INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue X, October 2025
www.ijltemas.in Page 198
Relationship between Social Dysfuction Factors and Femicide
Cases among Young Women in Juja Sub-County, Kenya
1
Winfred Kagwiria,
1
Japheth Lumadede,
2
Benjamin Mugambi
1
Department of Social Sciences, Tharaka University, Kenya
2
Department of Social Sciences, Tharaka University
DOI: https://doi.org/10.51583/IJLTEMAS.2025.1410000027
Received: 30 September 2025; Accepted: 07 October 2025; Published: 06 November 2025
Abstract: Femicide has increased in the last decade in the global index. In Kenya, femicide occurs under diverse age brackets but
recently the trend is high among young women (16-35 years). In particular, the brutal murders of young women: emerging and
with increasing trends and incidents in various places call for immediate solutions. The purpose of this study was to examine
relationship between social dysfuctions and femicide cases among young women in Juja sub-county. The Feminist theory and
Routine activity theory guided the study. The study was done at Juja sub-County in Kiambu County, Kenya. The study adopted a
mixed–method research approach, utilizing the embedded design. The target population was 300,948 respondents. The accessible
population was 200,510 respondents, out of which a sample of 278 was drawn through the Slovin's formula. The total sample was
288 respondents consisting of the 10 key informants included in the study. The researcher drew the sample using Stratified simple
random sampling. Questionnaires and interview schedules were the main instruments of data collection where 245 questionaires
were fully filled and returned and all the 10 interviews were conducted succesfully. Reliability was measured through Cronbach
Alpha a statistic coefficient (a value between 0 and 1) used to rate the reliability of an instrument and was 0.813. Poisson
regression analysis was conducted on the quantitative data with the help of Statistical Package for Social Sciences (SPSS) version
26.0 software. Qualitative data was analysed through thematic content analysis utilizing MAXQDA tool. The study established
no statistically significant relationship between social factors and femicide with a p-value of .367, well above the .05 threshold,
the model indicated that social variables do not have a direct, independent effect on the number of femicide cases. The study
concluded that social dysfunctions alone are not the cause of femicide but is as a result of a more complex, multi-faceted
dynamic at play than a simple cause-and-effect and recommended that to reduce these risks, there is a need for strengthened
social support systems. The study aimed to benefit the policymakers on formulating relevant social policies, inform the society
on better societal practices, contribute to the academia theories and literature on femicide, as well as inform young women in
adopting femicide prevention strategies.
Keywords: Social dysfunctions and femicide among young women
I. Introduction
political issue of global concern. Femicide is the killing of women and girls because of their gender (Websdale, 2014). It is
currently seen through the view of violations of human rights. The problem of femicide is present everywhere in the world, so
there is no country that does not need to deal with this phenomenon systematically (Richards, 2023).
According to the United Nations on Drugs and Crimes, globally, approximately 51,100 young women were killed by their
intimate partners or other close people known to them during 2023, higher than the 2022 estimate of 48,800 victims. The 2023
figures mean that 60 per cent of the almost 85,000 young women killed intentionally during the year were murdered by their
intimate partners or other close people known to them (UNODC 2023). In other words, an average of 140 young women,
worldwide lost their lives every day at the hands of their partner or a close person. Worldwide. While the research has been done
previously Further investigation on these factors and contexts of young women femicide, including victims aged 16 to 35 years,
would improve our understanding of young women femicide and potentially guide the implementation of prevention strategies
specifically tailored to the victims.
II. Literature Review
Social dysfunctional elements play a dominant role in femicide among young women because they allow violence, abuse and
gender disparities to silently persist. The failure by social institutions to respond properly to gender-based violence stands as a
major dysfunctional element (Cohen, 2013). The social institutional inadequacies combined with weak social support and
agreements to fight femicide cases that occurs within the society is an important factor in femicide. Many young women
especially those who face vulnerability feel unsafe and unsupported by the society hence increased chances of femicide (Twaine,
2012). The ongoing impunity of violent perpetrators exists because these environments provide them with complete assurance
that their harmful acts will go unchecked. The absence of institutional responsibility gives abusive behaviors enough freedom to
evolve into dangerous situations leading to femicide. According to Radford (2022), institutions with uncertain responses toward
violence foster environments of silence and panic which push young women toward dangerous partner agreements and
unsafeguarded violent encounters.
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue X, October 2025
www.ijltemas.in Page 199
The social training process frequently establishes a setting that supports justifications for violence against women leading to
discontinued interventions (Banyard, 2015). Members of these communities typically accept violence as a suitable means to
manage conflict because it helps them maintain control yet this habitual behavior results in tragedies of femicide, when violence
escalates beyond controlled levels. Gender stereotypes held by peer groups because males need dominance together with females
needing obedience remain persistent according to (Tepperman, 2017). The harmful environment found in the social spaces creates
settings that accept violent dynamics while reducing awareness around femicide cases.
Victim blaming notion enable the rapid dissemination of narratives that frame victims as responsible for their own murders. This
is evident in high-profile femicide cases in Kenya, such as the deaths of Sharon Otieno and Ivy Wangeci. Sharon Otieno, a
university student who was brutally murdered in 2018, was vilified for her relationship with a married politician. Some labeled
her as a "slay queen" and accused her of greed and immorality, effectively shifting blame from the perpetrators to her lifestyle
choices (Mutua, 2020). Similarly, Ivy Wangeci, a medical student killed by a rejected admirer in 2019, faced posthumous scrutiny
online. They suggested that her rejection of the perpetrator's advances and alleged material demands provoked her murder
(Macharia J.2020). These narratives not only dehumanized her but also diverted attention from the act of violence to her
perceived faults (Macharia, 2016).
Despite these important insights, there are also limitations in these analyzed factors and further studies are needed to explore
more on issues of intimate relationships which is a crucial issue among young women, social media influences especially with the
developed technological dynamics and also legal systems inadequacies and how these factors interconnect to cause femicide
among young women. Eventually, such studies will help to develop better strategies to fighting the issue of femicide.
Conceptual Framework
III. Methodology
This study was carried out at Juja sub-County in Kiambu County, Kenya. The target population was 300,948 residents of Juja
sub-county as per the official demographic statistics (Census, 2019). This sub-county has five wards; these are Murera, Theta,
Juja,Witeithie and Kalimoni ward. The study used a mixed-method research design called embedded design. Research design are
the various approaches that the researcher uses in answering research questions (Creswell, 2019). In this design, quantitative and
qualitative components were combined at the same time. One method is given less priority, either quantitative or qualitative and
then it will be embedded within the dominant method (Greene & Caracelli,1997). The quantitative method was dominant and
qualitative method was embedded in this study
The target population for this research was 300,948 population of Juja sub-county as per the 2019 Kenya Population and Housing
Census. Saunders (2007) defines the target population as the members of the real and hypothetical set of people, objects or events
that the researcher intends to generalize the results of the research. The accessible population was 200,510 adult population of
Juja sub-county. It consisted of both male and female gender. The population comprised respondents from different locations in
the sub-county among the five wards. A sample size of 278 respondents was used in this study. 10 key informants from all the
wards were purposively selected, comprising of the security officials i.e., police, and the community activists who provided a
complementary insight on femicide.
From the accessible population of 200,510, a sample size was drawn using Slovin’s formulae (Slovin, 1960):
n =
n =
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue X, October 2025
www.ijltemas.in Page 200
n =
n = 277.393
Thus, the sample size is 278 respondents
The research instruments were questionnaires and interview schedules. The questionnaires leaned towards the objectives of the
study to ensure validity. The questionnaires were used to collect data from the Juja subcounty residents and the interviews were
used to collect data from key informants(police and gender activists). The questionnaires had closedended questions for the
female inmates. The questionnaire also used Likert scale items extensively. There were four sections in the questionnaires; three
addressing the objectives of this study and one section gathering biodata. The questionnaires were a key part of this research.
From the interview schedule, the key informants provided information useful to the research and suggested solutions. 10 of them
participated in addressing the interview schedules.
The questionnaires were evaluated before the actual study to check for appropriateness and validity. The objectives of the study
were also checked with the questionnaires to ensure alignment of purpose. Seeking counsel from supervisors and experts in the
field assisted to sight gaps and areas that needed adjustments. Corrections were part of the process of ensuring the questionnaires
were effective. Validity is about how well research findings show similarity between participants and true findings among
individuals outside the study. A study instrument is valid when it can measure appropriately the different variables and how the
variables influence each other and interact (Bryman, 2016).
A pilot study took place at Nairobi County, Kenya. Meru GK Prison and Embu GK Prison are similar only that the latter has
fewer inmates. Nairobi county was chosen because it has similar characteristics with the primary study area and also, its
proximity to Kiambu County for easier logistics.
A pilot study aims at testing for reliability and gauging the internal consistency of the questionnaire. The researcher employed the
Cronbach alpha coefficient to test for internal consistency. The Cronbach alpha is a statistical coefficient that is a value between 0
and 1, used to test how reliable an instrument is. After getting the feedback, the data set was split into two and a score for
participants was calculated from each half of the scale. Cronbach alpha value of at least 0.75 was acceptable and indicated
reliability (Allibang, 2020). The same scores on each half indicated high correlation and reliability. The test gave a 0.816 result
meaning that the research instruments were reliable.
The table below shows the Cronbach alpha test results:
Table 1: Reliability Test on Pilot Data
Cronbach's Alpha
N of Items
.816
28
Data collection is the process of collecting data for purposes of study using the relevant sources (Allibang, 2020). The data
collection procedure had the researcher secure an introductory letter from the Tharaka University Research Ethics Committee that
assisted in obtaining a research permit from the National
Council of Science and Technology (NACOSTI) before starting the research process. The researcher sought permission from the
Commissioner of Prisons before conducting any physical visits to the prisons. The researcher visited various institutions of
interest and informed them of the intention to collect data. The researcher proceeded to sample and obtain the participants from
the list of names offered. The researcher introduced the participants to the purpose of the research and sought their consent to
participate in the research.
IV. Results And Discussions
The analysis was both qualitative and quantitative. Descriptive statistics like proportions, means, frequencies, and standard
deviation were the data analysis and data were presented in tables. The researcher used Poisson regression to analyze the
relationship of the independent variables and the dependent variable and thematic content analysis was used to analyze qualitative
data. Quantitative data was analyzed using the Statistical Package for Social Sciences (IBM SPSS Statistics V26).
Respondents Response Rate
A sum of 278 questionnaires was distributed to residents across the five wards of Juja Sub-County. Out of these, 245 were
successfully filled and returned, representing a response rate of 88.1%, which is within the acceptable range of 80%–100% as
recommended for research standards (Morton, 2012).In addition, all 10 key informants purposively selected for interviews
comprising 5 gender activists and 5 security officials fully participated, giving a 100% response rate for the qualitative
component. This strong response enhanced the reliability and completeness of the study findings. The response distribution is
summarized in Table 3 below.
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue X, October 2025
www.ijltemas.in Page 201
Table 3: Response Rate Summary
Respondent
Sample
Responses
Percentage (%)
Residents
278
245
88.1
Informants
10
10
100
Total
288
255
88.5
Distribution of Respondents by Age
This subsection presents the age distribution of respondents who participated in the study. Out of the 245 participants, the
majority were within the 25–30 age group, representing 45.4% of the total sample. The age group with the fewest respondents
was 46 and above, accounting for 10.6% of the sample. This distribution reflects the focus of the study on young women and the
population dynamics within the sub-county. Compared to some international trends for example, a study by Bronson (2017) that
noted a higher prevalence of certain age groups in violent crime victimization in the U.S .This study’s findings show a higher
concentration of affected individuals to be young , aligned with the Kenyan femicide trend among women aged 16–35.
The age data are summarized in Table 4 below.
Table 4: Distribution of Respondents by Age
Age
Frequency
Percent
Valid Percent
Cumulative Percent
16–24
58
23.7
23.7
23.7
25–33
77
31.4
31.4
55.1
34–42
42-45
30
30
12.25
12.25
12.25
24.5
24.6
46 and above
4
20.4
100.0
Total
245
100.0
100.0
Distribution of Respondents by Level of Education
The majority of respondents had completed secondary education, accounting for 43.3% of the sample. This reflects the dominant
educational attainment among young women in the area and suggests a basic level of literacy and awareness. The second most
common level was primary education, representing 35.1%, while college or university graduates made up 15.5%. A small portion
(6.1%) of the respondents had no formal education, which may point to disparities in educational access within the sub-county.
Table 5: Distribution of Respondents by Education Levels
Education Level
Frequency
Percent
Valid Percent
Cumulative Percent
No Education
15
6.1%
6.1%
6.1%
Primary Education
86
35.1%
35.1%
41.2%
Secondary Education
106
43.3%
43.3%
84.5%
College/University
38
15.5%
15.5%
100.0%
Total
245
100.0%
100.0%
—
Distribution of Respondents by Gender
While the main focus of the research was on femicide cases among young women, male respondents were also included to
provide a broader community perspective on the contextual risk factors contributing to femicide in Juja Sub-County.
Out of the 245 respondents who successfully returned the questionnaires, approximately 61.2% were female, and 38.8% were
male. This was consistent with the sampling design, which prioritized the inclusion of young women due to their direct relevance
to the study topic. The presence of male respondents added value by offering insights into societal norms, gender dynamics, and
community awareness surrounding femicide cases.
The distribution is summarized in Table 6 below.
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue X, October 2025
www.ijltemas.in Page 202
Table 6: Distribution of Respondents by Gender
Gender
Frequency
Percent
Valid Percent
Cumulative Percent
Female
163
61.2%
61.2%
61.2%
Male
82
38.8%
38.8%
38.8%
Total
245
100.0%
100.0%
—
Normality Test
To determine whether the dataset met the assumptions required for parametric statistical analysis, a normality test was conducted
using the Kolmogorov–Smirnov and Shapiro–Wilk tests.
Factor
W-statistic
p-value
Interpretation
Eco_Mean
0.894
0.377
Data is normal (p > 0.05)
Cul_Mean
0.777
0.052
Data is normal
Social_Mean
0.900
0.409
Data is normal (p > 0.05)
Data Analysis
All research instruments were consistently applied to this objective. The questionnaires used a Likert scale to measure the degree
of agreement, ranging from 1 to 5, where: 1 = Strongly Disagree, 2 = Disagree, 3 = Neutral, 4 = Agree, and 5 = Strongly Agree.
Additional views were gathered from key informant interviews with gender activists and police officers. The analysis of the
findings is presented as follows: In regard to social media facilitating deceptive romantic approaches. The mean score for this
item is 4.1, indicating a high level of agreement among respondents. About 66.9% agreed or strongly agreed that social media
platforms are used to deceptively approach young women. Only 18.8% disagreed, while 14.3% were neutral. This reflects a
strong perception that digital platforms are playing a role in exposing young women to risky relationships that may escalate to
femicide.
A majority of 71.1% of respondents agreed or strongly agreed with this claim that Young women are lured into dangerous
spaces via social media. The mean score is 3.9, showing strong agreement. Just 15.1% of respondents disagreed and 13.8%
were neutral. These findings affirm the concern that social media is an active tool in drawing young women into exploitative or
dangerous interactions.
On the statement that there are no strict law enforcement measures, only 14.3% disagreed or strongly disagreed, while a
much larger 73% agreed or strongly agreed. The mean score of 3.5 and relatively high SD of 1.28 reveal a weak perception of
law enforcement effectiveness in protecting women. These results suggest systemic legal inadequacies that could enable femicide.
A substantial 73.5% agreed or strongly agreed that legal delays are common in cases concerning young women. The mean score
of 3.9 further supports this concern. Only 14.3% of respondents disagreed. These delays may reduce access to justice and
increase victim femicide vulnerability.
A strong 78.4% of participants agreed or strongly agreed that violence is present in many young women's relationships, with a
mean of 4.0. This validates the frequent connection between intimate partner violence and femicide. These conflicts combined
with poor decision making, poor self-control and lack of experience leads to poor solving of issues with physical abuse being
involved and to extreme levels femicide.
In regards to experiences of jealousy, control and possessiveness, most 72.8% agreed or strongly agreed while 8.6% disagreed
or strongly disagreed, and 12.7% were neutral. With a mean of 3.9, this suggests that toxic behaviors such as control,
unfaithfulness, and jealousy are prevalent and likely contributing factors to femicide. This aligns with prior findings that femicide
is often preceded by emotional abuse and coercive control.
Table 12: Influence of Social Dysfunctions on Femicide (N = 245)
Statement
1 (%)
2 (%)
3 (%)
4 (%)
5 (%)
Mean
SD
Social media facilitates deceptive approaches
27 11.0%
19 7.8%
35 14.3%
65 26.5%
99 40.4%
3.8
1.20
Young women lured via social media
176.9%
20 8.2%
34 13.8%
92 37.6%
82 33.5%
3.8
1.18
No strict law enforcement measures
23 9.4%
39 15.9%
4116.7%
8032.7%
62 25.3%
2.6
1.26
Legal process delaysfor women’s cases
14 5.7%
218.6%
30 12.2%
96 39.2%
84 34.3%
3.9
1.09
Intimate partner conflict observed
10 4.1%
17 6.9%
26 10.6%
101 41.2%
91 37.2%
4.0
1.03
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue X, October 2025
www.ijltemas.in Page 203
Jealousy, control in marriage
13 5.3%
22 9.0%
31 12.7%
89 36.3%
90 36.7%
3.9
1.10
Average Mean: 3.67 Average SD: 1.14
The data clearly shows that social dysfunctions in the immediate environment of young women plays a crucial role in shaping
violent tendencies that may eventually lead to femicide.
The Poisson regression was calculated to measure the strength of the relationship between social dysfunction indicators and
femicide. The analysis also found no statistically significant relationship between social factors and femicide.
Parameter
Value
df
Significance (p-value)
Interpretation
Social
.814
1
.367
The relationship between social factors and femicide is
not statistically significant.
With a p-value of .367, well above the .05 threshold, our model indicates that social variables do not have a direct, independent
effect on the number of femicide cases. This outcome suggests that social dynamics contributing to femicide may not be easily
isolated but are instead part of a larger systemic issue.
Our findings reveal that none of these factors, when considered in isolation, have a statistically significant relationship with
femicide. This suggests a more complex, multi-faceted dynamic is at play than a simple cause-and-effect model would indicate.
Despite the lack of significance for individual predictors, the overall model fits the data very well. The Deviance and Pearson
Chi-Square values were .013 with 1 degree of freedom. Both of these values, when divided by their degrees of freedom, are very
close to one, which indicates that our model adequately describes the observed variation in the data. The goodness of fit confirms
that the model is well-specified and provides a reliable framework for future research into the intricate relationships between
these factors and femicide.
To better understand how social dysfunctions, contribute to femicide among young women, the researcher interviewed gender
activists and police officers within Juja Sub-County. The responses were categorized according to three key themes: social media
influence, legal system inadequacies, and relationship-related risks.
These themes reflected the indicators in the conceptual framework. Interviewees were asked to share their perspectives and lived
experiences dealing with cases of femicide, gender-based violence, and systemic challenges. Their insights provided in-depth
qualitative context to the statistical results captured from the questionnaire.
Stakeholders observed that social media has increasingly become a deceptive gateway through which young women are trapped
into relationships that expose them to harm, including femicide. According to both gender activists and police officers, platforms
like Facebook, WhatsApp, and Instagram are commonly used by perpetrators to present false identities, offer fake job
opportunities, or lure women with material promises. Here are some of the views.
“We have had many cases where young women were contacted online by men pretending to be offering jobs or relationships,
only for them to end up in abusive environments—some were found dead days later. It’s becoming common, especially where
young women are desperate for work or attention.”
(Informant 3, Female, 12th July 2025)
“There are real cases we’re following right now where the victim was last seen going to meet a man she met on Facebook. These
digital interactions are not being monitored well, and many young women fall for them.”
(informant1, Male, 13th July 2025)
These responses indicate that media platforms vulnerability, especially in emotionally unstable situations, can expose young
women to fatal outcomes. Social media creates easy access for predators, especially when the victims lack digital literacy or
oversight.
In discussing how laws and law enforcement contribute to or prevent femicide, most respondents were critical of the effectiveness
and implementation of gender protection laws in Juja Sub-County. While Kenya has a legal framework in place (e.g., the Sexual
Offences Act, Protection against Domestic Violence Act), enforcement was described as slow, dismissive, and under-resourced.
A police officer admitted that:
“We do have laws, yes, but most officers are not properly trained to handle femicide cases. Sometimes, reports are treated as
‘domestic misunderstandings’ until it’s too late and someone dies.”
(Informant 5, Male, 14th July 2025)
Similarly, a gender activist pointed out that
“Young women file complaints, but due to case backlog, lack of legal aid, and sometimes corruption, the suspects are released
and end up harming or killing more women. The system does not protect them.”
(Respondent 2, Female, 13th July 2025)
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue X, October 2025
www.ijltemas.in Page 204
Some respondents also noted that delayed investigations and poor witness protection discourage many survivors from reporting
threats or abuse, even when their lives are at risk. These sentiments align with questionnaire findings where over 70% of
respondents agreed that legal delays contribute to femicide vulnerability. A major consensus among stakeholders was that
intimate partner relationships are the leading drivers of femicide cases in the area. Respondents noted a pattern: most femicide
cases are not random acts of violence, but rather final escalations of long-term abuse, control, jealousy, unfaithfulness and
emotional manipulation.
A gender activist reflected:
“We’ve lost too many girls because they were trying to leave abusive relationships. The boyfriends couldn’t handle rejection and
decided to kill them. It always starts with control, checking her phone, isolating her, threatening her and it ends in death.”
(Respondent 4, Female, 12th July 2025)
A police officer also revealed:
“The hardest part is that most of these killings were predictable. There were threats, there were past assaults, but the women
never felt protected enough to run or report. Some felt ashamed, others feared being judged.”
(Respondent 6, Male, 13th July 2025)
These qualitative findings mirror international and regional studies. For instance, UN Women (2022) emphasized that most
femicide cases globally are committed by intimate partners, and early warning signs often go ignored. Similarly, Africa UNiTE
Campaign flagged weak justice systems and social media exploitation as emerging contributors to femicide in urban and peri-
urban Kenya. The insights from gender activists and police officers highlight the multidimensional risk posed by social
dysfunctions. From digital deception, systemic failure, to toxic intimacy, young women in Juja Sub-County face compounding
threats that can lead to fatal outcomes. The stakeholders' voices reveal urgent gaps in prevention, legal protection, and public
awareness
V. Summary, Conclusions, And Recommendations
The following are the summary findings of this study:
Social dysfunctions particularly social media manipulation, weak legal enforcement, and toxic romantic relationships were
revealed to be a major contributor to femicide. The majority of respondents agreed or strongly agreed that social dysfunctions
play a significant role in femicide. Specifically, 82.9% agreed that social media manipulation contribute to femicide, while 81.6%
agreed that intimate relationship issues is also a major risk in femicide cases. Moreover, a majority cited delays in legal
proceedings as persistent problem in causing femicide. The overall mean score for this objective was 4.05. The Poisson regression
analysis found no statistically significant relationship between social factors and femicide. With a p-value of .367, well above
the .05 threshold, the model indicated that social variables do not have a direct, independent effect on the number of femicide
cases.
Qualitative data from interviews further reinforced these findings. They highlighted how jealousy, control, and infidelity were
recurring indicators in femicide cases.
Policy Recommendations
i. Public campaigns targeting both young women and men should promote awareness on digital safety, healthy online
relationships, and how to detect manipulative behaviors online.
ii. Law enforcement agencies should fast-track investigations and prosecutions related to femicide cases. Gender desks in
police stations must be strengthened and made youth-friendly and survivor-centered.
iii. The county government and NGOs should invest in psycho-social support, particularly for couples and young people.
Relationship education should be integrated into youth programs, churches, and social media platforms.
iv. The government should integrate findings on social dysfunctions into the review and update of current gender policies in
the country.
v. They should also prioritize multi-sectoral responses that address root causes such as unemployment, mental health, and
social exclusion.
vi. Establishing community watch networks and safe spaces for at-risk women and girls.
vii. Training local leaders, social workers, and law enforcement on identifying early signs of social dysfunction that may
escalate to femicide.
References
1. Abrahams, N., Jewkes, R., Laubscher, R., & Hoffman, M. (2013). “Gender-based violence and femicide in South
Africa.” Journal of Interpersonal Violence
INTERNATIONAL JOURNAL OF LATEST TECHNOLOGY IN ENGINEERING,
MANAGEMENT & APPLIED SCIENCE (IJLTEMAS)
ISSN 2278-2540 | DOI: 10.51583/IJLTEMAS | Volume XIV, Issue X, October 2025
www.ijltemas.in Page 205
2. Adhiambo, L. (2018). "Patriarchy and Femicide in Rural Kenya: Cultural and Economic Dimensions." African Studies
Quarterly. [22].
3. Amnesty International Kenya (2021). "Femicide in Kenya: Social Media Trends and Societal Impacts."
4. Abrahams, N., Jewkes, R., Laubscher, R., & Hoffman, M. (2013). “Gender-based violence and femicide in South
Africa.” Journal of Interpersonal Violence
5. Baldry, A. C. and Winkel, F. (eds) (2009) Intimate partner violence, prevention and intervention, Hauppauge, NY: Nova
Science. Bourdieu, P.
6. Baldry, A. C., & Winkel, F. W. (Eds.). (2018). Intimate partner violence prevention and intervention: The risk
assessment and management approach. Nova Publishers.
7. Banyard, V. L. (2015). The impact of violence prevention education: A review of recent literature. Journal of
Interpersonal Violence, 30(13), 2126-2149.
8. Baldry, A. C. and Winkel, F. (eds) (2009) Intimate partner violence, prevention and intervention, Hauppauge, NY: Nova
Science. Bourdieu, P.
9. Baldry, A. C., & Winkel, F. W. (Eds.). (2018). Intimate partner violence prevention and intervention: The risk
assessment and management approach. Nova Publishers.
10. Banyard, V. L. (2015). The impact of violence prevention education: A review of recent literature. Journal of
Interpersonal Violence, 30(13), 2126-2149.
11. Campbell, J. C., Webster, D., Koziol-McLain, J., Block, C., Campbell, D., Curry, M. A., ... & Laughon, K. (2017). Risk
factors for femicide in abusive relationships: Results from a multisite case control study. In Domestic violence (pp. 135-
143). Routledge.
12. Caputi, J. and Russell, D.E.H. (2019) ‘Femicide: speaking the unspeakable’, Journal of Personality and Social
Psychology, Sep–Oct: 424–30.
13. Cattaneo, L. B., & Goodman, L. A. (2021). What is the relationship between economic abuse and intimate partner
violence? Journal of Interpersonal Violence, 20(7), 860-870.
14. Cohen, S., Janicki-Deverts, D., & Miller, G. E. (2013). Psychological stress and disease. JAMA, 298(14), 1685-1687.
15. Connell, R. W. (2015). Masculinities (2nd ed.). University of California Press.
16. Corradi, C., & Piacenti, F. (2016). ANALYZING FEMICIDE IN ITALY. OVERVIEW OF MAJOR FINDINGS AND
INTERNATIONAL COMPARISONS. Romanian Journal of Sociological Studies, (1).
17. Crenshaw, K. (2020). “Demarginalizing the Intersection of Race and Sex: A Black Feminist Critique.” University of
Chicago Legal Forum
