1 Introduction

The Syrian crisis, ongoing since 2011, is one of the worst humanitarian crises occurring globally [1]. As of October 2021, there are an estimated 13.4 million displaced Syrians [2]. Among the 6.7 million Syrian refugees forcibly displaced to other countries, 5.65 million are formally registered refugees [2]. Lebanon currently hosts the largest number of Syrian refugees per capita with refugees accounting for greater than a fifth of the country’s population [2]. Syrian families in Lebanon are often unable to meet survival needs of food, health care and shelter, 69% of households live below the poverty line, and over 70% of them lack legal status consequently facing ongoing risk of detention or deportation [3]. In Lebanon, over half of the refugee population is under age 18, [3] with approximately 500,000 school-aged Syrian children [2]. Refugee children are at risk of child labour, domestic violence, and sexual violence [4]. Refugee children also often experience food insecurity, barriers to education, lack of basic health services, and physical or psychological stress [3,4,5]. Syrian girls face additional gendered risks including sexual harassment, gender-based violence, and increased rates of child marriage [6,7,8,9,10].

Tripoli, located in north Lebanon, is the nation’s second largest city and was host to a 27% of Syrian refugees in Lebanon in 2021 [2]. Tripoli and its adjacent districts, Batroun, Bcharre, Kourra, Minnieh-Dennieh, and Zgharta, have high rates of urban poverty that are greater than the national average [11]. Tripoli is also home to Palestinian refugees who reside in camps within the metropolitan area. Some Syrian refugees reside in the Palestinian camps while others are integrated in the host community. In 2019, Syrian refugees constituted 20% of the population in greater Tripoli [12] and the spatial distribution of Syrian refugees had extended the population density away from the centre of the municipality [13].

Tripoli stands out with respect to security due to its long history with Syria [14]. Tripoli has been subject to considerable repercussions from the Syrian conflict resulting in significant sectarian, political and socioeconomic tensions. Specifically, the sectarian conflict in Syria was mirrored in armed clashes between two Tripoli neighborhoods, Bab al-Tabbaneh and Jabal Mohsen, in 2012 and 2013 [15].These tensions meant that Syrian families were both facing insecurity as refugees as well as physical danger resulting from ongoing conflict within their host community. Furthermore, the Tripoli security plan implemented in 2014 gave the Lebanese Armed Forces (LAF) control over Tripoli [16], causing Syrians to become targets at checkpoints and to be subjected to arrests [14]. Finally, with the arrival of more recent Syrian refugees in Tripoli, the city’s already precarious infrastructure was overstretched, creating greater competition for services between refugee and host communities. In fact, Syrian refugees in northern Lebanon were more likely to report negative relationships with the host community (9%), were more likely to list competition for jobs as a source of tension (54%), and more often noted competition for resources as a source of tension [3].

In previous research examining the experiences of Syrian refugee girls in Tripoli, Beqaa, and Beirut, particularly concerning child marriage, our quantitative findings revealed notable differences in responses from Tripoli compared to the other two locations. With this observation in mind, we conducted a secondary mixed-methods analysis with the objective of helping to explain these disparate responses. Our overarching objective was to understand Tripoli’s specific societal, economic, security, religious, and psychosocial context to better appreciate the unique needs of refugee girls in the area. These insights are valuable for government officials, policymakers, non-governmental organizations, and United Nations (UN) agencies to guide and improve humanitarian efforts to be more targeted and impactful. This is particularly relevant for Syrian refugee girls who have often not been included in research resulting in their unique experiences and needs going unrecognized.

2 Methods

The current work was a secondary analysis of data derived from a larger mixed-methods, cross-sectional study implemented in Lebanon in July and August 2016 by the ABAAD Resource Center for Gender Equality in Lebanon and Queen’s University in Canada [7]. Full details of study implementation have been previously published [17] and will be summarized here.

2.1 Participants

Individuals had to be 13 years of age or older to participate. A variety of participant subgroups were targeted for recruitment including married and unmarried Syrian girls, Syrian mothers, and Syrian fathers as well as men in the community (Syrian, Lebanese, or Palestinian). In many patriarchal societies of the Middle East, men predominantly make key decisions, including ones that affect girls, such as child marriage. Therefore, including both men, who offer insights into cultural dynamics and potential intervention points, and mothers, who may have contributed to decision-making for their daughters and might have personally experienced child marriage, was vital to our study design. Engaging a range of perspectives, from both genders, enriched our understanding of societal norms and bolstered collective efforts for positive change. In each of the three locations, a convenience sample of participants were interviewed in their naturalistic settings such as shops, transportation depots, cafes, and service delivery locations. Efforts were made to recruit participants from a diverse range of sites and socioeconomic backgrounds.

2.1.1 SenseMaker survey

The original research was designed to examine the experiences of Syrian refugee girls in Lebanon with a particular focus on child, early and forced marriage. ‘Girls’ were defined as children and adolescents under the age of 18. The parent study used Cognitive Edge’s SenseMaker® as a narrative capture tool to gather perspectives from narratives shared by participants about their experiences in Lebanon [18]. SenseMaker is based on the premise that storytelling is a natural way for people to share complex experiences and a way for people to make sense of the world around them [19, 20].

The SenseMaker survey began by offering three open-ended story prompts asking about the experiences of Syrian girls in Lebanon (see Appendix 1). Participants were requested to share a brief, anonymous narrative in response to their chosen prompt. Narratives were audio recorded and participants then interpreted the experiences shared in the narratives by plotting their perspectives on a series of predefined questions written by the study team. Dyad questions offered two possible response options and participants were asked to plot their perspectives anywhere along the spectrum from one extreme to the other. Triad questions offered three possible responses and participants plotted their perspectives between the three options on a triangle. Examples of both triads and dyads are provided in Appendix 1. The SenseMaker survey was written in English and then translated before piloting in May 2016 with further refinements made for clarification and ease of administration prior to data collection. These refinements included simplifying and clarifying language and improving the instructions about how to respond to the SenseMaking questions. The English survey was translated to colloquial Syrian Arabic (to ensure comprehension for all participants) by a Syrian translator, and then back translated to English to confirm accuracy.

We used SenseMaker as a valuable tool for obtaining comprehensive insights into complex and nuanced issues. It uses indirect questions to yield more genuine responses, especially on sensitive topics. Unlike typical surveys that have limited response options, SenseMaker allows for a broader range of answers, mitigating reporting bias. Social desirability is also minimized since within any given question, the possible answers are all neutral, all positive, or all negative, such that there is no one response that could be perceived as more acceptable. Participants interpret their stories, thus reducing researcher-induced bias. Data is digitally collected, making the process efficient: surveys typically last 15 min, and hundreds of narratives can be gathered weekly.

2.1.2 Data collection

Data were collected by a team of nine Syrian interviewers (six female, three male) as well as three male Lebanese interviewers, all of whom were selected by ABAAD based on nationality, place of residence, and prior work experience. Six interviewers were Syrian females, specifically chosen to interview Syrian girls and mothers. Three interviewers were Syrian men designated to engage with male Syrian participants, while three Lebanese men were selected to interview male Lebanese participants. The entire interview team completed a four-day training session immediately prior to data collection. The training covered SenseMaker methodology, research ethical practices, iPad usage, obtaining informed consent, an in-depth examination of the survey through various role-playing exercises, data uploading, and handling adverse events and program referrals. Interviews were conducted in a private location out of earshot of others using the SenseMaker application on iPad Mini 4’s. Narratives were transcribed and translated from Arabic to English by a native Arabic speaker. Since typical SenseMaking narratives are briefer than would be expected with traditional qualitative research, we refer to them as ‘micronarratives’.

2.1.3 Analysis

Responses to SenseMaker’s interpretation questions (i.e. dyads and triads) generated the quantitative data. The collective triad and dyad plots were examined visually in Tableau® version 10.1.5 to identify patterns in participants’ perspectives [18, 21]. Given the observed divergent responses among participants in Tripoli (in comparison to Beirut and Beqaa), we conducted a quantitative analysis to better understand these patterns.

For the dyads, in SPSS (IBM SPSS Statistics 24.0.0.0), the Kruskal–Wallis H test with a chi-squared test statistic was used to determine if the bar areas were statistically different between defined subgroups [22, 23]. A post-hoc analysis using Fisher’s Least Squares Difference then identified which groups differed from one another. A p-value of < 0.05 was considered statistically significant. Graphically, the dyad results are presented as violin plots representing the spectrum of possible responses from one extreme to the other. The shape of each plot represents the density of responses at each bar of the histogram and the asterisks indicate the mean for all responses for each respective subgroup.

For the triads, a geometric mean with a 95% confidence interval (CI) was generated for each subgroup using R scripts (R version 3.4.0). The 95% CIs are presented graphically as confidence ellipses around each geometric mean [24, 25]. If the 95% confidence ellipses for given subgroups did not overlap, we concluded that those subgroups responded in a statistically different way on that question [26]. After identifying patterns of perspectives in the quantitative data, we conducted an integrative mixed-methods analysis to aid in the interpretation of the quantitative findings. While data from all locations were included in the quantitative analysis (N = 1422), only narrative responses from Tripoli were included in the qualitative analysis since it was intended to understand why Tripoli was an outlier on some of the interpretation questions. Furthermore, only narratives that were about child marriage (as determined by the research assistant at the time of the interview) had been transcribed and translated from Arabic to English in the parent study and the current qualitative analysis is therefore limited to those transcripts. Data selection is illustrated in Fig. 1.

Fig. 1
figure 1

Flow diagram for narrative selection

Full size image

Our integrative mixed-methods analysis was performed through a content analysis of open-ended text narratives using an approach adapted from that of Castro et al. [27]. We first established thematic categories based on the quantitative data and then formulated a specific question designed to garner responses pertinent to our research topic: what was unique about Tripoli’s societal, economic, security, religious, and psychosocial context that impacted the experiences and needs of refugee girls in the area? We then coded the relevant narratives, treating each participant's experience as an individual case, generating response codes from their inputs, and choosing significant quotes that addressed our central question [28]. Representative quotes illustrating the main quantitative and qualitative findings are presented below.

2.1.4 Ethics

No identifying information was collected and thus the data were anonymous from the point of collection. Informed consent was reviewed in Arabic and indicated by tapping a consent box on the handheld tablet. Since the study consisted of a brief survey and was deemed to involve minimal risk, written informed consent was waived. Participants did not receive any monetary or other compensation. Individuals aged 13 and older were eligible to participate. Parental approval was not obtained since the survey was open-ended and avoided asking about anything of a sensitive nature. This aligns with current guidelines indicating that seeking parental permission might lead to bias and potential disagreements between teens and their parents [29]. The Queen’s University Health Sciences and Affiliated Teaching Hospitals Research Ethics Board approved this study protocol (#6014981).

2.2 Reflexivity and positionality

As part of the ongoing reflexive practice engaged throughout this work, we found it necessary to contextualize our positionality and recognize how our subjectivities have shaped the inquiries within this work. VD, CM, and SB are non-Lebanese/non-Syrian scholars based in Canada. They acknowledge their Western-influenced perspectives and how these might limit their ability to grasp cultural nuances and complexities of participants’ experiences. NB is a Lebanese scholar based primarily in Canada and similarly appreciates both her position as a Western scholar and as a member of the host country as a place of privilege that may bias data interpretation. SM is a Palestinian scholar living in Lebanon and working with Syrian refugee communities for many years. While being familiar with Arab cultural nuances, she also recognizes possible biases due to her Western education. The study was also approved by the Queen’s University Ethics Board as opposed to a local ethics process thus contributing to institutional level biases in the design of the study. To help minimize these biases, the parent study survey was piloted within the community to elicit and incorporate feedback both from the interviewers and the participants. Furthermore, we avoided interpretations of the shared narratives, instead allowing participants to interpret their own experiences.

3 Results

Demographic characteristics of Tripoli study participants are provided in Table 1. Overall, the sample was young (83.4% under the age of 35) with a slight predominance of men (51.7%). In total, 40.1% of the Tripoli respondents were young men. A quarter of the sample was Syrian girls under the age of 18 with 32% of these girls (n = 38) married at the time of the survey and another 6% divorced (n = 7). The sample largely self-identified as Sunni (93.8%) and another 5% declined to provide their religious affiliation.

Table 1 Demographic characteristics of Tripoli participants
Full size table

Our analysis noted three statistically significant differences in response patterns between participants in Tripoli versus other locations. These differences focused on: 1. perceptions among Syrian fathers that additional programs and services were needed to support Syrian girls in Tripoli; 2. perspectives among men in Tripoli that financial insecurity had contributed to the experiences of Syrian girls; and 3. a view among Syrian girls and mothers that girls were ‘protected too much’ in Tripoli. Each of these findings is presented below with illustrative participant quotes.

3.1 I. Call for additional programs and services by Syrian fathers in Tripoli

As shown in Fig. 2, Syrian fathers in Tripoli were significantly more likely to identify ‘programs and services’ as being helpful for Syrian girls as compared to all other subgroups in the three study locations, including Syrian fathers in Beqaa and Beirut.

Fig. 2
figure 2

Triad about what helps to support Syrian girls in Lebanon

Full size image

Fathers in Tripoli were statistically more likely to indicate that programs and supports were needed to support Syrian girls in Lebanon.

‘Programs and services’ was left open to the participant’s interpretation. From the accompanying narratives, however, it ranged from financial support to educational programs, health care and registry systems for residency status, marriages, and births. One Syrian father in Tripoli, who favored programs and services as a form of support, shared a story about limited access to education:

I know a family here in Lebanon that have been here 4 years now. They have a girl who’s always the first in class, she dreamed of becoming a doctor. She got to 9th grade and couldn’t finish her education because there aren’t classes here beyond the 9th grade. In public schools they do not accept her because she is Syrian. She had to stop her education. This is a very big problem. NarrID 1460: Syrian Father in Tripoli, in Lebanon for 5–7 years

Another Syrian father in Tripoli who answered similarly on the triad shared a story about limited health care and treatment options for a Syrian girl:

I know the story of a Syrian family who came here to Lebanon 2 and a half years ago. They have a 16-year daughter who was sick and couldn’t get treatment in Syria. It was also impossible in Lebanon because it was expensive. She is travelling now—thank God. NarrID 1327: Syrian Father in Tripoli, in Lebanon for 1–3 years

The following Syrian father in Tripoli highlighted the barriers to official registration and identification cards for his children:

We got married here in Lebanon, my family was in Syria. We lived in a tent. Our financial status is zero. I have two kids and I can’t register them. They are registered in the United Nations, but they don’t have an official identification card. We have nothing to do. Wherever we go the military is asking about the registered papers. I can’t let my wife work, because she has kids to care for, and my son this year got burnt by tea. NarrID 401: Syrian Father in Tripoli, unknown amount of time in Lebanon

As illustrated, many Syrian fathers in Tripoli referenced the unmet needs of their daughters in terms of access to education, health care, and legal documentation when they responded regarding what would help to support Syrian girls.

3.2 II. Tripoli men’s perspectives on the influence of financial insecurity in the lives of Syrian girls

Men in Tripoli (including Lebanese, Palestinian, and Syrian) were significantly more likely to identify financial resources as a driver in the lives of Syrian girls in comparison to all other groups. Similarly, as illustrated in Fig. 3 Syrian fathers in Tripoli were significantly more likely to identify financial resources as contributing to the shared girls’ experiences.

Fig. 3
figure 3

Triad asking what had contributed to the events in the shared narrative

Full size image

Men in Tripoli were more likely to indicate that the events in the shared story had happened because of financial resources. Syrian Fathers in Tripoli responded similarly.

In the following narrative, the family’s financial situation prompted the Syrian girl to join the work force, placing herself at risk due to police checkpoints, and eventually resulting in marriage for financial and protective reasons:

I will tell the story of a 17-year-old Syrian girl. She displaced to Lebanon with her family… As any other Syrian family, their financial situation was bad. The girl searched for a job to help her parents. Her father was working, and she has three younger siblings. So, she started working in a shop. Since she doesn’t have a permit, she was fearful at the police checkpoints. She feared the checkpoints to the extent she didn’t want to go to work anymore. Her employer told her that he’d be her guarantor if she marries him. She accepted, but he divorced her after approximately three months. NarrID 195: Lebanese Married man in Tripoli, in Lebanon for > 7 years

In many narratives, the presence of a man, be it the father, brother, or husband, was viewed by fathers and men in Tripoli as a source of financial security for Syrian girls. When marriages ended, the absence of a husband again contributed to financial hardship, particularly if the marriage had produced children. Another participant shared the following story where the loss of a Syrian girl’s father increased the family’s vulnerability to poverty, requiring her to work. Marriage was perceived as a financial benefit.

There was a girl, she had her mom and dad and three other sisters. Because of the events in Syria and because of the loss of her father, they had to come to Lebanon. In Lebanon, they do not have a backbone, so the girls had to stop school and work in houses and in several other places. So the girls were slandered. The girl had a marriage proposal but from a guy from another sect. But due to their financial status and poverty she had to marry him, because he had a good financial status. After a while, she delivered a baby girl, her husband left her and the baby. He said he was in Lebanon, but he travelled outside, and he never recognized neither his wife nor his daughter. The girl had to return back to working and return to the humiliating and poverty phase she was living in before. NarrID 230: Married man married in Tripoli, unknown amount of time in Lebanon.

Other narratives highlighted the ongoing financial strains experienced by married Syrian couples because of limited employment opportunities and lack of documentation. One man in Tripoli explained as follows:

The husband is 15 years-old and the wife is 14 years-old. They are cousins… They got married because the husband is the family provider… as it’s practiced, it’s better for the girl to marry her cousin, and this way the husband will hold her responsibility. They got married and all live in the same house, with the husband’s mother. The husband works in construction like all other Syrians, as they say in slang language ‘one day work and ten days nothing’. And this is the story which drew my attention and I called it ‘underage in the golden cage’. NarrID 225: Married man in Tripoli, in Lebanon for 1–3 years

Syrian fathers in Tripoli also shared similar narratives. In the following example, a Syrian father identified the financial situation as limiting both his daughter’s safety and her access to education:

“I am a Syrian immigrant who moved to Lebanon. What is bothering me the most is the situation on education regarding girls. I had to stop my daughter from attending school because education is very expensive here in Lebanon, and the roads are not safe. If I want to send my daughter in a safe bus to school, then I have to pay more money and I cannot afford that. This makes me very sad. NarrID 404: Syrian Father in Tripoli; in Lebanon for 1–3 years

3.3 III. Perceived overprotection of girls among Syrian girls and mothers in Tripoli

As shown in Fig. 4, when asked about protection, responses by Syrian girls in Tripoli were more likely to be evenly distributed across the spectrum and more likely than respondents in other groups to indicate that the girl in the shared story was ‘protected too much’. Syrian mothers in Tripoli had a similar distribution and were also more likely than others to indicate that the Syrian girl in the shared narrative was be ‘protected too much’ in comparison to all other groups.

Fig. 4
figure 4

Dyad asking whether the girl in the shared narrative was ‘protected too much’ or ‘not protected enough’

Full size image

Girls and mothers in Tripoli were more likely to indicate that the Syrian girl in the story was protected too much as highlighted in red. The * indicates the mean response for each subgroup and again highlights that girls and mothers in Tripoli were more likely to perceive that the Syrian girl in the story was overprotected. Differences were statistically significant for Tripoli girls (p < 0.001) and Tripoli mothers (p = 0.001).

When girls in Tripoli spoke of too much protection, they referred to facing restrictions such as not being able to go to school, not being able to leave the home unaccompanied by a male chaperone, and a lack of freedom primarily related to safety concerns in the community. Their perceptions of protection were often related to oppressive forces or gender-based violence. The following participant shared how she and her sisters were not permitted to leave the home out of concern for their safety:

I was married when I was 17 years old, and I have a child. I met my husband in Sabra. We got to know each other for a week only, and then we got married. We have been going through a lot of marital problems. My room is very small, and he pressures me a lot. My sisters, 14- and 15-years old, were forced by my parents to get married. Our parents imprisoned us at home; we were not allowed to go out. I wanted to get married in order to run away from my parents’ treatment, but marriage was even worse. NarrID 1058: Married Syrian girl in Tripoli, in Lebanon for 1–3 years

An unmarried Syrian girl, also identified ‘protection’ as a prison that limited her access to education:

When I was in Syria, I reached the sixth grade in my education. However, in Lebanon, I had to live in camps and my uncles refused to allow me to attend school or get out of the tent. In other words, I was imprisoned in camp. I am not living happily. Currently, I like a man who is 23 years old, and if he proposes, I will say yes for two reasons: one, I like him and I want to marry him. Two, I want to leave the prison that my family is forcing me into. NarrID360: Unmarried Syrian girl in Tripoli, in Lebanon for 3–5 years

In both narratives, early marriage was perceived by the girls as a potential escape from ‘too much protection’ by families. The following participant also decided to marry after her parents prevented her from going to school:

I displaced from Homs when I was 14 years old. In Lebanon, my parents prohibited me from going to school. A man proposed to marry me. I didn’t know anything about him, but we got married and had a daughter. I got pregnant the 2nd time by mistake, and my husband wanted me to have an abortion because he thought that it’s a girl. I tried to miscarry the child, but I didn’t succeed. When we did an ultrasound, we found out that it’s a boy. My husband was very happy. Now, I am only 16 years old, and I have two children. NarrID 1058: Married Syrian Girl in Tripoli, in Lebanon for 1–3 years

Similarly, Syrian mothers in Tripoli were more likely to indicate that girls were ‘protected too much’. For instance, in the following narrative, the family opted for marriage as a form of protection, despite the mother not really supporting it:

In Syria, we were safe and comfortable. When we displaced to Lebanon, we lived in a camp. I was compelled to wed my daughter at the age of thirteen. I feared for her safety. NarrID 359: Syrian Mother in Tripoli, in Lebanon for 3–5 years

Another Syrian mother in Tripoli shared how the protection of her daughter had become the responsibility of the family and significantly limited the girl’s freedom:

Here the atmosphere has changed. In Syria she used to move around, go to school, her friends, her uncles and cousins’ place. Here I can’t let her go out like that—it’s not safe. If she wants to leave she has to be accompanied by someone, her dad or her siblings. In school everyone joins her, if she’s alone she doesn’t go anywhere. NarrID 876: Syrian Mother; in Lebanon for 3–5 years

It is important to note that while Syrian mothers in Tripoli identified their daughters as being ‘protected too much’, many narratives did speak to the sexual and gender-based violence risks faced by Syrian girls confirming pervasive concerns for the girls’ safety:

She used to attend school in Syria but when we came to Lebanon, we were so scared that something might happen to her, we banned her from getting an education. She would see all of her friends going to school and leaving the house, but we would not allow it. One time, during the evening, as she was going down the stairs, a man sexually harassed her, and since then, we have become even more protective and have forbidden her from doing any activity. NarrID169: Syrian Mother in Tripoli, in Lebanon for 3–5 years

The shared perception among girls and their mothers of ‘too much protection’ often referred to limited opportunities for education and restricted freedom of movement resulting from safety concerns in the community. The increased control and restrictions of Syrian girls’ freedom as protection likely stemmed from a perceived greater risk to girls, who were recognized as being at risk of sexual harassment as well as sexual and gender-based violence especially in the public sphere.

4 Discussion

In the current study, we present mixed-methods data to understand why the experiences of Syrian refugee girls in Tripoli may differ from those of Syrian refugee girls in Beirut and Beqaa. From 464 participants in Tripoli, we observed three notable variations in responses between those from Tripoli and other areas: (a) Syrian fathers in Tripoli felt there was a need for more programs/services to assist Syrian girls, (b) men from Tripoli indicated that financial instability influenced the experiences of Syrian girls, and (c) both Syrian girls and their mothers in Tripoli believed the girls were overly protected. This more nuanced appreciation of the unique experiences and needs of Syrian refugees in Tripoli is critical for informing programs, services, and policies aimed at supporting Syrian refugee girls and improving their well-being and outcomes.

The vulnerabilities and challenges experienced by Syrian girls in Tripoli mirror those reported in other settings impacted by armed conflict and migration. For instance, our finding that Syrian girls in Tripoli faced sexual harassment and exposure to violence resonates with a systematic review on gender-based violence during complex emergencies [30]. Furthermore, many Syrian girls in Tripoli struggled to continue their formal education due to security and economic challenges, a phenomenon similarly highlighted in a literature review concerning educational access in crisis-stricken areas [31]. Previous research has also underscored the economic hardships encountered by girls affected by crises. In conflict and post-conflict situations, girls often assume labor roles, sometimes in unprotected informal sectors, placing them at heightened risk [32, 33]. While our primary findings align with previous research, our study's pivotal insight lies in the substantial differences in experiences and perceived needs among refugee girls within the same host country, Lebanon. Such intra-national variations, often overlooked, bear significant consequences for strategic planning and response.

It is important to contextualize the results. Tripoli, located on Lebanon's northern coast, is a conservative society with deeply rooted traditions. It has seen less urban development than Beirut and less population diversity with a largely Sunni Muslim population. While historically a center for trade, in recent years, Tripoli’s economy has faced significant challenges. Security-wise, the city has experienced sporadic clashes often linked to local politics or socioeconomic issues. Relevant to the current analysis, Tripoli has accommodated a large influx of Syrian refugees, which has exacerbated social and economic tensions. This contrasts with regions like Beirut, which is more cosmopolitan, and Beqaa, known for its agricultural richness and proximity to the Syrian border, making it a significant hub for refugees.

Syrian fathers in Tripoli indicating a need for programs and services may derive from increased financial strain in Tripoli or from a gap in available programming through government and non-governmental sources. Alternatively, perceived greater insecurity in Tripoli could have contributed to these results. Given that Syrian fathers are often perceived as a source of financial and physical security for their families [34], with more financial strain and unsafe conditions in Tripoli, Syrian fathers may have understandably prioritized programs and services over friends and family as well as community and religious values, to address their family's unmet needs. It is important to note that Tripoli’s basic urban services were already at capacity before the Syrian crisis. The influx of Syrian refugees resulted in further strain on the already scarce and inadequate infrastructure [13].

With respect to district-specific distribution of multi-purpose cash assistance in January 2016, Tripoli had the lowest proportion of eligible people receive the grants with only 37.1% receiving cash assistance compared to 46.2% in Beirut and 51.2% in West Beqaa [35]. The mismatch between need and available programming is possibly the reason Syrian fathers were more likely to identify a need for programs and services. Furthermore, refugee men are often made invisible within the distribution of humanitarian aid and resources which perhaps furthers the perceived need [36]. The availability of resources also affects the sense of protection and safety within Tripoli. For instance, beneficiaries of a multi-purpose cash assistance program reported feeling eight times more secure and having five times more trust in their host communities, compared to non-beneficiaries.

Tripoli has experienced decades of urban poverty, hosting a disproportionate 46% of Lebanon’s extremely poor population. While Tripoli is like Beirut in that refugees are living in an urban context, it is also like Beqaa in that it has some of the highest rates of poverty in the country [37]. It is perhaps the overlap of these factors, higher rates of and more extreme poverty, in conjunction with the higher costs of living associated with urban centres, that sets Tripoli apart in the current subnational analysis. For example, Syrian refugees in Tripoli compete for jobs in a limited and oversaturated labour market and have been found to have the lowest average monthly incomes, lowest employment rate for Syrian women, and the longest working hours per week compared to other regions in Lebanon [38].

In their shared narratives and in the literature, female-headed households are often presumed to be especially vulnerable by virtue of the family arrangement [39]. The absence of men is often framed as a factor in financial vulnerability. This perception was shared among Syrian refugee women who stated that men in their families provided financial, social, and physical protection, both in this study and in other reports [2]. Furthermore, Syrian men displaced to the Netherlands were found to construct their masculinities in relation to labour market access and paid employment [40]. This may also be true for men in Tripoli due to the limited opportunities to earn a livable wage. Other needs for Syrian girls, including protection and education, were often also framed with a financial lens as families could not afford education, or their protection was compromised by unsafe employment conditions. Despite a clear need for financial support among Syrian refugees in Tripoli, programs aimed at financial support, improved livelihoods, and/or employment opportunities were lacking in the region.

The shared perception among girls and their mothers of ‘too much protection’ often referred to limited opportunities for education and restricted freedom of movement resulting from safety concerns in the community. The increased control and restrictions of Syrian girls’ freedom as protection likely stemmed from a perceived greater risk to girls, who were recognized as being at risk of sexual harassment as well as sexual and gender-based violence, especially in the public sphere. Tripoli has continued to face security challenges due to its historical ties with Syria and the fallout from the Syrian conflict, leading to sectarian tensions. For instance, armed clashes in 2012–2013 mirrored Syria’s sectarian conflict and Syrian refugees in Tripoli experienced insecurity and became targets due to a security plan in 2014 [16]. The arrival of more Syrian refugees further strained the city’s resources, causing competition for jobs and tensions over resources, resulting in negative relationships [3, 14, 14,15,16].

In the face of these challenges and barriers, it is not unexpected that Syrian families in Tripoli might have greater fears for their daughters’ safety. However, our current analysis introduces important new insights regarding the perceptions of Syrian girls and mothers in this context, namely that from the girls’ perspective, over-protection was perceived negatively and as an oppressive force, restricting their freedom. These constraints in turn limited their access to education, opportunity to socialize with peers, and their ability to access programs and services. Other reports have similarly found that these barriers were so great that Syrian refugee parents often felt the need to keep their kids at home or to send their kids to different schools out of safety concerns [13].

4.1 Strengths and limitations

The results must be interpreted in the context of the study’s limitations. First, the findings are not generalizable due to the non-representative sampling. It is possible that some subgroups of Syrian families were not reached during recruitment and girls less than age 13 were excluded altogether for ethical reasons. Additionally, recruiting Syrian girls directly was difficult as they were seldom found in public places like shops and cafes where we engaged participants. Consequently, men are disproportionately represented in the sample. Second, SenseMaker collects micronarratives that are briefer than traditional qualitative interviews or focus groups since there are no follow up probing questions asked by the interviewer. As a result, the narratives are typically less detailed and may be less rich than these other qualitative data sources. Furthermore, the translated narratives available for qualitative analysis were limited to those that were about child marriage which may not be representative of the other narratives. Finally, since this was a secondary analysis, some key variables such as socioeconomic were not included since they had not been captured in the original survey.

The study also has some noteworthy strengths. For example, since SenseMaker avoids direct questioning, experiences tend to emerge more naturally from participants’ broader lived experiences. Moreover, through the inclusion of both men and women as well as Lebanese host community members, Palestinian and Syrian refugees, the results reflect a diverse range of perspectives. Since SenseMaker empowers participants to interpret the experiences shared in their micronarratives, the analysis was able to identify several nuanced differences (such as those for Tripoli) that would have otherwise likely been missed.

5 Conclusions

This subnational analysis identified significant differences in participant responses in Tripoli versus Beirut and Beqaa in three key realms: fathers in Tripoli perceived more of a need for programs and services, men in Tripoli thought financial security had more of an impact on Syrian girls, and girls’ safety in Tripoli was a concern for men but both girls and women perceived that girls were overprotected despite the security concerns. The intersectionality of the three realms remains evident. The protection and safety of Syrian girls is likely linked closely to the family’s financial security and the availability of programs and services. If families lack the necessary resources to protect their daughters such as safe access to education, healthcare, and identification cards, they may be more protective as a result. The increased protection sometimes resulted in gender-based violence such as child marriage, and restricted access to education and social recreation.

These different perspectives in Tripoli highlight that the experiences, concerns and needs of Syrian refugees in Lebanon are not homogenous across locations. Localized understanding of these experiences can guide the development of tailored approaches in aid and service provision to better address the needs and concerns of refugee families. For instance, in Tripoli, there is a need for additional programs and supports for Syrian girls, including programs that will contribute to financial security. While safety risks are real, including sexual harrassment and gender-based violence, this insecurity leads to the restriction of girls’ movement and a loss of freedom to attend school, engage with peers, and so on. Programming for Syrian girls must therefore take these circumstances and concerns into account and must consider things like providing safe transportation and access to safe education. As highlighted in this analysis on the experiences of Syrian girls in Tripoli, a better understanding of the contexts and needs at a subnational level would better inform both the distribution of resources and the types of assistance most needed by communities facing varied challenges.