77	.82	89.0	10.4	0.6	0.65
Social & relational	.44	.49	55.6	41.1	3.3	0.27

Note. ICC = intra-class correlation coefficient.

Strength of correlations (r, ICC) is qualified as “modest” (>.3), “strong” (>.5), and “very strong” (>.7).

Proportion raw agreement is qualified as “reasonable” (40% to 59%), “high” (60% to 79%), and “very high” (80% to 100%).

Kappa is qualified as “fair” (0.21-0.40), “moderate” (0.41-0.60), “good” (0.61-0.80), and “very good” (0.81-1.00).

Analyses

Analyses were performed with the SPSS 21.0 statistical package (SPSS Inc., Chicago, IL, USA). Chi-square tests were performed to determine differences between offenders with and without ID concerning the prevalence rates and severity of the majority of criminogenic risk factors and the prevalence rates of different types of offenses committed during young adulthood. ANOVAs were used to determine differences between the two groups on mean scores on the total number of committed offenses and the criminogenic risk factors: severity of violent offense, severity of sexual offense, age of first non-violent criminal offense, and age of first violent behavior.

Results

Sample

From our sample of 146 young adult violent offenders, 21% was considered to have a MBID (see Table 2). The mean age of offenders in our sample was 22.8 years (SD = 2.0) at the end date of the available police data. On average, MBID offenders were 1.4 years younger than offenders without MBID. The sociodemographic composition of both groups was roughly the same. The sample as a whole consisted of violent offenders with varying ethnic backgrounds (43% North African, 23% Surinamese, 10% native Dutch) who grew up in big and unstable households. A large majority had at least three siblings. More than half of all offenders had witnessed a parental divorce, around 45% grew up in a one-parent household, and 15% under unfavorable conditions (e.g., shifts from living as a child with parent(s) to living with other family members, with foster parents, or in boarding school; see Table 2).

Table 2.

Sample Characteristics of Young Adult Violent Offenders With and Without MBID.

Youth care history at	n	%
JBRA (no MBID)	116	79.5
WSG (MBID)	30	20.5
	MBID		No MBID		F	p
Mean age (SD)	21.7 (1.90)		23.1 (2.01)		10.72	.001
	n	%	n	%	χ²	p
Ethnicity					7.78	.557
Dutch	2	6.7	12	10.8
North African	11	36.7	52	46.8
Surinamese	10	33.3	22	19.8
Antillean	3	10.0	8	7.2
Western	2	6.7	4	3.6
Non-Western	2	6.7	13	11.7
Parental divorce	14	56.0	50	56.2	0.00	.987
Stepparent	7	29.2	22	25.9	0.10	.748
Dominant living situation during childhood					6.07	.299
Both parents	9	34.6	37	41.1
Mother	13	50.0	36	40.0
Father	—	—	3	3.3
Foster family	1	3.8	1	1.1
Other family members	2	7.7	2	2.2
Varying	1	3.8	11	12.2
Number of siblings					7.05	.632
None (only child)	2	8.0	6	6.5
1	3	12.0	17	18.5
2	3	12.0	17	18.5
3 or more	17	68.0	52	56.5

Note. MBID = mild to borderline intellectual disability; JBRA = Jeugdbescherming regio Amsterdam (Amsterdam’s central youth care agency); WSG = William Schrikker Groep (youth care agency specialized in guidance of youth with intellectual disability).

History of Criminal Behavior, Upbringing and Environment, and Offense Situation & Drugs

On these domains, no significant differences were found. On average, offenders from both groups were between 12 and 13 years old when they had committed their first non-violent offense and a little younger than 14 when they had displayed their first disruptive aggressive offense. Around 80% of offenders from both groups committed their offenses during childhood mainly in a group. Nearly all offenders from both groups (98%) had committed violent offenses during childhood, and, on average, their offenses were equally violent.

More MBID offenders displayed severe problem behavior before the age of 12 (i.e., early age of onset of problem behavior) and fewer were physically abused by their parent(s). No differences were observed on the other risk factors from this domain. No differences were observed on the risk factors from the domain offense situation & drugs (see Table 3).

Table 3.

Differences Between Young Adult Violent Offenders With and Without MBID in Terms of Prevalence and Severity of Criminogenic Risk Factors.

Domain 1. History of criminal behavior	Minimum-maximum	MBID (N = 30)				No MBID (N = 116)				χ²
Domain 1. History of criminal behavior	Minimum-maximum	n	% none	% moderate	% severe	n	% none	% moderate	% severe	χ²	p
Property crime, ever	0-1	30	0.0	—	100.0	113	3.5	—	96.5	0.18	.673
mainly without/with violence	0-1	26	42.3	—	57.7	91	40.7	—	59.3	0.00	1.000
Violent crime, ever	0-1	30	0.0	—	100.0	112	1.8	—	98.2	0.00	1.000
Sexual offence, ever	0-3	30	83.3	—	16.7	112	90.2	—	9.8	1.80	.625
Arson, ever	0-1	30	90.0	—	10.0	115	86.1	—	13.9	0.32	.572
Offenses: mainly solo (0) or in a group (1)	0-1	29	20.7	—	79.3	95	22.1	—	77.9	0.00	1.000
		n	Minimum-maximum	M (SD)		n	Minimum-maximum	M (SD)		F	p
Age of first non-violent offense		29	8-17	12.38 (2.46)		106	6-17	12.97 (2.18)		1.59	.209
Age of first violent behavior		28	11-16	13.71 (1.43)		100	4-19	13.85 (2.79)		0.06	.805
Maximum severity of violent crime, ever		30	0-4	2.63 (0.49)		112	0-4	2.79 (0.57)		1.99	.160
Domain 2. Upbringing & environment	Minimum-maximum	MBID (N = 30)				No MBID (N = 116)				χ²
Domain 2. Upbringing & environment	Minimum-maximum	n	% none	% moderate	% severe	n	% none	% moderate	% severe	χ²	p
Outplacement (due to own behavior)	0-1	30	80.0	—	20.0	115	76.5	—	23.5	0.03	.873
Outplacement (due to parents’ behavior)	0-1	30	90.0	—	10.0	115	89.6	—	10.4	0.00	1.000
Early age of onset of problem behavior	0-2	29	0.0	31.0	69.0	115	0.0	59.1	40.9	6.26	.012
Availability and accessibility of parents	0-2	30	26.7	36.7	36.7	115	31.3	48.7	20.0	3.73	.155
Adoption problems	0-2	30	96.7	3.3	0.0	116	97.4	1.7	0.9	0.727	.695
Peer rejection	0-2	27	59.3	33.3	7.4	109	70.6	22.0	7.3	1.48	.478
Parenting skills	0-2	30	3.3	43.3	53.3	110	6.4	32.7	60.9	1.38	.501
Authority problems	0-2	30	20.0	40.0	40.0	113	9.7	41.6	48.7	2.50	.287
Involvement with criminal peers	0-2	29	3.4	37.9	58.6	114	2.6	43.9	53.5	0.36	.837
Criminal behavior in the family	0-2	28	32.1	17.9	50.0	109	32.1	12.8	55.0	0.51	.775
Physical abuse by parents	0-2	27	59.3	33.3	7.4	100	52.0	20.0	28.0	6.59	.037
Physical abuse by others	0-2	29	86.2	10.3	3.4	112	81.2	14.3	4.5	0.410	.815
Neglect	0-2	29	62.1	31.0	6.9	108	64.8	22.2	13.0	1.53	.465
Sexual abuse by parents	0-2	30	100.0	—	—	112	100.0	—	—	—	—
Sexual abuse by others	0-2	29	96.6	3.4	0.0	113	95.6	2.7	1.8	0.57	.753
Domestic violence	0-2	20	65.0	15.0	20.0	92	57.6	15.2	27.2	0.479	.787
History of prior youth care	0-2	30	0.0	16.7	83.3	115	4.3	15.7	80.0	2.37	.306
Abnormal family (disrupted hierarchy)	0-2	29	51.7	24.1	24.1	111	49.5	25.2	25.2	0.04	.978
Escalating family situations	0-2	28	78.6	14.3	7.1	104	81.7	10.6	7.7	0.29	.859
Truancy/school dropout	0-2	29	17.2	17.2	65.5	112	15.2	25.9	58.9	0.94	.624
Parental substance abuse	0-2	26	80.8	7.7	11.5	107	78.5	5.6	15.9	0.43	.807
Parental mental health problems	0-2	27	74.1	14.8	11.1	97	78.4	12.4	9.3	0.22	.895
Parental somatic problems	0-1	26	46.2	—	53.8	88	54.5	—	45.5	0.28	.597
Threat to end up in prostitution	0-2	30	96.7	3.3	0.0	116	99.1	0.9	0.0	0.03	.875
Avoiding supervision/escape	0-2	30	30.0	66.7	3.3	115	13.9	80.0	6.1	4.45	.108
Domain 3. Offense situation & drugs	Minimum-maximum	MBID (N = 30)				No MBID (N = 116)				χ²
Domain 3. Offense situation & drugs	Minimum-maximum	n	% none	% moderate	% severe	n	% none	% moderate	% severe	χ²	p
During medicine abstinence/psychosis	0-2	28	100.0	—	—	115	98.3	0.9	0.9	0.49	.781
Under influence of any substance	0-2	22	59.1	36.4	4.5	65	40.0	58.5	1.5	3.50	.174
Under influence of alcohol	0-1	26	73.1	—	26.9	74	58.1	—	41.9	1.83	.176
Under influence of drugs	0-1	20	70.0	—	30.0	51	60.8	—	39.2	0.53	.468
Against familiar person	0-2	30	60.0	30.0	10.0	110	69.1	23.6	7.3	0.90	.639
Sexually related, in search of a victim	0-1	30	100.0	—	—	116	99.1	0.9	—	0.26	.610
Domain 4. Psychological factors	Minimum-maximum	n	% none	% moderate	% severe	n	% none	% moderate	% severe	χ²	p
Lack of empathy	0-2	30	6.7	56.7	36.7	100	10.0	61.0	29.0	0.79	.673
Impaired conscience development	0-2	30	0.0	40.0	60.0	104	1.0	60.6	38.5	4.72	.095
Ego-strength/susceptible (to peers)	0-2	30	3.3	46.7	50.0	109	15.6	60.6	23.9	8.97	.011
Low impulse control	0-2	30	3.3	40.0	56.7	101	13.9	37.6	48.5	2.58	.276
Lack of problem awareness/problem insight	0-2	28	0.0	42.9	57.1	105	2.9	49.5	47.6	1.41	.495
Domain 5. Psychopathology	Minimum-maximum	n	% none	% moderate	% severe	n	% none	% moderate	% severe	χ²	p
Addiction problems: gambling	0-2	29	100.0	—	—	99	98.0	1.0	1.0	0.60	.743
Addiction problems: alcohol	0-2	26	69.2	26.9	3.8	101	66.3	27.7	5.9	0.20	.907
Addiction problems: drugs	0-2	27	29.6	63.0	7.4	107	25.2	60.7	14.0	0.92	.631
ADHD/hyperactivity	0-2	30	70.0	20.0	10.0	111	77.5	15.3	7.2	0.73	.695
Anxiety disorder	0-2	30	96.7	3.3	0.0	108	92.6	5.6	1.9	0.83	.662
Depressive disorder	0-2	30	86.7	13.3	—	106	84.9	14.2	0.9	0.30	.860
Neurological problems	0-2	29	89.7	10.3	—	106	99.1	0.9	—	7.00	.008
Development personality disorder type B	0-2	27	22.2	37.0	40.7	81	25.9	25.9	48.1	1.22	.543
Feelings of hostility/aggressiveness	0-2	30	100.0	—	—	112	96.4	3.6	—	1.10	.294
Autism spectrum disorder	0-2	30	96.7	3.3	—	114	97.4	1.8	0.9	0.55	.760
Psychotic symptoms	0-2	29	100.0	—	—	113	94.7	4.4	0.9	1.61	.448
Sadism	0-2	30	90.0	10.0	—	116	95.7	4.3	—	1.49	.222
Sexually deviant behavior	0-2	29	89.7	10.3	—	114	88.6	7.9	3.5	1.19	.553
Pedosexual behavior	0-2	30	100.0	—	—	116	98.3	1.7	—	0.52	.469
Domain 6. Social & relational	Minimum-maximum	n	% none	% moderate	% severe	n	% none	% moderate	% severe	χ²	p
Negative affect/attitude	0-2	28	46.4	42.9	10.7	112	42.0	40.2	17.9	0.84	.657
Emotional support	0-2	29	65.5	34.5	0.0	107	64.5	32.7	2.8	0.84	.658
Total social network	0-2	28	10.7	75.0	14.3	114	12.3	79.8	7.9	1.12	.573
Secondary social network	0-2	28	0.0	50.0	50.0	112	2.7	58.9	38.4	1.80	.406
Social skills	0-2	27	22.2	40.7	37.0	109	54.1	36.7	9.2	16.13	.000

Note. MBID = mild to borderline intellectual disability; ADHD = attention-deficit/hyperactivity disorder.

Psychological Factors, Psychopathology, and Social & Relational

Concerning psychological factors, almost all offenders had a (severely) impaired conscience development. MBID offenders had a severely low ego strength/susceptibility to (peer) pressure more often. Most risk factors in the psychopathology domain did not differ between offenders with and without MBID. Among MBID offenders (indications for) neurological problems (e.g., atypical electroencephalogram [EEG], pregnancy toxemia, premature birth, sensory or motor deficiencies) were found more often. On average, more MBID offenders had problems with social skills than offenders without MBID (see Table 3).

Types of Committed Criminal Offenses During Young Adulthood

Overall, the most commonly committed types of offenses were property and/or violent crimes, followed by traffic offenses, general offenses in the public domain, and drugs- and weapon-related offenses. Sexual offenses and arson were committed by relatively few offenders. No differences in type of committed offenses were found between offenders with and without MBID (see Table 4).

Table 4.

The Proportion of Young Adult Violent Offenders With and Without MBID Who Committed an Offense at Least Once Between 2007 and 2012 and Differences of the Mean Number of Committed Offences Between Offenders With and Without MBID.

Types of offenses committed	MBID (N = 30)			No MBID (N = 116)			F	p
Types of offenses committed	n	%	M (SD)	n	%	M (SD)	F	p
Any	30	100.0	21.6 (13.6)	116	100.0	19.8 (13.8)	3.32	.071
Property crime no violence	27	90.0	7.9 (12.3)	105	90.5	5.2 (4.2)	3.50	.064
Violent property crime	27	90.0	3.7 (2.4)	95	81.9	2.4 (1.8)	9.64	.002
Violent offense	23	76.7	2.6 (1.3)	104	89.7	3.5 (2.3)	3.35	.070
Drug-related offense	12	40.0	1.8 (1.5)	36	31.0	1.7 (1.2)	0.04	.843
Weapon-related offense	11	36.7	1.4 (0.5)	33	28.4	1.5 (0.9)	0.09	.762
Sexual offense	2	6.7	1.0 (0.0)	3	2.6	1.0 (0.0)	—	—
Traffic offense	31	54.4	3.5 (2.7)	95	64.6	3.3 (2.8)	0.08	.774
General offenses public domain	18	60.0	7.4 (10.9)	69	59.5	4.0 (6.5)	2.89	.093
Arson	—	—	—	4	3.4	1.0 (0.0)	—	—
Other	28	93.3	3.9 (3.1)	94	81.0	4.8 (3.8)	1.45	.232

Note. MBID = mild to borderline intellectual disability.

Number of Committed Criminal Offenses During Young Adulthood

Only offenders who had committed a specific type of offense at least once were included in the analyses to determine differences in the level of criminal activity between both groups. Overall, with around 20 offenses committed, both groups of offenders had been equally active. When considering specific types of offenses, MBID offenders had committed more violent property crimes than offenders without MBID (3.7 vs. 2.4).

Discussion

Among a specific group of young adult offenders of violent crimes, our study explored differences between offenders with and without MBID. We focused on the prevalence rates and severity of criminogenic risk factors and on types of committed offenses and criminal activity during young adulthood.

In anticipation of a lack of solid information on determined MBID, all offenders who had received juvenile probation from Amsterdam’s youth care agency that specialized in ID youth were considered MBID. Indeed, many offenders’ youth care files did not mention IQ scores (34%). Also, our study population consisted in large part (90%) of offenders with a non-ethnic Dutch background, and IQ scores tend to be less reliable among ethnic minorities (e.g., De Jong & Colijn, 2010; Resing & Hessels, 2001).

We found that, in comparison with offenders without MBID, more MBID offenders displayed severe externalizing and disruptive behavior (i.e., at a younger age), fewer suffered from physical abusive parents, more were susceptible to peer pressure, more had (indications for) neurological problems, and more had problems concerning social-relational skills. No differences in types of committed offenses during young adulthood were found. Concerning levels of criminal activity, however, MBID offenders had committed more violent property crimes than offenders without MBID.

Although a multitude of specific criminogenic risk factors were included in our study, the two groups differed on not so many. MBID seems inept to distinguish a meaningful subpopulation of young adult violent offenders. A likely explanation, just as Kaal et al. (2012) described in their study of PIJ juveniles, is that among our study population, childhood problems were present at such high levels that MBID’s distinctive capacity is largely negated. After all, we did find unmistakingly high prevalence and severity rates of many criminogenic risk factors in both groups (e.g., early age of onset of problem behavior, lack of parenting skills, high involvement with antisocial/criminal peers, truancy, authority problems, impaired conscience and impulse control, diluted pro-social networks, and development of type B personality disorders).

Nonetheless, the differences we did find do have important implications. As said, MBID offenders displayed externalizing and disruptive problematic behavior earlier in life than offenders without MBID. Some displayed these behaviors even in kindergarten. This finding warrants early recognition of problem behavior and the necessity of timely interventions. Specifically, because early preventive interventions, such as the Nurse-Family Partnership Program (NFP; Mejdoubi et al., 2011; Olds et al., 1998; Olds, Hill, Mihalic, & O’Brien, 2001; Van der Duin, 2006), which focuses on ineffective parenting techniques, are associated with a reduction of delinquent behavior (e.g., Lundman, 1993).

Although problems concerning parenting skills were found to be highly prevalent and mostly severe for both groups of offenders, it seems that among the MBID group, one should be more inclined to think of unable rather than unwilling parents (e.g., Tymchuk & Feldman, 1991; Wade, Llewellyn, & Matthews, 2008). This is a notion that requires further investigation. These findings imply that it is important that interventions aim to affect behavior—and/or attitude change—of parents of under-aged MBID offenders, if considered necessary. In essence, interventions would need to dissolve parent-related risk factors, but, in a best-case scenario, even to transform these into protective factors.

Timing is an important aspect to consider. It is known that risk factors from the parents/family domain are found to be most strongly related to problem behavior before the age of 12. Van der Put et al. (2012) found that during the pre-adolescence phase, risk factors with a heavy impact on behavior tend to shift from the parents to the peers domain around the age of 13 and to the school and peers domain from 14 years onward. Interventions aimed to enhance the role of parents (especially those of ID youths) should best be deployed in a timely manner, that is, well before the age of 12. When criminal behavior presents at an older age (too late for parental involvement), interventions should primarily focus on dynamic risk factors.

Another finding of our study, in line with the literature (e.g., Asscher et al., 2012; Kaal et al., 2012; Van der Put et al., 2014), is that juvenile offenders with MBID presented with more problems concerning social and relational skills and were found to be more often more severely susceptible to peer pressure than juveniles without MBID. This finding confirms that juvenile MBID offenders have problems in their cognitive abilities related to the processing of social information (e.g., Gardner & Moffatt, 1990; Van Nieuwenhuijzen, Orobio de Castro, Wijnroks, Vermeer, & Matthys, 2004). It also indicates that interventions should focus on enhancement of social-relational skills. The What Works principles (Andrews & Bonta, 2006) and, from the derivative risk-needs-responsivity (RNR) model, most notably the needs principle (Bonta & Andrews, 2007) offer useful suggestions to do so.

Despite the fact that no differences were found concerning substance use and having committed offenses under the influence, youth care should be vigilant on addiction behavior among MBID youth. Literature suggests that although alcohol/drug use is less prevalent among juveniles with MBID, those who do are more prone to suffer from alcohol and drug-related problems (e.g., Bransen, Schipper, & Blekman, 2009; McGillicuddy, 2006; Van der Nagel, Kiewik, & Didden, 2012), of which criminal behavior is just one example.

All offenders in our sample have been under guidance of youth care. The extent to which the, then still under-aged, offenders were willing to accept youth care’s interference in their lives was strikingly low. To increase the attained potential of future interventions, juveniles’ acceptance of and motivation for youth care should be enhanced. The establishment of a good working alliance and base of trust between youth care workers and the offenders themselves, but also with their parents, seems an important prerequisite.

Again, the What Works principles (Andrews & Bonta, 2006) offer strong footing to do so. Especially the responsivity principle (Andrews & Bonta, 2006; Andrews, Bonta, & Hoge, 1990) should be well incorporated to enhance future interventions’ effectiveness by tailoring these to individual offenders or at-risk juveniles (Wong & Hare, 2005). Aligning interventions to cognitive capacity, learning style, cultural background, language mastery, or fluency are some examples of how to do so.

To further inquire the relation between criminogenic risk factors present during childhood and violent offense behavior during (young) adulthood with a focus on MBID, a follow-up study with a positive reference group seems obvious. Such a reference group could, for example, consist of peers (similar age) who have also received juvenile probation during childhood but who do not display persistent criminal behavior. Another comparison could focus on the escalation into violent crime. Such comparisons could also identify protective factors that buffer against the persistence or escalation of criminal behavior during the transition from adolescence into young adulthood. Recently, we finished the data collection of such a study, and we expect to publish these results in the near future.

Important to bear in mind is that a reasonable proportion of our study population (21%) was strongly indicated to have MBID. Knowing that ID is related to aggressive and delinquent behavior (Taylor & Lindsay, 2010), in combination with the fact that we studied violent offenders in particular of whom a significant part is strongly indicated to have MBID, necessitates profound attention to MBID, in general, when executing diversion programs or interventions aimed at reducing violent offenses or at preventing at-risk youth from drifting toward a violent criminal career.

Strengths and Weaknesses

We performed a systematic observation of historical information archived in youth care files for which a reliable and valid quantification instrument was used. Observers were thoroughly trained and a good inter-rater reliability was attained. Besides historic file data, we used objective police registration data on offenses committed during young adulthood. Important limitations of our study are that results can only be generalized to male offender populations. Also, because the measurement instrument used consists solely of environmental risk factors, our study did not take into account heritable features associated with both externalizing and disruptive behavior (e.g., Hicks, Krueger, Iacono, McGue, & Patrick, 2004).

In many youth care files, information was insufficiently available (and sometimes contradictory) to indicate the presence or absence of all criminogenic risk factors included in the scoring instrument FPJ with a decent amount of certainty. Raters were instructed to grant every file a quality score on a scale from 1 to 10. Quality scores were interpreted as a grade (ranging from 1 = very poor to 10 = excellent) that was mainly dependent on the amount of missing information and, to a lesser extent, on the amount of contradicting information within a file. A mean quality score of 6.2 was recorded. Scores were most often missing for the risk factors domestic violence, parents’ problems (e.g., addiction, psychiatric, and somatic problems) and on all risk factors from the FPJ domain offense-related risk factors and substance use.

Finally, we recognize problems with the validity and international generalizability of our findings, considering that offenders with and without MBID were distinguished based on service use in the local Amsterdam situation. However, within the context of the study, we were unable to use assessment or screening instruments to determine the presence of (indications for) MBID directly. To endorse our current choice to define offenders who have been under guidance of the youth care agency specialized in ID youth (WSG) as MBID, the distribution of offenders with a history of youth care at WSG or JBRA (Amsterdam’s central youth care agency) over the internationally used cutoff scores for MBID (Schalock et al., 2010) is presented, alongside a sensitivity analysis. Again, we would like to stress that IQ scores were unavailable for a significant part of the study sample (34%), and that available IQ scores should not be considered highly reliable (partly due to the ethnic composition of the study sample).

Table 5 shows misclassifications. From 29 offenders with a history of youth care at WSG, one had an IQ ≥ 85. Of 67 offenders with a history of youth care at JBRA, 19 had an 70 < IQ < 85 and four an IQ <70 (χ² = 22.86; p < .001). This implies that the two groups do have some overlap that, besides (lack of) statistical power issues, may account for the relatively few significant differences found between the two groups.

Table 5.

Distribution of Offenders With a Known IQ Score and History at WSG or JBRA Over Three IQ Groups.

Classification	IQ score	WSG		JBRA
Classification	IQ score	n	%	n	%
Mild intellectual disability	55-70	18	62.1	4	6.0
Borderline intellectual functioning	71-84	10	34.5	19	28.4
Low average intelligence and higher	85+	1	3.4	44	65.7

Note. IQ = intelligence quotient; WSG = William Schrikker Groep (youth care agency specialized in guidance of youth with intellectual disabilities); JBRA = Jeugdbescherming regio Amsterdam (Amsterdam’s central youth care agency).

Therefore, we also conducted a sensitivity analysis in which an alternative composition of the two groups was used. Only offenders guided by WSG and with an IQ < 85 were considered MBID (n = 28). All other offenders were considered no MBID (n = 118). Results from this alternative comparison largely replicated our earlier findings, and all statistically significant differences pointed in the same directions. However, the criminogenic risk factors concerning physical abuse by parent(s) and neurological problems failed to reach statistical significance. Concerning recent offense behavior, all results were similar. In general, this sensitivity analysis did not necessitate alteration of our main conclusions.

Conclusion

At least one in five of a group of young adult violent offenders from Amsterdam is strongly suspected to have a mild to borderline ID. Relatively few differences concerning prevalence rates of criminogenic risk factors during childhood and offense behavior during young adulthood between violent offenders with and without MBID were found. The group as a whole is characterized by many and severe problems that largely negate MBID’s distinctive capacity. However, the observed differences do offer useful insights to enhance the effectiveness of youth care interventions among juvenile offenders with MBID.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Andrews

D. A.

Bonta

(2006). The psychology of criminal conduct (4th ed.). Newark, NJ: LexisNexis.

Andrews

D. A.

Bonta

Hoge

R. D.

(1990). Classification for effective rehabilitation: Rediscovering psychology. Criminal Justice and Behavior, 17, 19-52.

Asscher

J. J.

Van der Put

C. E.

Stams

G. J. J.

(2012). Differences between juvenile offenders with and without intellectual disability in offense type and risk factors. Research in Developmental Disabilities, 33, 1905-1913.

Barron

Hassiotis

Banes

(2004). Offenders with intellectual disability: A prospective comparative study. Journal of Intellectual Disability Research, 48, 69-76.

Beaver

K. M.

Schwartz

J. A.

Nedelec

J. L.

Connolly

E. J.

Boutwell

B. B.

Barnes

J. C.

(2013). Intelligence is associated with criminal justice processing: Arrest through incarceration. Intelligence, 41, 277-288.

Blokland

Palmen

Van San

(2012). Crimineel gedrag in de jongvolwassenheid [Criminal behavior in young adulthood]. Tijdschrift voor Criminologie, 54(2), 85-98.

Boertjes

M. J.

Lever

M. S.

(2007). LVG en jeugdcriminaliteit [Mild intellectual disability and juvenile crime]. Diemen, The Netherlands: Expertisecentrum Jeugdzorg—Gehandicaptenzorg William Schrikker.

Bonta

Andrews

D. A.

(2007). Risk-need-responsivity model for offender assessment and rehabilitation. Rehabilitation, 6, 1-22.

Brand

E. F. J. M

. (2005a). Een maat voor de kwaliteit van instrumenten voor risicotaxatie [A measure for the quality of instruments for risk assessment]. In Sjerps

M. F.

Coster van Voorhout

J. A.

(Eds.), Het onzekere bewijs. Gebruik van statistiek en kansrekening in het strafrecht (pp. 429-456). Deventer, The Netherlands: Kluwer.

10.

Brand

E. F. J. M

. (2005b). Onderzoeksrapport PIJ-dossiers 2003C: Predictieve validiteit van de FPJ-lijst [Research study PIJ-files 2003C: Predictive validity of the FPJ-list]. The Hague, The Netherlands: Department of Justice, National Agency of Correctional Institutions.

11.

Brand

E. F. J. M.

Van Heerde

W. K.

(2004). Handleiding FPJ-lijst. Forensisch profiel justitiële jeugdigen [Scoring Manual FPJ-list. Juvenile forensic profile]. Den Haag, The Netherlands: Dienst Justitiële Inrichtingen, Department of Justice, National Agency of Correctional Institutions.

12.

Brand

E. F. J. M.

Van Heerde

W. K.

(2010). Handleiding FPJ-lijst. Forensisch profiel justitiële jeugdigen [Scoring Manual FPJ-list. Juvenile forensic profile]. The Hague, The Netherlands: Department of Justice, National Agency of Correctional Institutions.

13.

Bransen

Schipper

Blekman

(2009). Middelengebruik door jongeren met een licht verstandelijke handicap [Substance use among juveniles with a mild intellectual disorder]. Verslaving, 5(4), 37-49.

14.

Cicchetti

D. V.

(1994). Guidelines, criteria, and rules of thumb for evaluating normed and standardized assessment instruments in psychology. Psychological Assessment, 6, 284-290.

15.

Clare

I. C. H.

Gudjonsson

G. H.

(1993). Interrogative suggestibility, confabulation, and acquiescence in people with mild learning disabilities (mental handicap): Implications for reliability during police interrogations. British Journal of Clinical Psychology, 32, 295-301.

16.

De Jong

Colijn

(Eds.). (2010). Handboek culturele psychiatrie en psychotherapie [Handbook cultural psychiatry and psychotherapy]. Utrecht, The Netherlands: De Tijdstroom.

17.

Farrington

D. P.

(1995). The development of offending and antisocial behaviour from childhood: Key findings from the Cambridge study in delinquent development. Journal of Child Psychology and Psychiatry, 36, 929-964.

18.

Frize

Kenny

Lennings

(2008). The relationship between intellectual disability, indigenous status and risk of reoffending in juvenile offenders on community orders. Journal of Intellectual Disability Research, 52, 510-519.

19.

Gardner

W. I.

Moffatt

C. W.

(1990). Aggressive behaviour: Definition, assessment, treatment. International Review of Psychiatry, 2, 91-100.

20.

Goodman

Simonoff

Stevenson

(1995). The impact of child IQ, parent IQ and sibling IQ on child behavioural deviance scores. Journal of Child Psychology and Psychiatry, 36, 409-425.

21.

Hicks

B. M.

Krueger

R. F.

Iacono

W. G.

McGue

Patrick

C. J.

(2004). Family transmission and heritability of externalizing disorders: A twin-family study. Archives of General Psychiatry, 61, 922-928.

22.

Hirschi

Hindelang

M. J.

(1977). Intelligence and delinquency: A revisionist view. American Sociological Review, 42, 571-587.

23.

Kaal

H. L.

(2010). Beperkt en gevangen. De haalbaarheid van prevalentieonderzoek naar verstandelijke beperking in detentie [The feasibility of prevalence research into intellectual disability in detention]. Den Haag, The Netherlands: Wetenschappelijk Onderzoek- en Documentatiecentrum, Ministerie van Veiligheid & Justitie.

24.

Kaal

H. L.

Brand

E. F. J. M.

Van Nieuwenhuijzen

(2012). Serious juvenile offenders with and without intellectual disabilities. Journal of Learning Disabilities and Offending Behaviour, 3, 66-76.

25.

Koolhof

Loeber

Wei

E. H.

Pardini

d’Escury

A. C.

(2007). Inhibition deficits of serious delinquent boys of low intelligence. Criminal Behaviour and Mental Health, 17, 274-292.

26.

Kroll

Rothwell

Bradley

Shah

Bailey

Harrington

R. C.

(2002). Mental health needs of boys in secure care for serious or persistent offending: A prospective, longitudinal study. The Lancet, 359, 1975-1979.

27.

Landis

J. R.

Koch

G. G.

(1977). The measurement of observer agreement for categorical data. Biometrics, 33, 159-174.

28.

Loeber

(1990). Development and risk factors of juvenile antisocial behavior and delinquency. Clinical Psychology Review, 10, 1-41.

29.

Loeber

Farrington

D. P.

(Eds.). (1998). Serious and violent juvenile offenders: Risk factors and successful interventions. Thousand Oaks, CA: Sage.

30.

Lundman

R. J.

(1993). Prevention and control of juvenile delinquency (2nd ed.). New York, NY: Oxford University Press.

31.

McCord

Zola

(1959). Origins of crime: A new evaluation of the Cambridge-Somerville Youth Study. New York, NY: Columbia University Press.

32.

McGillicuddy

N. B.

(2006). A review of substance use research among those with mental retardation. Mental Retardation & Developmental Disabilities Research Reviews, 12, 41-47.

33.

Mejdoubi

Van den Heijkant

Struijf

Van Leerdam

HiraSing

Crijnen

(2011). Addressing risk factors for child abuse among high risk pregnant women: Design of a randomised controlled trial of the nurse family partnership in Dutch preventive health care. BMC Public Health, 11, 823.

34.

Moffit

T. E.

Gabrielli

W. F.

Mednick

S. A.

Schulsinger

(1981). Socioeconomic status, IQ, and delinquency. Journal of Abnormal Psychology, 90, 152-157.

35.

Murray

Farrington

D. P.

(2010). Risk factors for conduct disorder and delinquency: Key findings from longitudinal studies. Canadian Journal of Psychiatry/Revue canadienne de psychiatrie, 55, 633-642.

36.

Novaco

R. W.

Taylor

J. L.

(2004). Assessment of anger and aggression in male offenders with developmental disabilities. Psychological Assessment, 16, 42-50.

37.

Olds

Henderson

C. R.

Jr. Cole

Eckenrode

Kitzman

Luckey

. . . Powers

(1998). Long-term effects of nurse home visitation on children’s criminal and antisocial behavior: 15-year follow-up of a randomized controlled trial. Journal of the American Medical Association, 280, 1238-1244.

38.

Olds

Hill

P. L.

Mihalic

S. F.

O’Brien

(2001). Nurse-family partnership: Blueprints for violence prevention. Boulder: Center for the Study and Prevention of Violence, Institute of Behavioral Science, University of Colorado.

39.

Resing

W. C. M.

Hessels

M. G. P.

(2001). Het meten van de cognitieve mogelijkheden en het schoolgedrag van allochtone kinderen [Measuring cognitive abilities and school behavior in immigrant children in the Netherlands]. In Bleichrodt

van de Vijver

(Eds.), Diagnostiek bij allochtonen: mogelijkheden en beperkingen van psychologische tests (pp. 89-118). Lisse, The Netherlands: Swets & Zeitlinger.

40.

Schalock

R. L.

Borthwick-Duffy

S. A.

Bradley

V. J.

Buntinx

W. H.

Coulter

D. L.

Craig

E. M.

. . . Yeager

M. H.

(2010). Intellectual disability: Definition, classification, and systems of supports. Washington, DC: American Association on Intellectual and Developmental Disabilities.

41.

Stevens

Van Marle

H. J. C.

(2003). De PIJ-maatregel als opvoedingsinstrument. De PIJ-maatregel en de casuïstiek [The PIJ-order as a way of re-education]. Sancties, 4, 221-234.

42.

Taylor

Lindsay

W. R.

(2010). Understanding and treating offenders with learning disabilities: A review of recent developments. Journal of Learning Disabilities and Offending Behaviour, 1(1), 5-16.

43.

Teeuwen

(2012). Verraderlijk gewoon: Licht verstandelijk gehandicapte jongeren, hun wereld en hun plaats in het strafrecht [Deceitfully normal: Mild intellectually disabled juveniles, their world and their place in criminal law]. Amsterdam, The Netherlands: Uitgeverij SWP.

44.

Tymchuk

A. J.

Feldman

M. A.

(1991). Parents with mental retardation and their children: Review of research relevant to professional practice. Canadian Psychology/Psychologie Canadienne, 32, 486-496.

45.

Van der Duin

. (2006). Effectieve jeugdinterventies: VoorZorg [Effective youth interventions: Precaution]. Utrecht, The Netherlands: Nederlands Jeugdinstituut. Retrieved from http://www.nji.nl/jeugdinterventies

46.

Van der Linden

A. P.

Ten Siethoff

F. G. A.

Zeijlstra-Rijpstra

A. E. I. J

. (2001). Jeugd en Recht [Juvenile law]. Houten, The Netherlands: Bohn Stafleu Van Loghum.

47.

Van der Nagel

Kiewik

Didden

. (2012). Iedereen gebruikt toch? Handboek LVB en verslaving [Everybody uses, right? Handbook mild intellectual disability and addiction]. Amsterdam, The Netherlands: Boom.

48.

Van der Put

C. E.

Asscher

J. J.

Stams

G. J. J. M.

Moonen

X. M. H

. (2014). Differences between juvenile offenders with and without intellectual disabilities in the importance of static and dynamic risk factors for recidivism. Journal of Intellectual Disability Research, 58, 992-1003. doi:10.1111/jir.12078

49.

Van der Put

C. E.

Stams

G. J. J. M.

Hoeve

Deković

Spanjaard

H. J. M.

Van der Laan

P. H.

Barnoski

R. P

. (2012). Changes in the relative importance of dynamic risk factors for recidivism during adolescence. International Journal of Offender Therapy and Comparative Criminology, 56, 296-316. doi:10.1177/0306624X11398462

50.

Van Heerde

W. K.

Brand

E. F. J. M.

Van ’t Hoff

G. I. C. M.

Mulder

E. A.

(2004). Interbeoordelaarsbetrouwbaarheid en convergente validiteit van de Profiel Justitiële Jeugdlijst [Inter-rater reliability and convergent validity of the Juvenile Forensic Profile]. Den Haag, The Netherlands: Ministerie van Justitie, Dienst Justitiële Inrichtingen.

51.

Van Heerde

W. K.

Mulder

E. A

. (2005). Rapport PIJ-dossiers 2003-B. Praktijkevaluatie FPJ-lijst & SAVRY te Eikenstein en FORA [Research study PIJ-files 2003-B. Clinical evaluation FPJ-list & SAVRY in Eikenstein and FORA]. The Hague, The Netherlands: Department of Justice, National Agency of Correctional Institutions.

52.

Van Marle

(2004). Hulpverlening of justitie: Keuzes in de zorg voor verstandelijk gehandicapten [Care or justice: Choices in the care for mentally handicapped]. In Geus

Roelvink

(Eds.), Verstandelijk gehandicapten in aanraking met politie en justitie: Verslag van een congres (pp. 43-56). Utrecht, The Netherlands: Nederlands Instituut voor Zorg en Welzijn.

53.

Van Nieuwenhuijzen

Orobio de Castro

Wijnroks

Vermeer

Matthys

. (2004). The relations between intellectual disabilities, social information processing, and behavior problems. European Journal of Developmental Psychology, 1, 215-229.

54.

Van Nieuwenhuijzen

Vriens

Scheepmaker

Smit

Porton

. (2011). The development of a diagnostic instrument to measure social information processing in children with mild to borderline intellectual disabilities. Research in Developmental Disabilities, 32, 358-370.

55.

Wade

Llewellyn

Matthews

(2008). Review of parent training interventions for parents with intellectual disability. Journal of Applied Research in Intellectual Disabilities, 21, 351-366.

56.

West

D. J.

Farrington

D. P.

(1973). Who becomes delinquent? Second report of the Cambridge study in delinquent development. London, England: Heinemann Educational.

57.

Wong

S. C. P.

Hare

R. D.

(2005). Guidelines for a psychopathy treatment program. Toronto, Canada: Multi-Health Systems.

58.

Yun

Lee

(2013). IQ and delinquency: The differential detection hypothesis revisited. Youth Violence and Juvenile Justice, 11, 196-211. doi:10.1177/1541204012463410