Development and Validation of Attitude towards Honour Killing Scale

Received 26 May, 2017

Accepted 10 Jan, 2018

Published 30 Jun, 2018

The present study aimed to develop a scale that measures attitude towards honour killing. The scale initially contained 21 items, which were reduced to 10 scenarios after qualitative item analysis. For quantitative item analysis two phases of Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) were planned. The sample for exploratory factor analysis comprised of 326 individuals (men = 190 and women = 136) from general population, honour killers, and murderers in general. A factor loading of .40 was set as a selection criterion for an item to be retained in the scale. Using oblique rotation in EFA, all the items were retained proposing one factor solution. Another similar sample of 374 individuals (men = 198 and women = 176) was selected for Confirmatory Factor Analysis (CFA) which confirmed the factor structure proposed in EFA. According to the findings, men and women in general population had a negative attitude towards honour killing; if the cause of the honour killing was adultery then general population slightly favored honour killing, whereas, honour killers followed by the ordinary murderers had a highly positive attitude towards honour killing.

Every civilization believes in the sacredness of a human life. Despite the worldwide growth in the recognition of human rights, there still are individuals who believe in taking the life of another individual to save their pride. Honour killing is a part of general crime in different societies which needs to be understood, that is, what makes people think justified in taking the life of their loved ones (Jafri, 2008).

Honour killing is a crime usually committed by close family members of the victim if they come to know that victim did something which was against the norms of a society (Jafri, 2008). What makes honour killing different than other kinds of murder is that a family member, usually a father, brother or husband takes the life of a female (daughter, sister, wife etc) to protect the family’s honour, or to restore the bad name associated with the family. It is an effort to wipe off the dark spots on the honour of the family with the blood of the accused. According to a report published by Human Rights Group (2011), 720 honour killings incidents took the lives of 605 women and 115 men. While the report of Pakistan Human Rights Commission states that 791 honour killings took place in the year 2010 alone (Karimjee, 2011). Roughly 4000 honour killing incidents have taken place in Pakistan from 1998 to 2004 in which 2700 women and 1300 men lost their lives. The reported cases of honour killing tried in the courtrooms are 3451. The highest number of honour killing incidents took place in Punjab followed by Sindh province. In KPK 94 women were murdered in the incidents of honour killings in the year 2017 (Felix, 2004). No authentic data is available in other countries. According to the official data 1100 women were killed in 2015 in honour killings incidents which indicate that the crime of honour killings is on the rise, however there is no compatibility and consistency between different reports. Human Rights Commission of Pakistan has reported 280 murders from October 2016 to June 2017.This report is not accurate because it has not taken into account the unreported cases (Pakistan Honour Killings on the Rise, Report Reveals, 2016).

In the aftermath of Qandeel Baloch’s murder by her bother public protests intensified in the year 2016 and the parliament enacted anti honour killing legislation. The new law incorporated drastic punishment for the murderers who could no longer be pardoned by the relatives of the victims. The rise in honour killing is indicative of the fact that harsher law alone cannot ensure justice for women. The law enforcement agency should guarantee that the police impartially investigate the cases without fear or favor (Ijaz, 2017).

In the male dominated society, women are considered as a commodity of males who control their lives and keep a watch on their social interaction. Dr. Kiran Ishfaq (consultant clinical psychologist) stated that the misconceived idea of male being a protector and owner of women; as well a social pressure of society to regulate the behavior of women, including religious beliefs that the honour of a man is associated with his wife’s dignity; are reasons leading to honour killing. She added that low self-esteem is also responsible for honour killings. Besides, other reasons that contribute to honour killing include inability to control anger, mood disorders, jealously, paranoid schizophrenia or other psychological disorders (Sheikh, 2016).

There is a very scant literature available on honour killing as it is not a very common practice but still an astonishing one which needs explanation. Nasrullah, Haqqi, and Cummings (2009) tried to explore the attitude of people towards honour killing, by associating it with different demographic variables. They revealed that mostly women are killed in the name of honour, some of such cases are registered but many incidents are not reported. The study also revealed that low education and socioeconomic status had a strong relationship with the positive attitude towards honour killing. In another study, Eisner and Ghuneim (2013) concluded that besides low educational level, the conservative family structure was also associated with positive attitude towards honour killing.

Sheeley (2007) conducted a study in Jordan investigating the opinion on punishment awarded to honour killers. According to the result 95.5% individuals responded that honour criminals should be treated strictly. On the contrary, another study conducted in Jordan by Hadidi, Kulwicki, and Jahshan, (2001) concluded that courts in Jordan took lenient action in honour killing cases, and awarded 6 months to 15 years of punishment to honour killers. Similarly, in Pakistan the opinion of university students was solicited regarding honour killing incident, outside Lahore high court of a girl who had contracted love marriage, and was killed by her family. The survey result indicated that there was a difference between the opinion of males and females. Majority of males justified the killing of female as compared to half of the females (Shaikh, Shaikh, Kamal, & Masood, 2010).

Honour killings takes place for various reasons in different cultures/communities e.g., in India honour killings are carried out for marrying outside the caste, over dowry and the like. Whereas, in Pakistani society the main reason behind honour killings is adultery. As in Islam there are strict sanctions against an adulterous act, this might be the reason that people cannot tolerate such acts and consider it justified to kill a person involved in such an offense. Most of the previous researches conducted on the honour killings also focused on finding people’s attitude toward honour killing caused by adultery. However, the scale developed in the present study is unique in the sense that it measures aspects/reasons other than adultery which can cause honour killings in the Pakistani society.

As it is highlighted in the literature that in Pakistan, 791 honour killings took place only in 2010 (Karimjee, 2011), whereas, about 4000 such incidents occurred during 1998 to 2004. These statistics yield high need to develop a culturally relevant scale measuring attitude toward honour killing, encompassing all possible honour killing situations. Carrying out a research on such a sensitive aspect without a valid and reliable scale is not going to give promising results. The present study aimed at the establishment of a psychometrically sound scale of attitude towards honour killing and then field testing it on general population, honour killers and killers in general. Most of the studies conducted up till now made use of general public, the present study took a lead in asking honour killers and other murders too about what they think justifies honour killing. There can be a difference in a pre and post thinking of a person before and after honour killing but the results show one tries to justify the act once committed as compared to regretting it afterwards. Keeping in mind the above stated scenario, present study was planned to measure the following objectives:

1. To develop a valid and reliable scale of attitude towards honour killing 2. To compare general population, honour killers and murderers in general on attitude towards honour killing

METHOD

Phase I: Development of Attitude towards Honour Killing Scale
Objective of this phase was to develop the scale measuring attitude towards honour killing.

Step.1. Generation of items Pool for Attitude towards Honour Killing Scale
To measure the attitude of people towards honour killing, a new scale was developed based on the reports about honour killing indifferent daily newspapers, in other electronic and print media and through information gathered from general population. On the basis of these reports 21 different scenarios were presented with 6 possible options. The options given for every scenario varied in severity i.e., in a continuum from victims should not be punished at all---- to---- the murder of the victim was justified. The first draft was assessed by 10 academics, 3 doctors, 1 school and 3 college teachers, and 3 police officers. The panel was also asked to rank the different options in order of severity for each of the scenarios. According to their feedback, few of the items were deleted as they were considered a mere repetition and some options given for the scenarios were considered irrelevant. The scale was revised on the basis of their feedback.10 scenarios and 2 options for each scenario were removed .The questionnaire was again given to 20 academics and professionals, who appraised the scale and suggested few more changes, which were incorporated. A third draft was then given to 20 academicians, who reported no problems with the questionnaire, and finally 11 items with four options were retained.

Step.2. Exploratory Factor Analysis (EFA)
Sample. The sample for exploratory factor analysis comprised of 326 (men = 190 and women = 136) individuals. Among which 52 were criminals (Honour killers=26 & Ordinary murderers = 26) and the rest was general population (n = 274). The data of the prisoners was collected from Bannu, Charsada, and Mardan jails. The data was collected by the male research assistant, as the females were not allowed to go inside the jails. The data from the general population was gathered from different universities in Peshawar and from people in markets, bus stops and other public places. Mean age of female sample was 25.4 years (age range= 20 - 54 yrs) and of male sample was 28 years (age range= 23 - 40 yrs).

Procedure. To collect data from the honour killers and other murderers, a permission letter was taken from Inspector General of Prisons in Khyber Pakhtun Khawa. The researcher and research assistant traveled repeatedly to Peshawar, Timergara, Mardan, Swabi, and Charrsadda jails to collect data from honour killers and murderers. The data was collected in the presence of superintendent jail and one constable in the office of the superintendent. Convenience sampling technique was used to collect the data from general population, by approaching them in different public places. The principles/Heads of three women colleges, two boy’s colleges, and two universities were requested and data was collected from the students and teachers after their verbal consent. Maximum number of people, agreed to fill out the questionnaires.

Results. The exploratory factor analysis was conducted on sample. The analysis showed that one factor solution was the best option. Kaiser-Meyer-Olkin test is equal to .83, verifying that sample size is adequate for the factor analysis. Bartlett sphericity test was also significant. Principal Axis factoring was applied as extraction method. One component was extracted from analysis of Attitude towards Honour Killing Scale.

Table 1:	Factor analysis of Attitude towards Honour Killing Scale (N=326)

Table 1 shows the unidimensionality of the scale. All the items are loaded on a single factor and their item loadings are higher than the cutoff point i.e., .40. The scale explains 31 % of variance.

Table 2:	Descriptive statistics and alpha reliability coefficients Univariate normality of Attitude towards honour killing scale (N=326)

Table 2 depicts the alpha reliability of attitude towards honour killing scale, and all the other indicators of the sample univariate normality are within the acceptable range. Items total correlation range from .45 to .69 for all the 11 items.

Phase II: Validation of Scale through Confirmatory Factor Analysis. Objective of this phase was to confirm the factor structure proposed in EFA for the attitude towards honour killing scale.

Sample. The sample for confirmatory factor analysis included 374 individuals (men = 198 and women = 176), including general population, murderers and honour killers. Incarcerated sample (n = 72) was taken from Peshawar, Mardan, Charsada, and Teemargara jails (26 = honour killers & 46 = ordinary murderers). Mean age of the inmates was 32 years. Data from the general population was taken from universities, colleges and public places. Age range of the general population was 18 to 60 years with the mean age of 22 years.

Procedure. Same procedure mentioned in phase 1 was followed to collect data from sample as sample characteristics were same.

Results. Confirmatory analysis was run on the sample and the finding of the analysis confirmed that one factor solution was the best option. The original one factor model is subjected to a priori confirmatory factor analysis. A total of 11 indicators are used to predict attitude towards honour killing. The hypothesized model had a barely fit with respect to parsimony (MLM mean χ² = 142.42, df = 44; CFI= 0.93; RMSEA = 0.077 (C.I = 0.06 - 0.09). Examination of modification indices indicated correlation of item 11 is with item 9 and 10 will improve the model fit. Hence, two correlational pathways were added (pathways 1: item 11- item 9 and pathways 2: item 11 – item 10). This improved the model fit with better RMSEA value than the hypothesized model. (MLM mean χ² = 109.01, df = 42; CFI =0.96; RMSEA = 0.06 (C.I = 0.05 - 0.08). The standardized loadings ofindicators are moderate in magnitude. The smallest standardized estimate (0.45, p < 0.0001) and R² statistic (0.2, p< 0.0001) is of item11. The remaining estimates for the standardized regression weights range from 0.6 (item 6) to 0.75 (item 2) and R² statistics ranging from0.3 (item 4) to 0.56 (item 2).

Table 3:	Factor Loadings of Attitude towards Honour Killing Scale (N=374)

Table 3 shows the same results of CFA as that of EFA carried out in phase I. All the 11 items are greater than .40 loading on a single factor.

Table 4:	Variance accounted (R2) by each indicator of Attitude towards honour killing (N=376)
Note. *p < 0.001. **p < 0.05. S.E = standard error; R2 = variance.

Table 4 confirmatory factor analysis was also conducted in Mplus to check the possibility of a dual factor model due to the correlational pattern between item no 9, 10, and 11. EFA also suggested that a single solution is the best for measuring attitude towards honour killing. As the three items that are significantly contributing towards a model fit are the last consecutive ones, the reason for their correlation can be their position in the scale. By answering the initial questions, respondents come to know about the nature of the scale as well as develop a mental set of responding to the questions due to which there is a correlation between them, hence it does not suggest the dual factor model. Following table presents the variance explained by each indicator as R².

Table 5:	Scores of Men Women, Honour killers, and Murderers on attitude towards honour killing Questionnaire (N=278)

The above mentioned table indicates that majority of the general population especially women have a negative attitude towards honour killing. However, as compare to the other items even women in general population score higher (168, M=1.86) on the item no 3, which was about adulteress woman.

DISCUSSION

Among the incidents of honour killing witnessed in our society, majority of the victims happen to be women in those incidents. These killings are justified on the basis of various reasons like social pressure, culture, and restoration of family honour. It is a general perception as if honour killing is an acceptable norm in our society and that people favor it. In order to understand whether this perception is valid about our society, it is important to solicit public opinion on this issue. Very few researches are carried out to elucidate public opinion about honour killing, as well to explore the stance of the honour killers in this regard. Honour killing can take place due to various reasons like, violations of social norms, elopement, adultery, etc., but in Pakistan and around the world mostly researches focus on opinion about people being killed for their adulterous behavior. The lack of literature suggested a need to develop a measure the assess attitude of general public and/or honour killers on this issue, encompassing various scenarios of honour killing.

For this reason a newly developed scale comprised of 11 items was subjected to exploratory analysis. As all the items raised questions about honour killing, oblique rotation was suitable to be applied on the data. A cut off point of .40 was set to retain the item in the scale. According to the analysis it was found that all the 11 items were loaded on the same factor, and none of the item was discarded as all the items had greater than .40 loading. Initially single and two factor solutions were applied using oblique rotation, but one factor solution was the best option, because theoretically the scale was designed to measure one factor that is attitude towards honour killing and it nearly explained 31% of variance. In the second phase of the study the original one factor model was subjected to confirmatory factor analysis which also supported the unidimensionality of the scale.

After the establishment of the psychometric properties of the scale, it was field tested on general population and incarcerated sample. The finding suggested that there existed a significant difference between the two populations (criminals and general population) on each item of attitude towards honour killing scale (see Table 5). As was expected, women scored significantly low than men on honour killing, followed by honour killers and other murderers respectively who scored significantly high on attitude towards honour killing. Pakistani society is patriarchal society dominated by men, usually women are suppressed and that is why they are the victims of the honour killing. This can be the reason that their attitude is highly negative towards honour killing as compared to men. The study findings are also in accordance with other researches. Shaikh, Shaikh, Kamal, and Masood (2010) also conducted a survey to find out the people’s opinion about honour killing. Their findings suggested that there was a substantial difference between males and females’ attitude towards honour killing, as compared to males only half the number of females justified the killing of adulteress women.

Results in the present study showed that the criminals i.e., murderers and honour killers were having highly positive attitude towards honour killing. A possible reason for this finding can be that one tries to stick to what one has done to justify the stance the person has taken and to avoid guilt afterwards. It can be attributed to aggressive, impulsive and criminal tendencies in such individuals, but it would need another study to prove these assumptions.

In Pakistan honour killing can take place for reasons like adultery, elopement, going against social norms etc. If we look at the results on 10 items majority of the women are not having favorable attitude towards honour killing, but on item number 3 which is about adulteress women and her killing by her husband even 20.4% of women and 33.3% of men in general population showed positive attitude towards killing of such a women. The findings indicate that only if a person is killed for adultery then people in general population approve the killing. The possible reasons could be that Pakistan is Islamic state and adultery of a married person is considered a stark debauchery in Islam. But contrary to the popular assumption that general population is in the favor of honour killing is rejected by the current research as mostly individuals scored against honour killing. Same result was found in Turkish population (Caffaro, Ferraris & Schmidt, 2014). Three scenarios were given to the participants, first scenario was about the allegation of adultery, second was about adultery and third scenario was about an adulteress women without trying to hide her relations from her husband. The results indicated that the Turkish people held the victim responsible for her being killed by her husband and favored lenient punishment for the honour killers.

LIMITATIONS AND SUGGESTIONS

The main shortcomings of the present study as it did not probe comprehensively the other possible factors which could have influenced honour killing like peer-pressure, social pressure, and the like that could have fostered honour killing. The sample of general population mostly consisted of young people, whereas, wisdom gets developed at later ages of life. An inclusion of elderly population might have given different results on the issue at hand. Future researchers should study these kinds of social issues by using both qualitative and quantitative approaches simultaneously to delve into the bottom of the problem. Future researches can also investigate the role of different demographic variables behind honour killings. In depth interview should be carried out with the honour killers to understand what they feel after the incident took place, as it is not necessary an act committed once looks justified after its happening too.

CONCLUSION

The findings reveal that in general, people show negative attitude toward honour killing, contrary to the false perception against Pakistani society that that has indoctrinated the honour killing as a norm. In comparison to general public, actual honour killers and other murderers show positive attitude toward honour killing.

REFERENCES

1. Caffaro, F., Ferraris, F., & Schmidt, S. (2014). Gender differences in the perception of honour killing in individualist versus collectivistic cultures: Comparison between Italy and Turkey. Sex Roles, 71(9), 296-318.
2. Eisner, M., & Ghuneim, L. (2013). Honor killing attitudes amongst adolescents in Amman, Jordan. Aggressive Behaviour, 39(5), 405-417.
3. Felix, Q. (2004, July 22). Honour killing and “karo kari” in Pakistan.
4. Ijaz, S. (2017, September 25). Honor’ killings continue in Pakistan despite new law.
5. Jafri, A. H. (2008). Honour killing: Dilemma, ritual, understanding.Karachi:Oxford University Press.
6. Karimjee, M. (2011, December 20). Pakistan: 675 women killed in honour killings in 2011. The Global Post.
7. Hadidi, M., Kulwicki, A., & Jahshan, H. (2001).A review of 16 cases of honour killings in Jordan in 1995. International Journal of Legal Medicine, 114(6), 357-359.
8. Nasrullah, M., Haqqi, S., & Cummings, K. J. (2009). The epidemiological patterns of honour killing of women in Pakistan. European Journal of Public Health, 19, 193-197.
9. Pakistan Honour Killings on the Rise, Report Reveals. (2016, April 1).British Broad Casting News.
10. Sheikh, M. (2016, July 24). Psychology behind ‘honour killing’–not normal, a mental illness.
11. Shaikh, M. A., Shaikh, I. A., Kamal, A., & Masood, S. (2010). Attitudes about honour killing among men and women: Perspectives from Islamabad. Journal of Ayub Medical College, 22(3), 38-41.
12. Sheeley, E. R. (2007). Reclaiming honor in Jordan: A national public opinion survey on “honor” killings. Amman, Jordan: Black Iris.