Evidence of validity and reliability of the Desiderative Questionnaire with juvenile offenders and victimized adolescents

Evidências de validade e fidedignidade do Questionário Desiderativo com adolescentes infratores e vitimizados

Abstract

The objective of this work was to assess validity and reliability of the Desiderative Questionnaire with adolescent offenders and victims. The sample comprised 200 adolescents, of both genders, between 12 and 16 years of age, of whom 50 were sentenced to correctional measures, 50 were victims of domestic violence and 100 without suspected violence or transgression. Data analysis was performed using the Pearson Chi-Square Test, Kappa Concordance Index and Cronbach’s Alpha. The results showed statistically significant differences between the clinical and control groups in several categories, with more difficulties among adolescent victims and offenders. They also revealed precision of the technique used and of the evaluation system with high correlations among the judges as well as high internal consistency, with adequate reliability coefficients.

Keywords
Adolescence; Psychological assessment; Violence

Resumo

O objetivo deste trabalho foi realizar estudos de validade e de fidedignidade do Questionário Desiderativo com adolescentes infratores e vitimizados. Compuseram a amostra 200 adolescentes, de ambos os sexos, entre 12 e 16 anos, sendo 50 cumprindo medida socioeducativa, 50 vítimas de violência doméstica e 100 sem suspeitas de sofrerem violência ou terem praticado ato infracional. A análise dos dados foi realizada por meio do Teste Qui-Quadrado de Pearson, Índice de Concordância Kappa e Alfa de Cronbach. Os resultados mostraram que o teste diferenciou de forma estatisticamente significante os grupos clínicos e de controle em várias categorias, identificando mais dificuldades entre os adolescentes vitimizados e infratores, com menor força de ego. Também revelaram precisão da técnica e do sistema de avaliação com elevadas correlações entre os juízes e alta consistência interna, com adequados coeficientes de fidedignidade.

Palavras-chave
Adolescência; Avaliação psicológica; Violência

Violence is defined by the World Health Organization (WHO) as “[…] the intentional use of physical force or power threatened or actual, against oneself, another person or against a group or community, that either results or has a high likelihood of resulting in injury, death , psychic harm, maldevelopment or deprivations” (World Health Organization, 2002). This complex and multicausal phenomenon has involved adolescence and today is a cause of concern. Adolescents exposed to violence, whether as victims or offenders (authors of violence), are subjected to some form of exclusion and/or suffering, in addition to exhibiting personality aspects that need to be understood (Pinto Junior et al., 2020; Trabbold et al., 2016).

Victims, specifically of domestic violence, experience abusive and oppressive situations (Pinto Junior et al., 2015; Pinto Junior et al., 2020). As offenders, youngsters are prevented from experiencing their citizenship, through which they can acknowledge themselves and be recognized as subjects of rights and duties (Brambilla et al., 2015; Tardivo, & Moraes, 2016).

Studies show the existence of a positive correlation between the experience of domestic victimization in childhood and the manifestation of violent and/or disruptive behavior in adolescence. Being a victim of domestic violence and later becoming an aggressor or offender can represent two sides of the same coin, that is, the adolescent’s offending behavior can be a consequence of harm caused by the experience of domestic victimization (Baena-Vallejo et al., 2020; Paula & Assumpção, 2013; Tardivo et al., 2016).

When addressing child-juvenile domestic violence, in its different modalities (physical, sexual, psychological and resulting from negligence), it appears that it is currently considered a serious public health problem by the WHO, revealing itself as one of the main causes of morbidity and mortality in this age group (Malta et al., 2017; Martins & Romagnoli, 2017). Investigations in the area show that the experience of domestic victimization of children and adolescents is responsible for a lot of harm to the physical and psychological health of victims (Nemeroff, 2016; Tardivo, Pinto Junior et al., 2019; Thibodeau et al., 2015; van der Kolk, 2017).

Likewise, research on violence committed by adolescents in an offense situation demonstrates that there is a confluence of several factors associated with the etiology of the adolescent’s offending behavior, from restricted access to basic social goods and services and exposure to poverty (Pereira et al., 2015; Silva & Milani, 2015; Tardivo & Moraes, 2016), to the experience of a family environment with inadequate maternal and paternal care, exposure to marital violence, or being a victim of aggressive and violent parents (Paula & Assumpção, 2013; Pinto Junior, Tardivo, & Cassepp-Borges, 2017; Silva & Milani, 2015), all factors that enhance the development of disruptive, psychopathological and violent behaviors in adolescents.

But, in addition to these etiological factors, psychodynamic and symbolic characteristics linked to such practices are also observed (Pedro & Neves, 2015; Tardivo et al., 2018). In view of the psychic system difficulties in elaborating past serious emotional traumas such as victimization experience in childhood, the transition to violent acts may occur in adolescence. So, the infraction, in a psychodynamic perspective, appears as an attempt to protect narcissism and identity. In other words, the adolescent’s infraction acts as an attempt to symbolize the external reality of what cannot be internally elaborated, that is, a way to obtain relief from the anguish caused by the traumatic drive excitation (Pedro & Neves, 2015).

Specifically about the task of describing the psychodynamics of cases of child/adolescent victimization and of adolescents in conflict with the law, the psychoanalytical psychodiagnostic process, with all its resources, has shown to be very promising for this purpose. Subordinated to clinical thinking for the apprehension of psychic dynamics and the understanding of the individual’s problems, the psychoanalytical psychodiagnostic process uses the results of psychological instruments, such as clinical interviews and, mainly, projective techniques, as a dynamic work that is structured as a function of emergent and significant aspects of the clinical situation (Pinto Junior & Tardivo, 2017).

Thus, as indicated by Pinto Junior and Tardivo (2017), there is a need, especially in Brazil, to invest in research aiming at the construction and validation of instruments that facilitate the identification and psychodynamic understanding of individuals exposed to violence. It should be noted that projective techniques are an important resource in this type of psychological assessment (Tardivo & Moraes, 2016). These instruments are characterized by the presentation of a relatively unstructured task, asking the subject to interpret or structure the stimulus presented, based on the projection mechanism, revealing fundamental aspects of their psychological functioning (Grassano, 1997; Tardivo & Moraes, 2016).

The precursor of the Desiderative Questionnaire (QD), which is the subject of this article, was the Desiderative Test, developed in the 1940s by psychiatrists Cordoba and Pigem (1946) from Barcelona. In this proposal, the participant should be asked the question: If you couldn’t be a person, what would you want to be? The respondent would then explain his/her choice. The test was later re-elaborated by Bernstein (1973) and developed by Nijamkim and Braude (2000), with a dynamic criterion, modifying it with an expansion of the instructions in order to obtain in all cases a richer and more complete material, with six questions focusing on the respondent’s choices and rejections. It also aimed at expanding the foundation, integrating it with the psychoanalytic focus, deepening the analysis and interpretation, considering more the symbolic aspects, the sequence of responses (associations) and the gestalt overall (Nijamkim & Braude, 2000).

The way the subject captures the framework and elaborates his/her answers provides data about the personality structure, making it possible to analyze the test using several theoretical references, the most used being the Freudian and Kleinian approach. Along these lines, it allows us to become aware of the degree of ego strength, a construct also derived from psychoanalytic theory (Guimarães et al., 2008; Nijamkim & Braude, 2000; Pinto Junior & Tardivo, 2015; Pinto Junior et al., 2018).

The QD is studied and applied especially in Latin American countries. In a systematic literature review, Pinto Junior et al. (2018) describe the scientific production during the period between 2001 and 2017, all in Spanish and Portuguese, mainly from Argentina and Brazil. The article presents the studies, their objectives and the different designs and contexts of the research. Validity studies with this instrument were not found in this survey, especially in Brazil, nor investigations with adolescents in situations of domestic violence.

In view of the above, this study aimed to investigate the clinical validity and reliability of the analysis proposed by the authors for the QD with Juvenile Offenders (JO) and victims of Domestic Violence (DV) in Brazil, seeking, also, to present the response patterns of this population to the test. With these studies, we expect to bring relevant scientific contribution in the area of psychological assessment and collaborate to the certification of the Federal Council of Psychology for the professional use of this projective technique in Brazil.

Method

Participants

The sample consisted of 200 participants between 12 and 16 years of age, assisted in social assistance and education institutions in Southern Fluminense and São Paulo State, divided into three groups: (a) Group composed of 50 Juvenile Offenders (JO), of both genders, serving a correctional measure for transgressions; (b) Group composed of 50 adolescents who are proven victims of Domestic Violence (DV), of both genders; and (c) Control Group, composed of 100 adolescents, of both genders, without suspicion of suffering or having suffered domestic violence or having committed an infraction. Table 1 shows the distribution of participants by gender, age and education, with the aim of presenting the characteristics of each group with regard to these categories.

Table 1
Distribution of participants by gender, age and education

Characteristic		Control			Juvenile Offender			Domestic Violence		p-value
		M	SD		M	SD		M	SD
Total		100	100.0		50	100.0		50	100.0
Age
	Group 12	12	12.0		0	0.0		8	16.0	0.0002*
	Group 13	13	13.0		2	4.0		13	26.0	(χ2 = 30.32)
	Group 14	13	13.0		4	8.0		7	14.0
	Group 15	18	18.0		6	12.0		8	16.0
	Group 16	44	44.0		38	76.0		14	28.0
Gender
	Female	54	54.0		10	20.0		27	54.0	0.0002*
	Male	46	46.0		40	80.0		23	46.0	(χ2 = 17.48)
Education		9.58	1.65		7.84	2.48		6.84	2.06	< 0.001* (F = 34.67)

• Note:
• ^*
  Statistically significant.

Instrument and Procedures

After approval of the project by the Research Ethics Committee of the School of Medicine of the Universidade Federal Fluminense (FM/UFF/HU) under opinion nº 1.425.412, and signature of the Consent and Child Consent Forms, the application of the QD to each individual participating in one of the three research groups was carried out. For each Positive or Negative answer, it was initially identified what was answered by the participant and what was called symbol in this text. After tabulating the 200 questionnaires, the symbols were categorized and grouped. Thus, both for the application and for the analysis, the proposal of the original manual by Nijamkim and Braude (2000) was followed. The theoretical reference used was the psychoanalytic as described by the authors and by the studies of Pinto Junior et al. (2021).

QD Responses Analyses

A form of analysis of the QD was proposed by the authors, based on that presented by Nijamkim and Braude (2000); Pinto Junior and Tardivo (2015); Pinto Junior et al. (2018). The criteria of presence (1) and absence (0) were adopted for the general categories, which refer to the formal aspects of the responses to the QD, which are: complete test: if the participant responded to all kingdoms; perseveration: the subject presents more than one answer in one or more kingdoms in positive or negative cathexes; anthropomorphic response: elements in which human identity remains present, such as: superman, ghost, angels.

Regarding the specific categories for positive and negative cathexes, the following were evaluated: Reaction time: refers to the lapse of time between the probing (questions asked in order to obtain choices or rejections) and the appearance of the participants’ response. Appropriateness to the probing: refers to the participants’ ability to perform the test, that is, to imagine themselves temporarily as non-humans, which may be adequate, inadequate or by induction. Choice of kingdom: the positive cathexes of the test represent values that the subject wants to keep, while the negative cathexes refer to undesirable values, attributes and characteristics, that is, what the subject wants to reject (Pinto Junior et al., 2021). Response quality (of choice and rationalization): each symbol and referred rationalization must be evaluated in terms of its quality, and can be conventional (common), original (creative and adequate) or bizarre (inappropriate or weird). Dissociation: ability to discriminate the valued aspects from the undervalued aspects; it indicates the possibility of recognizing situations that generate anxiety and identify the resources to control it, which may be adequate or inadequate. Projective identification: mechanism by which the ego deposits an aspect of itself in a symbol with which it identifies himself/herself, and in the QD, this mechanism is recognized through the subject’s ability to choose a properly structured and consistent symbol, which may be structured or unstructured. Rationalization: individuals justify their choice and can do it within the formal logic (Nijamkin & Braude, 2000), and may be adequate or inadequate. Linking perspective of the response: refers to the type of responses to the QD, determined by the subject’s ability or not to establish a bond or relationship with the other, being narcissistic or including the other (Pinto Junior & Tardivo, 2015).

In order to be able to carry out comparisons between the results of the three groups with respect to the construct “ego strength” an index was created, with scores being assigned according to the level of elaboration of the responses produced, based on psychoanalytic theory (Guimarães et al., 2008; Nijamkim & Braude, 2000; Pinto Junior & Tardivo, 2015; Pinto Junior et al., 2018). In this analysis proposal, the more evolved the answer, the higher the score the individual receives, considering the items: complete test, kingdom perseverance, anthropomorphic response, adequacy of the probing/issue (adequate or inadequate), originality, bizarreness or degree of response convention, instrumental defenses (dissociation, projective identification, rationalization: adequate or inadequate), binding perspective (narcissistic or other-involving response). The score in this system can vary from zero to 56 points for the set of answered cathexes, allowing the identification of the ego’s strength level and the classification of its intensity into different levels, as shown in Table 2.

Table 2
Ego strength score for desiderative quiz

Categories	Responses	Score	Responses	Score	Responses	Scores
Full Test Maximum: 1 point	Yes - 1	1
Will Persevere Maximum: 2 points	0	1	1	0
Anthropological Response Maximum: 2 points	0	1	1	0
Adequate to Probing Maximum: 12 points	Adequate	2	Inadequate	0	By induction	1
Response quality Maximum: 9 points	Conventional	1	Original	1,5	Bizarre	0
Dissociation Maximum: 12 points	Adequate	2	Inadequate	0
Projective Identification Maximum: 6 points	Struct. Symbol	1	N. Struct. Symbol	0
Rationalization Maximum: 6 points	Adequate	1	Inadequate	0
Perspective link answer Maximum: 6 points	Narcissistic Response	0	Resp. includes the other	1

• Note: Maximum Total Ego Strength: 56 points.

Reliability study: agreement among Judges in the assessment of analysis categories and Internal Consistency

Thirty protocols were drawn and sent for blind analysis to three PhD professors, experts in the field of psychological assessment and projective techniques. The agreement indices (percentage and Kappa) were calculated considering the three evaluators jointly, that is, each question was considered concordant when the three evaluators reported exactly the same option. Percentage agreements above 70% were considered good results and the Kappa Coefficients of Agreement (k) were considered acceptable from 0.70, according to the Dancey and Reidy (2013) guidelines.

After evaluating the protocols of the 200 participants, the analysis of the internal consistency of the QD analysis categories was also carried out, using the Cronbach’s Alpha indicator, considering the categories of the QD attribute type in the calculation. The calculation of Cronbach’s alpha was performed for the qualitative questions, dichotomizing each one of them, that is, transforming the appropriate/inappropriate type response into values 0 or 1, since this (Cronbach’s alpha) is suitable for both quantitative values and for ordinal or dichotomized qualitative variables, as in the present study.

Comparison of categories between study groups

To assess the existence or not of statistically significant differences between the clinical and Control groups, the means and standard deviations were presented for the Response time measure (quantitative characteristic), and the differences were tested by the Analysis of Variance test followed by Multiple comparisons by the Tukey’s method.

For the other QD categories (all of them characteristics of the attribute type), the frequencies and percentages to the responses to each option were presented and the Pearson’s Chi-square test was used, in order to verify the statistical significance of the percentage differences between the clinical and Control groups. The percentages of choice for each symbol were evaluated in each of the Kingdoms and Positive/Negative responses separately, using the Fisher’s Exact Test to verify the statistical significance of percentage differences between the clinical and Control groups. For all tests, the significance level (α) equal to or less than 0.05 was considered.

Results

The reliability study, carried out through the agreement index, revealed a total of 87% of agreement among the three evaluators regarding the QD categories, thus showing an adequate level of agreement. The analysis of the Kappa index indicated a value of 0.8403, considered adequate agreement between the evaluators. Table 3 shows the results for each of the categories in relation to the questionnaire as a whole and all alpha values were adequate.

Table 3
Cronbach’s alpha for desiderative analysis categories

Categories	Cronbach’s Alpha
	Positive	Negative	1+	2+	3+	1-	2-	3-
Perseveration	0.903	0.903	-	-	-	-	-	-
Anthropomorphic Response	0.903	0.902	-	-	-	-	-	-
Probing Adequacy	-	-	0.902	0.902	0.903	0.904	0.901	0.901
Response Quality	-	-	0.902	0.898	0.901	0.898	0.897	0.894
Dissociation	-	-	0.902	0.898	0.899	0.901	0.900	0.897
Projective Identification	-	-	0.903	0.898	0.899	0.898	0.899	0.894
Rationalization	-	-	0.902	0.901	0.899	0.901	0.897	0.897
Link perspective of the answer	-	-	0.903	0.898	0.895	0.898	0.897	0.893

Table 4 and Table 5 show only the categories in which statistically significant differences were observed. Table 4 shows the distributions of the three groups in the categories referring to General Aspects and Table 5 presents the results of the QD content categories. The data showed that the test statistically differentiated the clinical and control groups in a significant manner. In the analysis of Ego Strength in the QD, the JO group presented results of lower Ego strength (mean of 37.36 points), followed by the DV Group (mean of 39.97 points), and finally by the Control Group (mean of 44.15 points).

Table 4
Distribution of participants in the General Aspects of Desiderative Questionnaire categories

Categories	Choice/Answer		Group			p-value
			C	JO	DV	C x JO	C x DV
Complete Test	No		3 (3.0%)	12 (24.0%)	2 (4.0%)	< 0.0001*	0.7477
	Yes		97 (97.0%)	38 (76.0%)	48 (96.0%)	(χ2 = 16.33)	(χ2 = 0.10)
Perseveration	Negative category		18 (18.0%)	6 (12.0%)	2 (4.0%)	0.0730	0.1272
	Positive category		13 (13.0%)	7 (14.0%)	7 (14.0%)	(χ2 = 6.97)	(χ2 = 5.70)
	Positive. and Negative category		19 (19.0%)	19 (38.0%)	11 (22.0%)
	None		50 (50.0%)	18 (36.0%)	30 (60.0%)
Anthropomorphic Response	Negative category		5 (5.0%)	7 (14.0%)	4 (8.0%)	0.0146*	0.5024
	Positive category		8 (8.0%)	6 (12.0%)	7 (14.0%)	(χ2 = 10.53)	(χ2 = 2.35)
	Positive. and Negative category		1 (1.0%)	4 (8.0%)	1 (2.0%)
	None		86 (86.0%)	33 (66.0%)	38 (76.0%)
Reaction Time	Kingdom Animal		8.4±8.0	15.1±18.3	12.6±16.6	0.0135 (t = 2.85)*	0.1723 (t = 1.80)
	Kingdom Inanimate		12.1±10.2	23.3±29.7	14.7±15.2	0.0015 (t = 3.53)*	0.6797 (t = 0.84)
Probing Adequacy	Kingdom	Level	Group			p-value
			C	JO	DV	C x JO	C x DV
Positive	Animal	Adequate	94 (94.0%)	36 (72.0%)	41 (82.0%)	0.0006*	0.0209*
		Inadequate	0 (0.0%)	2 (4.0%)	0 (0.0%)	(χ2 = 14.86)	(χ2 = 5.33)
		By induction	6 (6.0%)	12 (24.0%)	9 (18.0%)
	Inanimate	Adequate	78 (78.8%)	23 (47.9%)	20 (40.0%)	0.0005*	< 0.0001*
		Inadequate	0 (0.0%)	1 (2.1%)	0 (0.0%)	(χ2 = 15.30)	(χ2 = 22.20)
		By induction	21 (21.2%)	24 (50.0%)	30 (60.0%)
	Plant	Adequate	39 (39.4%)	10 (21.7%)	9 (18.0%)	0.0451*	0.0083*
		Inadequate	0 (0.0%)	1 (2.2%)	0 (0.0%)	(χ2 = 6.20)	(χ2 = 6.96)
		By induction	60 (60.6%)	35 (76.1%)	41 (82.0%)
Negative	Inanimate	Adequate	90 (89.1%)	27 (57.4%)	29 (58.0%)	< 0.0001*	< 0.0001*
		Inadequate	0 (0.0%)	0 (0.0%)	0 (0.0%)	(χ2 = 19.42)	(χ2 = 19.34)
		By induction	11 (10.9%)	20 (42.6%)	21 (42.0%)
Sequence of the Kingdoms	Kingdoms		Group			p-value
			C	JO	DV	C x JO	C x DV
Positive Sequence	Animal/Inanimate/Plant		51 (52.0%)	18 (40.9%)	25 (50.0%)	0.0421*	0.8707
	Animal/Plant/Inanimate		32 (32.7%)	25 (56.8%)	18 (36.0%)	(χ2 = 9.90)	(χ2 = 1.25)
	Inanimate/Animal/Plant		10 (10.2%)	1 (2.3%)	5 (10.0%)
	Plant/Animal/Inanimate		3 (3.1%)	0 (0.0%)	2 (4.0%)
	Plant/Inanimate/Animal		2 (2.0%)	0 (0.0%)	0 (0.0%)

• Note:
• ^*
  Statistically significant. C: Control Group; DV: Victims of Domestic Violence; JO: Juvenile Offenders.

Table 5
Distribution of participants in the Desiderative Questionnaire Content categories

Response Quality			Group			p-value
Cathexis	Kingdom	Quality	C	JO	DV	C x JO	C x DV
Positive	Animal	Bizarre	1 (1.0%)	3 (6.1%)	4 (8.0%)	0.0265*	0.0044*
		Conventional	83 (83.0%)	44 (89.8%)	45 (90.0%)	(χ2 = 7.26)	(χ2 = 10.86)
		Original	16 (16.0%)	2 (4.1%)	1 (2.0%)
	Inanimate	Bizarre	0 (0.0%)	5 (10.4%)	1 (2.0%)	0.0014*	0.1949
		Conventional	55 (55.6%)	30 (62.5%)	32 (64.0%)	(χ2 = 13.10)	(χ2 = 3.27)
		Original	44 (44.4%)	13 (27.1%)	17 (34.0%)
Negative	Inanimate	Bizarre	0 (0.0%)	3 (6.4%)	1 (2.0%)	0.0170*	0.0059**
		Conventional	70 (69.3%)	35 (74.5%)	22 (44.0%)	(χ2 = 8.15)	(χ2 = 10.27)
		Original	31 (30.7%)	9 (19.1%)	27 (54.0%)
	Plant	Bizarre	0 (0.0%)	3 (6.7%)	0 (0.0%)	0.0258*	0.1651
		Conventional	82 (84.5%)	33 (73.3%)	36 (75.0%)	(χ2 = 7.32)	(χ2 = 1.93)
		Original	15 (15.5%)	9 (20.0%)	12 (25.0%)
Dissociation	Kingdom	Quality	C	JO	DV	C x JO	C x DV
Positive	Animal	Adequate	96 (96.0%)	39 (78.0%)	41 (82.0%)	0.0005* (χ2 = 12.00)	0.0041* (χ2 = 8.26)
Negative	Animal	Adequate	78 (78.0%)	39 (79.6%)	28 (57.1%)	0.8241 (χ2 = 0.05)	0.0083* (χ2 = 6.97)
	Inanimate	Adequate	68 (67.3%)	28 (59.6%)	15 (30.0%)	0.3577 (χ2 = 0.85)	< 0.0001* (χ2 = 18.82)
Rationalization	Kingdom	Quality	C	JO	DV	C x JO	C x DV
Positive	Animal	Adequate	82 (82.0%)	34 (68.0%)	30 (60.0%)	0.0535 (χ2 = 3.73)	0.0035* (χ2 = 8.53)
	Plant	Adequate	51 (51.5%)	34 (73.9%)	26 (52.0%)	0.0108* (χ2 = 6.50)	0.9554 (χ2 = 0.003)
Negative	Inanimate	Adequate	57 (56.4%)	22 (46.8%)	14 (28.0%)	0.2744 (χ2 = 1.20)	0.0010* (χ2 = 10.86)
Response perspective	Kingdom	Quality	C	JO	DV	C x JO	C x DV
Positive	Animal	Narcissistic	63 (63.0%)	47 (94.0%)	34 (68.0%)	< 0.0001*	0.5459
		Include another	37 (37.0%)	3 (6.0%)	16 (32.0%)	(χ2 = 16.38)	(χ2 = 0.37)
	Inanimate	Narcissistic	43 (43.4%)	35 (72.9%)	18 (36.0%)	0.0008*	0.3835
		Include another	56 (56.6%)	13 (27.1%)	32 (64.0%)	(χ2 = 11.28)	(χ2 = 0.76)
	Plant	Narcissistic	51 (51.5%)	31 (70.5%)	25 (50.0%)	0.0346*	0.8613
		Include another	48 (48.5%)	13 (29.5%)	25 (50.0%)	(χ2 = 4.47)	(χ2 = 0.03)
Negative	Animal	Narcissistic	39 (39.0%)	42 (85.7%)	18 (36.7%)	< 0.0001*	0.7892
		Include another	61 (61.0%)	7 (14.3%)	31 (63.3%)	(χ2 = 28.93)	(χ2 = 0.07)
	Inanimate	Narcissistic	22 (21.8%)	37 (78.7%)	16 (32.0%)	< 0.0001*	0.1733
		Include another	79 (78.2%)	10 (21.3%)	34 (68.0%)	(χ2 = 43.38)	(χ2 = 1.85)
	Plant	Narcissistic	53 (54.6%)	38 (82.6%)	29 (60.4%)	0.0012*	0.5090
		Include another	44 (45.4%)	8 (17.4%)	19 (39.6%)	(χ2 = 10.55)	(χ2 = 0.44)

• Note:
• ^*
  Statistically significant. C: Control Group; DV: Victims of Domestic Violence; JO: Juvenile Offenders.

Discussion

Regarding the sociodemographic characteristics, in terms of age, the Control and DV groups can be seen with very similar percentages. The more balanced distribution between different age groups, especially in the DV group, may be related to the fact highlighted by research on domestic violence in childhood and adolescence about the characteristics of victims in terms of age; as it is configured in a phenomenon that affects children and adolescents of different age groups, revealing itself as a “virally democratic” social problem (Malta et al., 2017; Pinto Junior et al., 2015). On the other hand, it appears that the JO group has greater differences, with the highest concentration in the 16-year-old group, with a significant difference. These data corroborate the findings in the literature which indicate that the majority of teenagers undergoing correctional measures is mostly concentrated in the age group between 15 and 17 years old (Paula & Assumpção, 2013; Silva & Milani, 2015).

The analysis of the gender category shows the Control and DV groups quite similar and the JO group with a greater number of male participants, with a significant difference (p-value = 0.0002). This characterization also seems to confirm the results of studies on the profile of adolescent offenders, which indicate that most of them are male (Paula & Assumpção, 2013; Silva & Milani, 2015). In contrast, the literature indicated that the distribution of victims of domestic violence in terms of gender is more equal (Malta et al., 2017; Pinto Junior et al., 2015).

As for education, participants in the Control group had more years of education than those in the clinical groups. These data may be related to research results with adolescents in conflict with the law, which indicate that the educational level of these young people is always identified as inferior, with frequent school dropouts (Pinto Junior et al., 2017; Silva & Milani, 2015). Likewise, victims of domestic violence commonly show learning difficulties, delay and drop out of school and some studies also highlight that the presence of severe learning difficulties can be considered, along with other signs and symptoms, an important indicator of domestic victimization (Pinto Junior et al., 2015; Tardivo et al., 2019).

With regard to reliability, as pointed out in the description of the results, the data revealed an adequate level of agreement between the evaluators for this evaluation model for the test. Kappa index analysis also revealed adequate agreement among raters. We can conclude that the evaluation model presented here for the QD is reliable, as this is one of the characteristics necessary for a psychological test to be considered reliable in its use.

Regarding the analysis of the general aspects of the QD categories, in the Complete test category, the Control and DV groups showed quite similar results, but the JO group exhibited a higher percentage of non-completed tests, and the difference was significant when compared to the Control group. It can be suggested that the greater number of incomplete tests in the JO group is associated with a fragility of the ego, resulting from a difficulty in discriminating the reality from fantasy (Nijamkim & Braude, 2000).

With regard to the anthropomorphic response category, the Control Group and the DV Group are very similar in rates. However, in the JO group, differences are found, with indicators of difficulties in dissociating from human identity, evidencing failure to respond to the test and suggesting dysfunction in the level of adequacy to reality (Guimarães & Pasian, 2009; Guimarães-Eboli & Pasian, 2020) and ego fragility.

As to the data on reaction time, it was found that, in general, the mean times of the JO group were greater than those of the Control and DV, but being significant only for positive/animal and positive/inanimate. These data confirm those obtained by Nijamkim and Braude (2000) and those obtained by Guimarães and Pasian (2009), with an average time between 10 and 30 seconds. However, new studies should be carried out so that it is possible to interpret the results with the Brazilian population, in other age groups.

Regarding the adequacy of the probing, in the Positive/Animal, Positive/Plant and Positive/Inanimate cathexes, the vast majority of subjects answered the test adequately, highlighting the Control group that scored the highest. It can be seen that the JO and DV groups had a slightly higher percentage of responses by induction, and these groups showed significantly different results from those of the Control group. According to Nijamkim and Braude (2000), the need for induction in the QD is related to the individual’s difficulties in properly following the test instructions and, projectively, with the problems in the psychological functions related to the adaptation to reality.

When analyzing the sequence of choices, it can be seen that in the positive ones there was a higher prevalence of Animal/Inanimate/Plant for the Control and DV groups. In contrast, for the JO group there was a predominance of the Animal/Plant/Inanimate sequence, with more than 80% of the responses, with significant differences in relation to the Control group. According to Nijamkim and Braude (2000), the sequence of kingdoms in the QD is related to the expectation of a healthy personality in which the conservation instinct predominates over the death drive. The authors argue that the expected sequence, then, for positive cathexes should be: Animal/Plant/Inanimate. It is noteworthy that Bunchaft and Vasconcellos (2001), in a study on the standardization of QD among university students in Brazil, also found a sequence different from that suggested by Nijamkim and Braude (2000).

Regarding the QD content categories, the data on the quality of the response show that, in the Positive/Animal and Positive/Inanimate cathexes, the vast majority of the categories were classified as conventional, but highlighting the Control Group with a slightly higher percentage of responses of the original type. The original type responses, which are related to creativity, intuition, cognitive and emotional flexibility, were found mostly in the Control group and afterwards in the DV group. In general, these data point to the predominance of responses of the conventional type, i.e., the one that most subjects respond to and is related to the adequacy to reality. These results corroborate the findings of the study by Bunchaft and Vasconcellos (2001).

When checking the quality of the responses to the test, it was found that the JO and DV groups showed very similar and significantly different results compared the Control group, with more bizarre responses among the JO. Such responses are very different (uncommon), with markedly idiosyncratic inappropriate or strange associations, being consistent with psychopathological conditions, according to Grassano (1997). A tendency for the JO group to exhibit more psychopathological indicators is suggested.

Regarding the data on dissociation, these show that, in the Positive/Animal cathexis, the larger majority of responses were considered adequate, with emphasis on the Control group with a slightly higher percentage. Participants in the JO and DV group had similar and significantly different results from those in the Control group. In the Negative/Animal cathexis, it appears that most were evaluated as adequate for the three groups, but highlighting the DV group with a slightly lower percentage and significantly different from the Control group, which is also observed in the Negative/Inanimate cathexis. Thus, the DV group stands out with the highest percentages of inadequate responses regarding dissociation, which, according to Nijamkin and Braude (2000), concerns the subject’s ability to discriminate the valued from the devalued aspects and indicates the possibility of recognizing situations that generate anxiety and identify the resources to control it.

With regard to projective identification, that is, the degree of structuring of the chosen symbol (Nijamkin & Braude, 2000), the data, in general, denoted very few unstructured responses. Thus, according to the results, it was found that the three research groups did not reveal difficulties in issuing symbol responses with structure and consistency.

The data analysis on rationalization shows that, in Positive/Animal cathexis, there are differences in the responses of participants in the clinical groups when compared to the Control group, which exhibited a higher percentage of adequate responses, followed by the JO group and then the DV group, with lower percentage. In the Positive/Plant cathexis, the Control and DV groups presented very similar results, but the JO group exhibits a higher percentage of adequate rationalization, being significantly different from the Control group. Regarding the Negative/Inanimate cathexis, the Control group also exhibited a higher percentage of adequate responses, followed by the JO group and then the DV group, with a lower result. As rationalization is the procedure by which the subject tries to offer a coherent explanation from a logical or socially accepted point of view for an attitude, act, idea or feeling (Guimarães-Eboli & Pasian, 2020; Nijamkin & Braude, 2000), it is clear that, in general, a good rationalization capacity prevailed in the three groups, showing adequacy of thought to the shared reality, despite the traumatizing antecedents of exclusion and/or suffering experienced, especially by the juvenile offenders and victimized adolescents.

From the data on the bonding perspective of the response, a category that refers to the person’s ability to establish a bond with the other or turn to oneself, it appears that a higher percentage of narcissistic response is found in the JO group, being significantly different from the control group. Thus, JO have greater difficulty in establishing a bond with each other, acting based on narcissistic motivations. These data are similar to the results of Peker and Rosenfeld (2013) in the Argentine framework, who highlighted that the majority of juvenile offenders in their sample (56%) showed difficulty in establishing a bond with each other, and that narcissistic performance seems to be a characteristic psychodynamics of adolescents in conflict with the law, which facilitates their involvement in infractions and violence.

The analysis of symbols evoked in the QD was also performed (by the Fisher test). For the Animal Kingdom, it was found that the highest frequencies were for birds (including small bird, eagle, etc.), pets (with a prevalence of dog and cat) and wild animals (lion, tiger, jaguar, etc.) in the positive responses. In the negative responses, the following stood out: poisonous animals (snake, scorpion, spider, etc.), domestic animal (especially dog and cat) and insect (bee, ant, etc.). These data corroborate the results of Guimarães and Pasian (2009) who also found birds and pets as the most evoked symbols in the positive animal category and insects and domestic animals in the negative animal category. It should be noted that the indication of a domestic animal as a rejected symbol in this investigation was mainly given by the participants of the clinical groups, and may be linked to failures in dissociation, since they are considered positive due to their characteristics of affection with the human being.

In the analysis of the symbols listed for the Plant Kingdom, the highest frequencies refer to trees, vegetables and flowers in the positive cathexes, and legumes, vegetables and trees in the negative ones. The same results were found in the research by Guimarães and Pasian (2009). In the analysis of symbols for the Inanimate Kingdom, it was noted that the number of categories was much higher, as in the research by Guimarães and Pasian (2009). The highest frequencies were for electronic equipment and means of transport in the positive responses and school supplies, garbage and weapons in the negative.

In the analysis of the Ego Strength, understood as the degree of structuring of the Self and the quality of its functions, especially the defenses against the symbolic attack represented by the probing of the QD (Nijamkim & Braude, 2000), there are differences between the groups. In general, the JO group showed results associated with a personality with less Ego strength, followed by the DV Group, with a moderate Ego strength personality, and finally by the Control Group, which showed results associated with a higher Ego strength, compared to the other groups. These data show that the QD is a sensitive instrument to discriminate the three research groups, finding theoretical support in the specialized literature in the area of violence and victimization, especially with regard to the consequences and mental health problems in adolescents exposed to or involved in situations of violence (Nemeroff, 2016; Tardivo & Moraes, 2016; Tardivo et al., 2019; Thibodeau et al., 2015; van der Kolk, 2017).

Conclusion

The objectives of the study were addressed and, using the different comparisons made, the findings pointed out to performance specifics in different groups of adolescents and in different categories of analysis of the QD, which suggests evidence of clinical validity. The reliability of the interpretive categories of the QD and the evaluation system used was also confirmed, based on the study of agreement between evaluators. Likewise, through the analysis of internal consistency, reliability coefficients were found for the instrument.

The study showed the response patterns of adolescent victims, offenders and adolescents without these clinical conditions for all cathexes and kingdoms assessed by the instrument. Clinical groups denoted greater difficulty in establishing bonds and lower ego strength.

The research described in this paper brings contributions to the field of Psychodiagnostics, especially for adolescents in situations of violence and psychological distress. QD, a technique little known in Brazil, presented adequate results in this investigation in terms of reliability and clinical validity. It is also evident that more studies with other clinical groups and ages should be carried out in order to obtain approval by the Federal Council of Psychology for the use of this technique by psychologists in their professional activities with the Brazilian population.

• Article drawn from the post-doctoral report by A. A. PINTO JUNIOR, entitled “O Questionário Desiderativo em adolescentes vítimas de violência doméstica e infratores”. Universidade de São Paulo, 2018.

How to cite this article

• Tardivo, L. S. L. P. C., Pinto Junior, A. A., Rosa, H. R., Belizario, G. O., & Medeiros, D. (2022). Evidence of validity and reliability of the Desiderative Questionnaire with juvenile offenders and victimized adolescents. Estudos de Psicologia (Campinas), 39, e200021. https://doi.org/10.1590/1982-0275202239e200021

References

• Baena-Vallejo, G. A., Carmona-Otálvaro, J. G., & Rengifo-Arias, C. G. (2020). Propuesta de intervención sobre la violencia intrafamiliar: abordaje de acuerdo con la función y sentido del fenómeno violento presente en la dinámica familiar. Estudos de Psicologia (Campinas), 37, e180104. https://doi.org/10.1590/1982-0275202037e180104
  » https://doi.org/10.1590/1982-0275202037e180104
• Bernstein, J. (1973). Análisis e interpreteción del Cuestionario Desiderativo Facultad de Psicología de la Universidad de Buenos Aires.
• Brambilla, B. B., Cardoso, J. T., Capelo, D. F., & Toledo, R. (2015). Psicologia e políticas públicas: uma leitura histórica sobre violências, invisibilidades e juventudes em conflito com a lei. Revista Ibirapuera, 10, 49-56. http://www.revistaunib.com.br/vol10/06.pdf
  » http://www.revistaunib.com.br/vol10/06.pdf
• Bunchaft, G., & Vasconcellos, V. L. P. (2001). Padronização do teste desiderativo no contexto da análise transacional: resultados preliminares. Psicologia: Teoria e Pesquisa, 17(1), 19-25. http://www.scielo.br/pdf/%0D/ptp/v17n1/5402.pdf
  » http://www.scielo.br/pdf/%0D/ptp/v17n1/5402.pdf
• Córdoba, J., & Pigem, J. M. (1946). La expressión desiderativa como manifestación de la personalidad. Medicina Clínica, 4(3), 20-23.
• Dancey, C. P., & Reidy, J. (2013). Estatística sem matemática para psicologia Penso.
• Grassano, E. (1997). Indicadores psicopatológicos nas técnicas projetivas Casa do Psicólogo.
• Guimarães, N. M., & Pasian, S. R. (2009). Adequação ao real de adolescentes: possibilidades informativas do questionário desiderativo. Psicologia: Teoria e Pesquisa, 25(3), 347-355. http://dx.doi.org/10.1590/S0102-37722009000300008
  » https://doi.org/10.1590/S0102-37722009000300008
• Guimarães, N. M., Pasian, S. R., & Loureiro, S. R. (2008). O Questionário Desiderativo: possibilidades teóricas e empíricas na atualidade. In A. E. Villemor-Amaral & B. S. G. Werlang (Orgs.), Atualizações em métodos projetivos para avaliação psicológica (1th ed., pp. 391-414). Casa do Psicólogo.
• Guimarães-Eboli, N. M., & Pasian, S. R. (2020). Evidências psicométricas do questionário desiderativo em adultos. Avaliação Psicológica, 19(2), 179-188. https://dx.doi.org/10.15689/ap.2020.1902.08
  » https://doi.org/10.15689/ap.2020.1902.08
• Malta, D. C., Bernal, R. T. I., Pugedo, F. S. F., Lima, C. M., Mascarenhas, M. D. M., Jorge, A. D. O., & Melo, E. M. D. (2017). Violências contra adolescentes nas capitais brasileiras, segundo inquérito em serviços de urgência. Ciência & Saúde Coletiva, 22(9), 2899-2908. https://dx.doi.org/10.1590/1413-81232017229.14212017
  » https://doi.org/10.1590/1413-81232017229.14212017
• Martins, F. F. D. S., & Romagnoli, R. C. (2017). A violência contra as crianças e adolescentes admitidos no Hospital João XXIII: uma análise quantitativa. Gerais: Revista Interinstitucional de Psicologia, 10(2), 148-161. http://pepsic.bvsalud.org/pdf/gerais/v10n2/02.pdf
  » http://pepsic.bvsalud.org/pdf/gerais/v10n2/02.pdf
• Nemeroff, C. B. (2016). Paradise lost: the neurobiological and clinical consequences of child abuse and neglect. Neuron, 89(5), 892-909. https://dx.doi.org/10.1016/j.neuron.2016.01.019
  » https://doi.org/10.1016/j.neuron.2016.01.019
• Nijamkim, G., & Braude, M. (2000). Questionário desiderativo Vetor.
• Paula, M. D. L. B., & Assumpção, F. B. Jr. (2013). Delinquência juvenil e família. Revista Psicopedagogia, 30(91), 43-51. http://pepsic.bvsalud.org/pdf/psicoped/v30n91/06.pdf
  » http://pepsic.bvsalud.org/pdf/psicoped/v30n91/06.pdf
• Pedro, J. C., & Neves, A. S. (2015). A família e o adolescente em conflito com a lei: vínculos, estratégias e recursos afetivos. Horizonte Cientifico, 9(1), p.1-27. http://www.seer.ufu.br/index.php/horizontecientifico/article/view/22491/15714
  » http://www.seer.ufu.br/index.php/horizontecientifico/article/view/22491/15714
• Peker, G. M., & Rosenfeld, N. G. (2013). La ausencia de lo vincular en los adolescentes disociales. Anuario de Investigaciones, 20(1), 395-401. http://www.scielo.org.ar/pdf/anuinv/v20n1/v20n1a42.pdf
  » http://www.scielo.org.ar/pdf/anuinv/v20n1/v20n1a42.pdf
• Pereira, T. C. S., Reis, J. N., & Costa, L. A. (2015). Autor e vítima: a vulnerabilidade social de jovens que cometeram atos infracionais em Belo Horizonte. Revista de Terapia Ocupacional da Universidade de São Paulo, 26(2), 258-266. https://doi.org/10.11606/issn.2238-6149.v26i2p258-266
  » https://doi.org/10.11606/issn.2238-6149.v26i2p258-266
• Pinto, A. A. Jr., & Tardivo, L. S. L. P. C. (2017). Escala de Exposição à Violência Doméstica (Coleção EEVD). Vetor.
• Pinto Junior, A. A., & Tardivo, L. S. L. P. C. (2015). Estudio del funcionamiento psicodinamico de agresores sexuales con el Cuestionario Desiderativo. Subjetividad y Procesos Cognitivos, 19(2), 187-207. http://dspace.uces.edu.ar:8180/xmlui/bitstream/handle/123456789/3461/Estudio_Pinto_Cury-Tardivo.pdf?sequence=1
  » http://dspace.uces.edu.ar:8180/xmlui/bitstream/handle/123456789/3461/Estudio_Pinto_Cury-Tardivo.pdf?sequence=1
• Pinto Junior, A. A., Cassepp-Borges, V., & Santos, J. G. (2015). Caracterização da violência doméstica contra crianças e adolescentes e as estratégias interventivas em um município do Estado do Rio de Janeiro, Brasil. Cadernos de Saúde Coletiva, 23(2), 124-131. https://dx.doi.org/10.1590/1414-462X201500020062
  » https://doi.org/10.1590/1414-462X201500020062
• Pinto Junior, A. A., Rosa, H. R., Chaves, G., & Tardivo, L. S. L. P. C. (2018). O Questionário Desiderativo: fundamentos psicanalíticos e revisão da literatura. Arquivos Brasileiros de Psicologia, 70, 274-287. http://pepsic.bvsalud.org/scielo.php?script=sci_arttext&pid=S1809-52672018000300019&lng=pt&tlng=pt
  » http://pepsic.bvsalud.org/scielo.php?script=sci_arttext&pid=S1809-52672018000300019&lng=pt&tlng=pt
• Pinto Junior, A. A., Tardivo, L. S. L. P. C., & Medeiros, D. (2021). Características de personalidade de adolescentes em conflito com a lei: estudo com o Questionário Desiderativo. In E. M. Ferreira (Org.), Psicologia: identidade profissional e compromisso social 2 (1st ed., pp. 26-42). Atena, https://dx.doi.org/10.22533/at.ed.4002129034
  » https://doi.org/10.22533/at.ed.4002129034
• Pinto Junior, A. A., Tardivo, L. S. L. P. C., & Cassepp-Borges, V. (2017). La escala de exposición del niño y adolescente a la violencia doméstica en Brasil: adaptación y validez. Subjetividad y Procesos Cognitivos, 21(1), 105-121. http://dspace.uces.edu.ar:8180/xmlui/bitstream/handle/123456789/3781/La%20escala_Pinto-Junior_Plata-Cury_yotros.pdf?sequence=1
  » http://dspace.uces.edu.ar:8180/xmlui/bitstream/handle/123456789/3781/La%20escala_Pinto-Junior_Plata-Cury_yotros.pdf?sequence=1
• Pinto Junior, A. A., Tardivo, L. S. L. P. C., Rosa, H. R., Chaves, G., Belizario, G. O. (2020). Traços de personalidade de adolescentes infratores e vitimizados por meio do Eysenck Personality Questionnaire Junior (EPQ-J). Mudanças (IMS), 27, 9-14. https://www.metodista.br/revistas/revistasmetodista/index.php/MUD/article/view/9979/7118
  » https://www.metodista.br/revistas/revistasmetodista/index.php/MUD/article/view/9979/7118
• Silva, C. Y. G. D., & Milani, R. G. (2015). Adolescência e tendência antissocial: o rap como expressão de uma privação emocional. Psicologia: Ciência e Profissão, 35(2), 374-388. https://dx.doi.org/10.1590/1982-370301572013
  » https://doi.org/10.1590/1982-370301572013
• Tardivo, L. S. L. P. C., Althanat, M., Miura, P. O., & Pinto, A. A. Jr. (2016). Family violence against children and adolescents in Brazil and relations with identity development. Acta Psychopathologica, 2(4), 1-5. https://doi.org/4172/2469-6676.100061
  » https://doi.org/4172/2469-6676.100061
• Tardivo, L. S. L. P. C., Avoglia, H. R. C., Pinto A. A. Jr., & Rosa, H. R. (2018). Rupturas na adolescência e sociedade: concepções sobre ser jovem por infratores. In C. K. Sakamoto & R. C. F. A. Giora (Orgs.), Rupturas: olhares criativos e novos tecidos conceituais (pp. 375-385). Gênio Criador.
• Tardivo, L. S. L. P. C., Pinto, A. A. Jr., Rosa, H. R., Moraes, M. C. V., Vagostello, L., & Miura, P. O. (2019). Depression in Brazilian children and adolescent victims of intrafamilial abuse: a discussion about gender. In S. A. Xu (Org.), Violence exposure: perspectives, gender differences and outcomes (pp. 129-156). Nova Science Publishers.
• Tardivo, L. S. L. P. C., & Moraes, M. C. V. (2016). The personality of juvenile offenders: psychodynamic aspects in Children´s Apperception Test Human Figures. International Journal of Psychological and Brain Sciences, 1(1), 9-12. http://article.sciencepublishinggroup.com/html/10.11648.j.ijpbs.20160101.12.html
  » http://article.sciencepublishinggroup.com/html/10.11648.j.ijpbs.20160101.12.html
• Thibodeau, E. L., Cicchetti, D., & Rogosch, F. A. (2015). Child maltreatment, impulsivity, and antisocial behavior in African American children: moderation effects from a cumulative dopaminergic gene index. Development and Psychopathology, 27(4), 1621-1636. https://dx.doi.org/10.1017/S095457941500098X
  » https://doi.org/10.1017/S095457941500098X
• Trabbold, V. L. M., Caleiro, R. C. L., Freitas Cunha, C., & Guerra, A. M. C. (2016). Concepções sobre adolescentes em situação de violência sexual. Psicologia & Sociedade, 28(1), 74-83. https://doi.org/10.1590/1807-03102015v28n1p074
  » https://doi.org/10.1590/1807-03102015v28n1p074
• van der Kolk, B. A. (2017). Developmental trauma disorder: toward a rational diagnosis for children with complex trauma histories. Psychiatricannals, 35(5), 401-408. https://dx.doi.org/10.3928/00485713-20050501-06
  » https://doi.org/10.3928/00485713-20050501-06
• World Health Organization. (2002). Informe mundial sobre la violencia y salud https://www.cevs.rs.gov.br/upload/arquivos/201706/14142032-relatorio-mundial-sobre-violencia-e-saude.pdf
  » https://www.cevs.rs.gov.br/upload/arquivos/201706/14142032-relatorio-mundial-sobre-violencia-e-saude.pdf

Publication Dates

• Publication in this collection
  28 Oct 2022
• Date of issue
  2022

History

• Received
  07 Feb 2020
• Reviewed
  11 June 2021
• Accepted
  18 Oct 2021