Forensic Science International 236 (2014) 146–150
Contents lists available at ScienceDirect
Forensic Science International journal homepage: www.elsevier.com/locate/forsciint
The reliability and validity of the rating scale of criminal responsibility for mentally disordered offenders Weixiong Cai *, Qingting Zhang, Fuyin Huang, Wei Guan, Tao Tang, Chao Liu Shanghai Key laboratory of Forensic Medicine, Institute of Forensic Science, Ministry of Justice, Shanghai 200063, PR China
A R T I C L E I N F O
A B S T R A C T
Article history: Received 6 July 2011 Received in revised form 2 April 2013 Accepted 17 December 2013 Available online 13 January 2014
In China, the criminal responsibility of the mentally disordered offenders is divided into three levels, there are the whole responsibility, diminished responsibility and irresponsibility. According to the Criminal Law, ‘‘If a mental disordered patient causes harmful consequences at a time when he is unable to recognize or control his own conduct, upon verification and confirmation through legal procedure, he shall not bear criminal responsibility.’’ That means there are two standards of assessing criminal responsibility, namely volitional and cognitive capacity. It is as equal as the Mc’Naughton Rule and the Irresistible Impulse Test. But for a long time, the criminal responsibility was assessed mainly by experience because of lacking of standardized assessment instrument. Recently, we have developed ‘‘the rating scale of criminal responsibility for mentally disordered offenders (RSCRs)’’. The scale includes eighteen items, namely criminal motivation, aura before offense, inducement of crime, time and place and object and tool selectivity of crime, emotion during the crime, shirking responsibility after offense, concealing the truth during inquest, camouflage, understanding the nature of the offense, estimating the consequence of the offense, impairment of life ability, impairment of learning or work, impairment of insight, impairment of reality testing, and impairment of self-control. This scale can be applicable for all cases and easy to use. This scale had been tried out in several forensic psychiatry institutes, the Cronbach a of the scale is 0.93, and all items have high correlation with the total score of the scale (r = 0.50–0.89). Two factors were extracted by the factorial analysis, and the cumulative squared loading was 68.62%. The scores of the three levels were 9.66 5.11, 26.54 5.21 and 40.08 7.90 respectively and highly significant differences were observed among groups. By establishing discrimination analysis among three levels, classification results suggested that 88.90% of the original grouped cases were correctly classified, and the discriminant value had high conformity with the experts’ opinions. The data showed that the scale would be the best validated instrument for the criminal responsibility in China. ß 2013 Elsevier Ireland Ltd. All rights reserved.
Keywords: Forensic psychiatry Criminal responsibility Rating scale Mentally disordered offender RSCR
1. Introduction The identification of criminal responsibility in China remains controversial, fueled by public antagonism and professional dissension. During the last several decades, many forensic psychiatrists were tempted to rely on empirical evaluation and traditional measures on which they were trained and gained years of intensive experience, the objectivity and impartiality were doubted [1,2]. Which measure is the most effective for the assessment of criminal responsibility? In the United States, the Rogers criminal responsibility assessment scales (R-CRAS) had been applied to the insanity defense [3–6]. However, the lack of standard assessment tools exists all the while in the forensic psychiatry field in China. Recently, we have developed the
* Corresponding author. Tel.: +86 21 52359576; fax: +86 21 52359576. E-mail address: [email protected] (W. Cai). 0379-0738/$ – see front matter ß 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.forsciint.2013.12.018
assessment measure named ‘‘the rating scale of criminal responsibility for mentally disordered offenders (RSCRs)’’. The purpose of the present study was to examine the reliability and validity of this rating scale in four hospitals and one institute from different provinces in China. 2. Method and object 2.1. Object 1187 objects were recruited from the Institute of Forensic Science, Ministry of Justice, the Second Xiangya Hospital of Central South University, the Mental Health Center of Shanghai, the Psychosis Hospital of Guangzhou and Nanjing Brain Hospital from January 2007 to May 2009. Among these objects, 891 were male and 296 were female, mean age was 35.52 years (SD = 10.635, range 14–78). The educational level: 19 were illiterate, 378 were primary school, 494 were junior high school, 203 were high school,
W. Cai et al. / Forensic Science International 236 (2014) 146–150
147
Table 1 The diagnosis type in different groups (number, %). Group
Schizophrenia and paranoid disorder
Affective disorder
Organic and mental retardation
Others
Irresponsibility (n = 448) Diminished responsibility (n = 276) Whole responsibility (n = 463)
381(85.0) 118(42.8) 48(10.4)
15(3.4) 42(15.2) 17(3.7)
14(3.1) 63(22.8) 59(12.7)
38(8.5) 53(19.2) 339(73.2)
Table 2 The statistic scores in different groups. Group
Mean
Standard deviation
Standard error
Irresponsibility (n = 448) Diminished responsibility (n = 276) Whole responsibility (n = 463)
9.12 26.33 43.73
6.29 7.59 7.03
0.30 0.46 0.33
and 93 were college or above. Marriage situation: 318 were married, 572 were unmarried, 92 were divorced, and 205 were unknown. Occupation: 276 were farmer, 97 were worker, 67 were official, 481 were unemployed, 22 were self-employed, 68 were others and 186 were no details. The criminal type: 322 were homicide, 347 were harm, 175 were robbery or extortion, 49 were rape, 48 were arson, 120 were theft, and 126 were swindle and other crimes. Diagnosis: 482 were schizophrenia, 74 were affective disorder, 96 were mental retardation, 40 were organic mental disorder, 65 were paranoid mental disorder, 8 were neurosis, 6 were stress related disorders, 35 were personality disorder, 12 were substance-dependent mental disorder, 267 were other mental disorder and 102 were without psychosis.
mental disorder was mild, and the capacity of cognition or volition was intact; diminished responsibility: suffered from mental disorder and the capacity of cognition or volition was impaired due to the psychiatric symptoms; irresponsibility: suffered from mental disorder and the capacity of cognition or volition was lost completely due to the psychiatric symptoms. Double-blind principle was complied in the research, i.e. the raters used the scale independently and did not take part in the discussion to the object so it is impossible for him/her to know the experts’ opinion to the object’s criminal responsibility, on other hand, the forensic psychiatrists did not know the rating result. The data were analyzed using the SPSS 16.0 software package. 3. Results
2.2. Method 3.1. The interrater reliability of the scale The RSCR was developed on the base of the Scale for Diminished Criminal Responsibility, the Rogers Criminal Responsibility Scale (R-CRAS) [4,6,7]. The scale consist of 18 items, namely criminal motivation, aura before offense, inducement of crime, time selectivity of crime, place selectivity of crime, object selectivity of crime, tool selectivity of crime, emotion during the crime, shirking responsibility after offense, concealing the truth during inquest, camouflage, understanding the nature of the behavior, estimating the consequence of the offense, impairment of life ability, Impairment of learning or work, impairment of insight, impairment of testing-reality, impairment of self-control. The point for 18 items is from 0 to 4. The diagnosis and criminal responsibility level of the object was also made by professional groups, which were made up by more than three forensic psychiatrists according to CCMD-3 (Chinese Classification of Mental Disorders, the third revision) and the criteria: whole responsibility: no psychosis or the severity of
10 cases were assessed respectively by six observers, there was a high inter-rater reliability among all items (kappa = 0.785, p < 0.001), which shows that the scale is stable. 3.2. The diagnosis of different responsibility groups Chi-square test was used to compare diagnosis among three groups, and there was significant difference across groups in diagnosis (x2 = 863.80, p < 0.01), the severity of the mental disorder was certainly related to the responsibility level (see Table 1). 3.3. The score of RSCR among different groups One way analysis of variance showed that there was also significant difference in the scores among groups (F = 2864.0,
Table 3 The statistic scores in different diagnostic groups. Diagnosis
Responbility
n
Mean
Standard deviation
F
Schizophrenia and paranoid disorder
Irresponbility Diminished responbility Whole responbility
381 118 48
8.77 24.92 40.77
6.21 8.36 6.71
635.43
Affective disorders
Irresponbility Diminished responbility Whole responbility
15 42 17
11.20 25.67 44.06
7.43 6.21 9.72
80.44
Organic and mental retardation
Irresponbility Diminished responbility Whole responbility
14 63 59
11.36 29.25 39.02
5.62 6.44 9.02
80.52
Others
Irresponbility Diminished responbility Whole responbility
38 53 339
10.00 27.00 45.00
6.48 7.23 6.02
645.75
W. Cai et al. / Forensic Science International 236 (2014) 146–150
148
Table 4 The statistic scores in different criminological groups. Criminal type
Responbility
n
Mean
Standard deviation
F
Homicide
Irresponbility Diminished responbility Whole responbility
161 59 102
7.00 26.00 43.00
5.79 7.16 7.26
890.59
Harm
Irresponbility Diminished responbility Whole responbility
167 83 97
7.00 26.00 42.00
5.65 8.17 6.27
850.66
Robbery or extortion
Irresponbility Diminished responbility Whole responbility
46 37 92
13.00 27.00 47.00
9.03 8.70 6.14
299.07
Rape
Irresponbility Diminished responbility Whole responbility
4 17 28
12.50 28.00 43.00
8.42 5.51 8.01
36.71
Arson
Irresponbility Diminished responbility Whole responbility
26 13 9
8.50 26.00 45.00
4.86 3.41 10.54
118.51
Theft
Irresponbility Diminished responbility Whole responbility
15 33 72
9.00 29.00 46.00
4.42 7.20 6.87
196.67
Swindle and others
Irresponbility Diminished responbility Whole responbility
29 34 63
10.00 26.00 44.00
5.10 7.16 7.19
259.30
Table 5 The Pearson related coefficient of items in different groups. Items
Mean
Standard deviation
Correlation to total score
p
Realistic motivation Aura before offense Inducement of crime Time selectivity of crime Place selectivity of crime Object selectivity of crime Tool selectivity of crime Emotion during the crime Shirking responsibility after offense Concealing the truth during the inquest Camouflage Understanding the nature of the behavior Estimating the consequence of the offense Impairment of life ability Impairment of learning or work Impairment of insight Impairment of reality testing Impairment of self-control
1.69 0.84 1.55 1.28 1.28 1.45 1.30 1.83 1.01 0.69 0.43 2.09 2.09 2.13 1.79 0.98 2.17 2.01
1.25 0.83 1.20 1.07 1.03 1.10 1.02 1.51 1.05 0.97 0.81 1.57 1.54 0.98 1.06 0.90 1.58 1.50
0.89 0.65 0.82 0.84 0.84 0.83 0.81 0.66 0.75 0.66 0.50 0.84 0.86 0.74 0.78 0.86 0.88 0.88
Contents lists available at ScienceDirect
Forensic Science International journal homepage: www.elsevier.com/locate/forsciint
The reliability and validity of the rating scale of criminal responsibility for mentally disordered offenders Weixiong Cai *, Qingting Zhang, Fuyin Huang, Wei Guan, Tao Tang, Chao Liu Shanghai Key laboratory of Forensic Medicine, Institute of Forensic Science, Ministry of Justice, Shanghai 200063, PR China
A R T I C L E I N F O
A B S T R A C T
Article history: Received 6 July 2011 Received in revised form 2 April 2013 Accepted 17 December 2013 Available online 13 January 2014
In China, the criminal responsibility of the mentally disordered offenders is divided into three levels, there are the whole responsibility, diminished responsibility and irresponsibility. According to the Criminal Law, ‘‘If a mental disordered patient causes harmful consequences at a time when he is unable to recognize or control his own conduct, upon verification and confirmation through legal procedure, he shall not bear criminal responsibility.’’ That means there are two standards of assessing criminal responsibility, namely volitional and cognitive capacity. It is as equal as the Mc’Naughton Rule and the Irresistible Impulse Test. But for a long time, the criminal responsibility was assessed mainly by experience because of lacking of standardized assessment instrument. Recently, we have developed ‘‘the rating scale of criminal responsibility for mentally disordered offenders (RSCRs)’’. The scale includes eighteen items, namely criminal motivation, aura before offense, inducement of crime, time and place and object and tool selectivity of crime, emotion during the crime, shirking responsibility after offense, concealing the truth during inquest, camouflage, understanding the nature of the offense, estimating the consequence of the offense, impairment of life ability, impairment of learning or work, impairment of insight, impairment of reality testing, and impairment of self-control. This scale can be applicable for all cases and easy to use. This scale had been tried out in several forensic psychiatry institutes, the Cronbach a of the scale is 0.93, and all items have high correlation with the total score of the scale (r = 0.50–0.89). Two factors were extracted by the factorial analysis, and the cumulative squared loading was 68.62%. The scores of the three levels were 9.66 5.11, 26.54 5.21 and 40.08 7.90 respectively and highly significant differences were observed among groups. By establishing discrimination analysis among three levels, classification results suggested that 88.90% of the original grouped cases were correctly classified, and the discriminant value had high conformity with the experts’ opinions. The data showed that the scale would be the best validated instrument for the criminal responsibility in China. ß 2013 Elsevier Ireland Ltd. All rights reserved.
Keywords: Forensic psychiatry Criminal responsibility Rating scale Mentally disordered offender RSCR
1. Introduction The identification of criminal responsibility in China remains controversial, fueled by public antagonism and professional dissension. During the last several decades, many forensic psychiatrists were tempted to rely on empirical evaluation and traditional measures on which they were trained and gained years of intensive experience, the objectivity and impartiality were doubted [1,2]. Which measure is the most effective for the assessment of criminal responsibility? In the United States, the Rogers criminal responsibility assessment scales (R-CRAS) had been applied to the insanity defense [3–6]. However, the lack of standard assessment tools exists all the while in the forensic psychiatry field in China. Recently, we have developed the
* Corresponding author. Tel.: +86 21 52359576; fax: +86 21 52359576. E-mail address: [email protected] (W. Cai). 0379-0738/$ – see front matter ß 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.forsciint.2013.12.018
assessment measure named ‘‘the rating scale of criminal responsibility for mentally disordered offenders (RSCRs)’’. The purpose of the present study was to examine the reliability and validity of this rating scale in four hospitals and one institute from different provinces in China. 2. Method and object 2.1. Object 1187 objects were recruited from the Institute of Forensic Science, Ministry of Justice, the Second Xiangya Hospital of Central South University, the Mental Health Center of Shanghai, the Psychosis Hospital of Guangzhou and Nanjing Brain Hospital from January 2007 to May 2009. Among these objects, 891 were male and 296 were female, mean age was 35.52 years (SD = 10.635, range 14–78). The educational level: 19 were illiterate, 378 were primary school, 494 were junior high school, 203 were high school,
W. Cai et al. / Forensic Science International 236 (2014) 146–150
147
Table 1 The diagnosis type in different groups (number, %). Group
Schizophrenia and paranoid disorder
Affective disorder
Organic and mental retardation
Others
Irresponsibility (n = 448) Diminished responsibility (n = 276) Whole responsibility (n = 463)
381(85.0) 118(42.8) 48(10.4)
15(3.4) 42(15.2) 17(3.7)
14(3.1) 63(22.8) 59(12.7)
38(8.5) 53(19.2) 339(73.2)
Table 2 The statistic scores in different groups. Group
Mean
Standard deviation
Standard error
Irresponsibility (n = 448) Diminished responsibility (n = 276) Whole responsibility (n = 463)
9.12 26.33 43.73
6.29 7.59 7.03
0.30 0.46 0.33
and 93 were college or above. Marriage situation: 318 were married, 572 were unmarried, 92 were divorced, and 205 were unknown. Occupation: 276 were farmer, 97 were worker, 67 were official, 481 were unemployed, 22 were self-employed, 68 were others and 186 were no details. The criminal type: 322 were homicide, 347 were harm, 175 were robbery or extortion, 49 were rape, 48 were arson, 120 were theft, and 126 were swindle and other crimes. Diagnosis: 482 were schizophrenia, 74 were affective disorder, 96 were mental retardation, 40 were organic mental disorder, 65 were paranoid mental disorder, 8 were neurosis, 6 were stress related disorders, 35 were personality disorder, 12 were substance-dependent mental disorder, 267 were other mental disorder and 102 were without psychosis.
mental disorder was mild, and the capacity of cognition or volition was intact; diminished responsibility: suffered from mental disorder and the capacity of cognition or volition was impaired due to the psychiatric symptoms; irresponsibility: suffered from mental disorder and the capacity of cognition or volition was lost completely due to the psychiatric symptoms. Double-blind principle was complied in the research, i.e. the raters used the scale independently and did not take part in the discussion to the object so it is impossible for him/her to know the experts’ opinion to the object’s criminal responsibility, on other hand, the forensic psychiatrists did not know the rating result. The data were analyzed using the SPSS 16.0 software package. 3. Results
2.2. Method 3.1. The interrater reliability of the scale The RSCR was developed on the base of the Scale for Diminished Criminal Responsibility, the Rogers Criminal Responsibility Scale (R-CRAS) [4,6,7]. The scale consist of 18 items, namely criminal motivation, aura before offense, inducement of crime, time selectivity of crime, place selectivity of crime, object selectivity of crime, tool selectivity of crime, emotion during the crime, shirking responsibility after offense, concealing the truth during inquest, camouflage, understanding the nature of the behavior, estimating the consequence of the offense, impairment of life ability, Impairment of learning or work, impairment of insight, impairment of testing-reality, impairment of self-control. The point for 18 items is from 0 to 4. The diagnosis and criminal responsibility level of the object was also made by professional groups, which were made up by more than three forensic psychiatrists according to CCMD-3 (Chinese Classification of Mental Disorders, the third revision) and the criteria: whole responsibility: no psychosis or the severity of
10 cases were assessed respectively by six observers, there was a high inter-rater reliability among all items (kappa = 0.785, p < 0.001), which shows that the scale is stable. 3.2. The diagnosis of different responsibility groups Chi-square test was used to compare diagnosis among three groups, and there was significant difference across groups in diagnosis (x2 = 863.80, p < 0.01), the severity of the mental disorder was certainly related to the responsibility level (see Table 1). 3.3. The score of RSCR among different groups One way analysis of variance showed that there was also significant difference in the scores among groups (F = 2864.0,
Table 3 The statistic scores in different diagnostic groups. Diagnosis
Responbility
n
Mean
Standard deviation
F
Schizophrenia and paranoid disorder
Irresponbility Diminished responbility Whole responbility
381 118 48
8.77 24.92 40.77
6.21 8.36 6.71
635.43
Affective disorders
Irresponbility Diminished responbility Whole responbility
15 42 17
11.20 25.67 44.06
7.43 6.21 9.72
80.44
Organic and mental retardation
Irresponbility Diminished responbility Whole responbility
14 63 59
11.36 29.25 39.02
5.62 6.44 9.02
80.52
Others
Irresponbility Diminished responbility Whole responbility
38 53 339
10.00 27.00 45.00
6.48 7.23 6.02
645.75
W. Cai et al. / Forensic Science International 236 (2014) 146–150
148
Table 4 The statistic scores in different criminological groups. Criminal type
Responbility
n
Mean
Standard deviation
F
Homicide
Irresponbility Diminished responbility Whole responbility
161 59 102
7.00 26.00 43.00
5.79 7.16 7.26
890.59
Harm
Irresponbility Diminished responbility Whole responbility
167 83 97
7.00 26.00 42.00
5.65 8.17 6.27
850.66
Robbery or extortion
Irresponbility Diminished responbility Whole responbility
46 37 92
13.00 27.00 47.00
9.03 8.70 6.14
299.07
Rape
Irresponbility Diminished responbility Whole responbility
4 17 28
12.50 28.00 43.00
8.42 5.51 8.01
36.71
Arson
Irresponbility Diminished responbility Whole responbility
26 13 9
8.50 26.00 45.00
4.86 3.41 10.54
118.51
Theft
Irresponbility Diminished responbility Whole responbility
15 33 72
9.00 29.00 46.00
4.42 7.20 6.87
196.67
Swindle and others
Irresponbility Diminished responbility Whole responbility
29 34 63
10.00 26.00 44.00
5.10 7.16 7.19
259.30
Table 5 The Pearson related coefficient of items in different groups. Items
Mean
Standard deviation
Correlation to total score
p
Realistic motivation Aura before offense Inducement of crime Time selectivity of crime Place selectivity of crime Object selectivity of crime Tool selectivity of crime Emotion during the crime Shirking responsibility after offense Concealing the truth during the inquest Camouflage Understanding the nature of the behavior Estimating the consequence of the offense Impairment of life ability Impairment of learning or work Impairment of insight Impairment of reality testing Impairment of self-control
1.69 0.84 1.55 1.28 1.28 1.45 1.30 1.83 1.01 0.69 0.43 2.09 2.09 2.13 1.79 0.98 2.17 2.01
1.25 0.83 1.20 1.07 1.03 1.10 1.02 1.51 1.05 0.97 0.81 1.57 1.54 0.98 1.06 0.90 1.58 1.50
0.89 0.65 0.82 0.84 0.84 0.83 0.81 0.66 0.75 0.66 0.50 0.84 0.86 0.74 0.78 0.86 0.88 0.88
