Throughout human history a phenomenon has been observed which is currently receiving major relevance which is extreme violence against women that is more frequent. Recently, this crime has been called femicide. Diana Russell y Van de Ven¹, who were pioneers in research on sexual violence, incest, rape, and femicide among other female related issues, define femicide as "the murder of women by men because they are women"¹. In Latin America, Lagarde² describes femicide as those historical conditions that promote socially aggressive practices, which harm the integrity, development, health, independence and life of women. Since the understanding that femicide is used specifically to identify the deaths of women in certain conditions, femicide has been categorized as a crime³. Mexico was one of the first countries in Latin America that created a law that categorizes and criminalizes femicide as a crime that is a product of femicide violence. Since July 14 2012, article 325 chapter V of the Federal Penal Code of Mexico has established femicide among the crimes against life and bodily integrity. Those who commit femicide are "those who take the life of a woman for gender reasons". These gender reasons include: sexual violence; history of family, work or school violence; a relationship of affection, romance or trust between the victimizer and victim; isolation of the victim; and exposing or putting on display the body of the victim in a public place.

Statistics worldwide are far from providing precise data about femicide and there are few organizations in different countries that have a real record of this crime. The World Health Organization (WHO) approved a global statistical framework to measure femicides. The Director of United Nation Organization (UN) Women, Sima Bahous, explains that the lack of data is a major obstacle for the fight against femicide, "since it is possible to act better against what can be measured"⁴.

Despite the fact that there is no statistical data recorded at a global level, in countries such as the United Kingdom the figures from 2009 to 2018 show that a woman was murdered by a man every three days, while in high income countries such as Slovenia, New Zealand, Switzerland, and Austria the female homicide rate is greater than or equal to the male homicide rate⁵. In 2017, the highest number of femicides was observed in Asia, followed by Africa, America, Europe and Oceania⁵. In Latin America, Honduras, the Dominican Republic, El Salvador, Ecuador, Costa Rica, and Panama were the countries that registered the highest number of femicides per 100,000 women in 2020. In 2021 in Mexico, more than three thousand women lost their lives; while 922 were femicide cases, 2,540 were victims of other crimes. The states within Mexico with the highest rate of femicide were the State of Mexico (132), Jalisco and Veracruz (66), Mexico City (64), and Nuevo Leon (57)⁶.

Femicide is a problem that has different variables and theoretically has been associated with some psychosocial factors such as personality disorders⁷, psychopathy⁸, cognitive distortions⁹, suicidal ideation¹⁰, violence¹¹ and criminality¹². The main studies on femicide focus on two areas, the first being the classification types originated by the contribution of Holtzworth-Munroe and Stuart¹³, who developed a theoretical typology identifying three descriptive dimensions based on the severity of domestic violence, the generality of the violence (towards the wife of towards others), and the psychopathology/personality disorders. This results in three subtypes of aggressors: only in the family; dysphoric/borderline; and generally violent/antisocial.

The subtype of only in the family has the characteristics of men who participate in domestic violence, which is less severe, and they are less likely to suffer from psychological and sexual abuse throughout their lifetime¹³. The violence of the men in this subtype is generally restricted to the family. These men do not participate in acts of violence outside of the home and they do not have legal problems. They demonstrate little psychopathology and do not have personality disorders or passive-dependent personality.

The subtype of dysphoric/borderline abusers are abusive towards their partners, which ranges from moderate to severe, including psychological and sexual forms. The violence is limited mainly to the family, but is also evident towards those outside of the family, and these individuals engage in criminal behavior¹³. These men are psychologically anxious and emotionally volatile. They can demonstrate characteristics of borderline personality disorder and schizoid personality disorder, and they can have problems with alcohol and substance abuse.

The third subtype of violent/antisocial aggressors engage in marital violence that ranges from moderate to severe, including psychological and sexual abuse. They also engage in aggression towards those outside of the family and have a more extensive history of criminal conduct. It is likely that they have problems with alcohol and substance abuse and it is more likely that they have antisocial personality or psychopathy¹³.

Following this line of thought, Dixón et al.¹², through their files analyses based on the classification of Holtzworth-Munroe y Stuart¹³ focused on constructing a classification system of men convicted for murdering their partners through the use of two variables: criminality and psychopathology. Three classifications were generated: delinquents with low criminality and low psychopathology; high criminality and low-moderate psychopathology; and low-moderate criminality and high psychopathology. The group classified as low criminality and low psychopathology has similarities with the subtype initially called only in the family in that the characteristics of low criminality indicate not having a history of intimate partner violence as well as not having psychopathology related to this. In contrast, the group classified as low-moderate criminality/high psychopathology is characterized by moderate to high criminality, by the illicit use of drugs at the moment of the crime (drug abuse), and these delinquents are more likely to react in interpersonal disputes¹².

The high psychopathology has a history of depressive and suicidal behavior (depression/suicide), suicide attempts, suicide after committing the homicide, high levels of anger through excessive hitting, high levels of dependency, preoccupied with being watched, and reacting to discussions with anger and violence. Lastly, the third group classified as high criminality/low-moderate psychopathology is comprised of delinquents who have been tried at an early age, are unemployed, have been in prison at an early age, and have been convicted for violent crimes towards those outside of the family. The psychopathology dimension indicates that the offenders classified in this group have low levels of psychopathology and engage in substance abuse¹².

The second area of study on femicide focuses on the criminality levels because criminality is an important variable which allows for the characterization of those who commit femicide through an empirical classification of men who are violent towards their partner¹². Criminality is a risk factor which can help to predict a fatal outcome in an intimate partner relationship¹⁴. In fact, Matías et al.¹⁵, have indicated that these risk factors are specific and play an important role in femicide.

Criminality has been indicated by extensive penal sentences, sentences for violence towards those outside of the family, first sentences for crime at an earlier age, and arrest for any type of crime. Men who commit femicide show a higher risk if they have made previous threats of murder, non- fatal strangulations, rape, spying, and harassment prior to the femicide, and easy access to firearms¹⁴. Matías et al.¹⁵, indicates that easy access to firearms is the risk factor that is most associated to femicide and differentiates between lethal and non-lethal aggressors. Buteau et al.¹⁶, add a history of alcohol and substance abuse as a risk factor associated with femicide. While Aguilar adds a history of addiction as a risk factor, Matías et al.¹⁵, mentions alcohol use as a risk factor and as a precipitating factor in which the perpetrator commits the crime under the influence of alcohol and/or drugs as well as having access to a firearm. For these reason the present study was aimed at understanding the indicators and criminality level on those sentenced for the crime of femicide in Mexico.

# Type of study

The present work is a qualitative study because it focuses on determining the strength of the association or correlation between variables, the generalization and objectivity of the results through a sample in order to make inference to a general population¹⁷. Acording to Sampieri et al¹⁸: using an exploratory level, "exploratory studies are carried out usually when the objective is to examine a theme or research problem little studied or has not been studied at all". It is a transversal type because the data is collected at a single time and it is aimed at describing variables and analyzing their behavior at a specific time¹⁸.

# Population and sample

The population incarcerated for the crime of femicide in Mexico State at the time of this study was 82 people who were sentenced. A non-probabilistic sample of intentional type with 62 volunteers, in three social rehabilitation centers in the State of Mexico, Mexico was used.

# Inclusion criteria

People sentenced for the crime of femicide who could read, write and accept voluntarily to participate in the study.

# Exclusion criteria

The exclusion criteria were: have been processed; have committed a different crime, being unable to read or write.

# Instrument

A check list based on the study by Dixon et al.¹² was used in the current study, in which the following indicators were evaluated as being present or absent: sentences after the age of 28; sentences before the age of 16; recidivism; sentences for violence unrelated to family; instrumental sentence; juvenile incarceration; unemployment; history of intimate partner violence; history of violence in other relationships, suicide attempt after the crime; order of protection; harassment of the partner; history of suicide attempt; frequent arguments with one"s partner; history of substance abuse; history of alcohol use; committing the crime under the influence of drugs; committing the crime under the influence of alcohol; admission of the crime. This last variable was included in this research.

# Procedure

Once approval by the General Office of the Prevention and Social Reintegration of the State of Mexico was obtained, the Office of the Department of Psychology and Criminology of each readjustment center organized the documents for the security director for each of the prevention centers in order to provide access to the researchers.

Then, the participants were found and informed of the purpose of the study and their willingness to participate was ascertained through informed consent. Psychologists and Directors of each center assigned a place for the applications, and participants were assisted individually. The response time varied depending on the education level and attitude of each participant.

# Data analysis

The data was analyzed through descriptive statistics in order to identify the average of each criminality indicator and to calculate the quartiles for each participant. In the first quartile, low criminality is explained by specific indicators and a low response of these and then increasing the gravity in each quartile until reaching the very high criminality, where the most indicators existed.

# Ethical considerations

The study followed both national and international ethical guidelines²⁰ for human studies, taking into account central elements such as just treatment, the physical and emotional integrity of participants, the anonymity and confidentiality of their responses through informed consent. The study had the approval of the Committee of Research Ethics of the Autonoma University from the State of Mexico, with registry number DOOPSIC-0221 as well as the approval of the General Office of Prevention and Social Reintegration of the State of Mexico.

The characteristics of the participants were between the ages of 25 and 64, with 24-40 being the most frequent ages. With respect to relationship to the victim, in the majority of the cases it was the wife, although there were times when it was the girlfriend, lover, acquaintance, relative or stranger. The level of education of participants ranged from no education to university. From the 62 participants who were evaluated, 34 had children with the victim and the number of children varied from one to four, with the most frequent being having at least one child (Table 1).

In reference to the years of incarceration, it varied from 1 to 40 years, being the most frequent being 1 to 20 years, with some participants being incarcerated for many years. Although, they were sentenced for homicide because it was against the partner according to current laws it is considered femicide (Table 2).

Femicide is considered a grave crime³ according to the Federal Penal Code of Mexico. Consequently, the sentences of the participants include a wide range such as 12 years and up to 220 years as its maximum sentence if you take into account the aggravating factors of the crime. The most frequent sentences ranged from 21 to 80 years. Lastly, the most frequent type of homicide was intimate partner femicide (Table 2).

The data that describes some of the characteristics of the participants were obtained thanks to the cooperation of the psychology area of each center in order to corroborate the information given when conversing with each participant.

With respect to criminality, after calculating the average of each indicator and the quartiles for each score, the results indicate four levels which correspond to low, medium, high, and very high criminality, during which it was observed that those incarcerated for the crime of femicide were in medium and high levels of criminality (Table 2).

The indicators which identify the differences between the levels of criminality include: instrumental benefit of the crime, history of violence towards one"s partner, harassment of one"s partner, arguing with one"s partner during the crime, substance use, alcohol use, and admission of the crime (Table 3).

The results of the current study coincide with those of Dixón et al.¹², y Holtzworth-Munroe y Stuart¹³ in that criminality plays an important role in the biography of those who commit femicide. The criminality indicators which were present in the participants of this study include: history of previous incarcerations; violence with non-relatives; violence and harassment of one"s partner; substance and alcohol use; and instrumental gains from committing the crime.

The concurrence of these indicators facilitated the establishing of criminality levels identified as "very high", which have not been reported previously. The majority of participants fell into medium and high levels and taking into account that 70% of participants committed femicide against their intimate partner, which contradicts previous findings where femicide delinquents were classified as "only family" or femicide of intimate partners showed low criminality levels.

Based on these results, the population incarcerated for the crime of femicide almost falls into the classifications of "borderline or dysphoria" and "generally violent"¹³, who present with moderate and high levels of criminality, respectively, with the two last subtypes being a great risk factor for committing femicide.

According to Dawson y Dawson²¹ and Santos²², within the criminality indicators that mark the differences in levels, they propose the dimension identified as violence, which includes a history of intimate partner violence, threats of homicide against one"s partner, isolation of the victim, and escalation of violence. This concurs with the result found in the present study where participants showed a history of violence, harassment, and that the crime took place during an argument with one"s partner, with this last indicator not being reported previously in other countries. Nevertheless, this can be explained by certain norms and cultural beliefs, which influence the type of violence committed by the aggressor against the partner in different countries. This can be interpreted as a deficient situational response due to the lack of effective strategies in conflict resolution.

Another relevant indicator is substance and alcohol use, which can play a double role since these can be both the facilitators of violence and precipitants of femicide.

The results of this study agree with those reported by Buteau et al.¹⁶, Aguilar⁷ and Matías et al.¹⁵, since the participants had a history of substance use and had committed the crime while under the influence of drugs and alcohol. Mathews et al.²³, believes that these men often use their substance use as an excuse to act violently towards their partner and even commit the crime.

The indicator of obtaining personal gain from the crime such as through money or prohibiting the victim from informing authorities of illegal activities or simply removing the victim from one"s life because she is an obstacle for other plans¹² was found in this study. This indicator discredits substance use as an excuse to commit the crime since this indicator implies a certain degree of conscious planning. Lastly, admission of the crime showed small differences between two levels, which could be explained through the levels of criminality as can be expected with high and very high levels of criminality in that these individuals have experience with denying their crimes, creating alibis, and creating an appearance of innocence.

It can be highlighted that men who commit the crime of femicide are a population rarely studied, who theoretically present homogeneity and various possibilities in classification. Thus, the criminality arises as a category of analysis, which tries to explain certain characteristics in people who have committed this crime. Nevertheless, it is necessary to consider other factors such as culture, social structure, individual characteristics of the imprisoned population.

In reference to the indicators that mark the differences between the levels of criminality, a history of violence towards one"s partner to resolve conflict is one of these. In addition, a history of alcohol and substance use can be found. It is these same indicators that makes one thing that not all indicators carry the same weight in risk factors for committing femicide, especially a history of intimate partner violence, which requires a more exhaustive study as and explanation of this phenomenon.

The main limitation of this study is related to the sample used since it is small and specific to a single federal entity, which has no national representation. Despite this, it is important to highlight that few studies such as the present one, have gathered information on criminality with individuals incarcerated for the crime of femicide, involving the perpetrator in a direct manner, since the majority of studies base their analysis on penitentiary documents which require multiple studies to identify this phenomenon.

Studies such as the present one are required to identify the factors of criminality, which are involved in femicide that can be used to prevent an episode of lethal violence as well as develop programs of reintegration for exconvicts.