Remember me
ARTICLE
Year : 2022 | Volume
: 20
| Issue : 1 | Page : 14-27
Violence and traumatic exposures among islamic high school students in thailand's subnational conflict
Mahsoom Sateemae1, Tarik Abdel-Monem2, Suhaimee Sateemae3, Abdullah Uma4, Denise Bulling5
1 PhD, MHSc, Faculty of Education, Fatoni University, Pattani, Thailand
2 JD, MPH, University of Nebraska Public Policy Center, University of Nebraska, Lincoln, NE, USA
3 MIS, Satree Islam Vitaya Mulniti School, Yala, Thailand
4 PhD, Faculty of Liberal Arts and Social Sciences, Fatoni University, Pattani, Thailand
5 PhD, University of Nebraska Public Policy Center, University of Nebraska, Lincoln, NE, USA
Correspondence Address:
Tarik Abdel-Monem
University of Nebraska Public Policy Center, University of Nebraska, Lincoln, NE 68588-0228
USA
Source of Support: None, Conflict of Interest: None
DOI: 10.4103/INTV.INTV_2_21
Since 2004, nearly 7,000 people have been killed in Thailand's subnational conflict in its southernmost provinces – one of the longest running domestic insurgencies in Southeast Asia. This study assesses exposure to conflict-related trauma among a sample of high school students (n = 419) in Islamic private schools within the conflict-affected area. Responses to the Posttraumatic Stress Disorder Check List 6-item version indicated that 18.6% of the students had symptomology consistent with probable posttraumatic stress disorder (PTSD). A survey of traumatic events specific to the conflict area was constructed, identifying the types and frequency of exposure to conflict-related traumatic events among our sample. Correlation analysis showed significant associations between exposure to trauma, with PTSD symptomology, life satisfaction, happiness and perceptions of neighbourhood security. Results of this study suggest that many youth in the region experience continuous and multiple forms of trauma that are detrimental to emotional wellbeing, heightening the need to provide protective interventions.
Keywords: conflict, Islamic school, minority education, Muslim, Muslim minority, Muslim student, PTSD, Thailand, trauma
Trauma typically plagues civilian and military survivors of conflict (Al-Krenawi et al., 2009; Hamama-Raz et al., 2008; McAloney et al., 2009). It is estimated that up to 350 million children around the world live in areas currently affected by violent conflict (Save the Children International, 2018). A large body of literature has investigated the detrimental impacts of violent conflict and trauma on youth in a variety of contexts, including Bosnia (Goldstein et al., 1997; Smith et al., 2002), Israel (Kasler et al., 2008; Pat-Horenczyk et al., 2007; Schwarzwald et al., 1993), Kuwait (Hadi & Llabre, 1998; Llabre & Hadi, 1994), Lebanon (Bryce et al., 1989; Karam et al., 2008; Macksoud & Aber, 2008), and Palestine (Khamis, 2005; Lavi & Solomon, 2005; Thabet et al., 2002). The full and lasting effects of conflict-related trauma on youth are difficult to measure. Much of the existing research on trauma affecting youth in conflict is cross-sectional in nature and captures point-in-time impacts and trauma symptomology. The core attributes of PTSD as defined in the American Psychiatric Association's Diagnostic and Statistical Manual of Mental Disorders (DSM-V) are sustained re-experiencing/re-living of the traumatic event(s), social and/or emotional detachment or numbing, hyperarousal, and other behaviours that are detrimental to wellbeing (American Psychiatric Association, 2013). It is possible that long-term mental health conditions from exposure to trauma-related conflict may develop into adulthood, depending on the nature of the conflict and its scope of violence, individual characteristics and exposures to traumatic events, and the availability of protective factors (Betancourt et al., 2013; Dimitry, 2012; Osofsky et al., 2015). Only a small body of longitudinal works exists which examine the long-term effects of traumatic exposure to conflict among youth (Gvirsman et al., 2016; Sharma et al., 2017). This is likely due to difficulties with conducting research in war-time or post-conflict environments. Nonetheless, research has found that exposure to chronic conflict-related violence among youth can lead to a wide variety of short- to medium-term effects on wellbeing (Mollica et al., 1997; Smith et al., 2002). However, impacts on youth are highly dependent on the specific context and experiences within that conflict (Sagi-Schwartz, 2008).
This study assesses experiences of conflict-related trauma and PTSD symptomology among Islamic high school students in the deep south of Thailand. Since 2004, Human Rights Watch (2020) has estimated that over 7,000 individuals have been killed in this subnational conflict, around 90% being civilians. There have been numerous works examining the conflict from historical, sociocultural or policy contexts, mainly by regional studies or security analysts (Abuza, 2011; Jitpiromsri & McCargo, 2010; Joll, 2010; McCargo, 2010). However, there are few studies on traumatic experiences and impacts of the Southern Thailand conflict on youth – many of whom are victims, some of whom have perpetrated violence, and all of whom have now lived for over 15 years in a state of chronic, low-intensity conflict.
Background on the Southern Thailand Conflict
The fundamental origins of the Southern Thailand conflict are political, but have assumed an ethnoreligious character due to the confrontation between the ethnically Thai, majority Buddhist society and central government, and the predominantly Malay and Muslim residents of the area. The three provinces were previously a Malay–Muslim sultanate prior to its annexation by Siam (Thailand) in the early twentieth century (Puaksom, 2008; Virunha, 2008). Following annexation, successive central governments imposed a combination of aggressive security and assimilationist policies in the region. Under the dictatorship of Field Marshall Pibulsongram in the 1940s, the central government forced conversions from Islam to Buddhism, abolished Islamic family and inheritance laws, prohibited Islamic cultural practices such as banning headscarves, replaced Islamic educational institutions with Thai educational requirements, and other measures designed to suppress Malay–Muslim identity and “modernise” the nation (Aphornsuvan, 2003, 2008; Croissant, 2007; Jory, 2007; Tissamana, 2015). These policies were met by local resistance and rebellions to central rule by Malay–Muslim groups in the region (Islam, 1998; Kraus, 1984). These assimilationist policies were generally relaxed with the demise of Thailand's ethno-nationalist regimes after World War II, and gradual adoption of liberal-democratic values (Forbes, 1982; Yusuf, 2010). Although Thailand maintains central governance over the three provinces, since the 1980s, Thai governments have offered conciliatory policies to the Malay–Muslim community, better integrated local voices into political affairs, improved attempts to address long-standing grievances, and implemented economic development and social policies supportive of the region and its residents (Ockey, 2008; Storey, 2008; Tamthai & Boonchoo, 2009).
Despite these constructive developments, a major increase in violence began in 2004 with a series of high-profile incidents, including the raid of a military armoury by antigovernment insurgents, storming of a mosque by soldiers and suffocation deaths of nearly 80 Muslim prisoners in military custody (McCargo, 2006; Satha-Anand, 2006). Shortly thereafter, martial law and aggressive security policies were imposed in the three provinces, a development which is widely regarded to have exacerbated the situation (Kazmin, 2007; Phongpaichit & Baker, 2010; Tan-Mullins, 2009). Since then, prevalent forms of violence have included attacks on government schools and educators, attacks on Buddhist and Islamic places of worship, drive-by shootings, and motorcycle or car bombs (Human Rights Watch, 2007, 2010). Because few claims are made after attacks, attributing responsibility to the violence has been difficult. Commentators believe the perpetrators are a mix of antigovernment Muslim insurgents, military and police acting outside the authority of law, civilian vigilantes, and organised criminal networks tied to drug trafficking or corrupt business interests (Askew, 2008, 2010; Fuller, 2012; Küng, 2018; Liow & Pathan, 2010; Pathamanand, 2007). Uniformed and nonuniformed government forces or their allies have been accused of torture and extrajudicial killings, unlawful detentions, and acting with general impunity (Human Rights Watch 2015a, 2015b). Since 2013, formal peace talks between the Thai government and antigovernment actors believed to represent insurgent groups have frequently stalled and have yet to reach meaningful breakthroughs.1
Review of Literature
Since the outbreak of violence in 2004, there have been a handful of studies assessing trauma and/or prevalence or symptomology of PTSD among residents of this area. However, this body of research has been growing. Wichaidit (2018) conducted an overview of studies examining mental health and trauma among residents in the area, including both youth and adults, and found widely varying rates of trauma among populations examined. Jayuphan and colleagues (2020) assessed PTSD symptomology among primary level students in the affected areas (n = 972). They found that 30% exhibited PTSD symptoms. Tokhani (2011) assessed behavioural problems among adolescents from communities in the area and found that those living within high alert zones had increased levels of behavioural problems compared to those who did not. Other notable works include Jinpanyakul and Putthisri's survey (2018) of high-school students in Narathiwat province (n = 341), in which they found associations between conflict-related mental health status and perceived quality of life. Panyayong and Juntalasena (2009) assessed traumatic experiences and PTSD symptomology among students in the three southern provinces of Thailand (n = 2,884) and found a symptomology rate indicating possible PTSD of 21.9% and prevalence of other behavioural problems. Panyayong and Lempong (2013) assessed traumatic experiences and PTSD symptomology among children in the three southern provinces (n = 198) and found a much lower rate of possible PTSD of 7.8%. Phothisat (2013) assessed PTSD among children of police officers deployed in the region, and found PTSD symptomology rates of 18%, driven by family-related problems and stress associated with the conflict. Wiwattanaworaset and Pitanupong (2015) administered a variation of the 12-item General Health Questionnaire among witnesses to a bombing one month and six months after the incident, and found suspected PTSD symptomology of 23% and 15%, respectively. Yongpitayapong and Wangthong (2014) assessed PTSD symptoms among Muslim students (n = 57) after the director was assassinated in front of the school. They found that 14% of the students had PTSD symptoms one month after the event. Overall, this body of research suggests that heightened levels of trauma are common among residents of the area. However, the violence and security situation has hindered the ability to collect more extensive research data. Continued documentation and studies are needed to further understand the development and characteristics of trauma among area residents to develop suitable public mental health interventions.
Research Approach and MethodsSample Population
This study builds on prior assessments of trauma among youth in the Southern Thailand conflict by focusing on students enrolled in Islamic high schools (n = 427). The population sampled were senior students at four Islamic private high schools in the region, traditionally known as ponok. Ponok are private Islamic schools that primarily cater towards the Malay–Muslim community in the region. Many though not all ponok schools are officially recognised by and receive funding from the Thai government, with the requirement that they teach a combination of the Thai national curriculum subjects, religious (Islamic) courses, and vocational training (Liow, 2009; Sateemae et al., 2015).2 Thai government statistics indicate there are approximately 197 such Islamic private schools in Thailand, with over 157,927 students enrolled (Office of the Private Education Commission, n.d.).3 The four schools in our sample were selected on the basis that they were all in the same general geographic proximity within two adjoining districts of the same province.
Research Approach
We utilised the PTSD Check List 6-item version (PCL-6) to assess symptomology of PTSD among our sample. Additionally, we were interested in mapping types of trauma specific to the Southern Thailand conflict as experienced by our sample. We developed a checklist of 25 traumatic items believed to be characteristic of the conflict, and examined how frequently they were experienced among our sample. Additionally, we used confirmatory factor analysis to model the individual trauma items into thematic categories and examined correlations between the frequency of experiencing trauma within these categories with PTSD symptomology derived from the PCL-6 scores, perceptions of life satisfaction and happiness, and perceptions of overall neighbourhood security.
We chose to use confirmatory factor analysis because this approach allows the researchers to identify factor constructions within data to test a theory or question (Cole, 1987; Suhr, 2006). We were interested in examining how the frequency of different types of traumatic experiences impacted the overall wellbeing of these students, particularly as this stage of violence has continued since 2004 and is pervasive throughout the three provinces of the conflict-affected region. Given the broad scope of the violence, confirmatory factor analysis allowed us to test a factor structure correlating PCL-6 scores with the frequency of exposure to traumatic events in four different categories: direct physical threats or harm to a person (e.g., personally experiencing a physically violent traumatic event such as being attacked or shot at), less physically proximate witnessed traumatic events (e.g., witnessing or being exposed to a physically violent traumatic event, such as hearing shooting or bomb blasts, or seeing someone else being harmed), and incidents involving a person's social network that may indirectly traumatise a person (e.g., having a family member or friend exposed to a physically violent traumatic event, such as having a family member or friend injured or killed). The fourth category of trauma items included items related to personally invasive intrusiveness or indignities, such as displays of authority to humiliate or disrespect a person or their religion – experiences which are commonly reported in ethnic conflicts (Dar & Deb, 2021; Giacaman et al., 2007). These parameters were defined through insights provided by group consultations with students and teachers in developing the trauma checklist, continuous reporting about the conflict in both Thai and international news media, and the authors' personal observations.
Measures
We developed a Southern Thailand Trauma Checklist that was similar in structure to the Gaza Trauma Checklist (Altawil et al., 2008; Thabet & Vostanis, 1999; Thabet et al., 2008) and Kashmir Conflict Exposure Checklist (Dar & Deb, 2021) used to measure occurrence of traumatising incidents in conflict zones. Our Southern Thailand Checklist was developed by the authors in consultation with mixed-gender peers from the three southern provinces to validate its content. We established the conceptual parameters of the checklist as potentially traumatic stressors salient to the conflict that were commonly experienced by residents of the area. After identifying the conceptual domain, draft items were developed based on reviews of violent incident data about the conflict collected by Prince of Songkhla University's Deep South Watch centre, local and international news media that regularly report about the conflict, and instruments developed by other researchers considered relevant to the Southern Thailand context (e.g., Altawil, 2008; Roberts et al., 2009). Similar to Altawil's Gaza Checklist (2008), simple “yes” or “no” response options were used to simplify the instrument for respondents. A draft Southern Thailand Trauma Checklist was translated and back-translated from Thai to English to ensure understandability. Three of the authors facilitated a discussion with mixed-gender youth and adult representatives (high school students and teachers) from the population of interest to revise the content for the checklist, probe for understandability of the draft items, discuss their applicability, and identify other potentially relevant themes or considerations. An inductive approach was used to eliminate items deemed unrelated to our conceptual parameters (Haynes et al., 1995). Two of the authors then reviewed remaining items for inclusion in the checklist in which a Cohen kappa of 1 was obtained, resulting in 25 items. The final checklist prompts respondents to indicate whether or not they had ever experienced the 25 traumatic events related to the current political situation in the region, and if so, estimate the number of times the events had been experienced. The Cronbach alpha coefficient to assess internal consistency relative to their sum score was 0.715 for the 25 items.
Trauma symptomology was assessed by using a 6-item abbreviated version of the PCL-6. The PTSD checklist is a commonly used self-report or interview administered questionnaire that corresponds to DSM criteria for PTSD symptomology (Blanchard et al., 1996; Weathers et al., 1993). The 6-item abbreviated version of the PCL asks respondents to indicate to what extent they have experienced the following symptoms or behaviour within the previous month: (1) “Repeated, disturbing memories, thoughts or images of a stressful experience from the past”; (2) “Feeling very upset when something reminded you of a stressful experience from the past?”; (3) “Avoided activities or situations because they reminded you of a stressful experience from the past?”; (4) “Feeling distant or cut-off from other people?”; (5) “Feeling irritable or having angry outbursts?” and (6) “Difficulty concentrating?”. Respondents identify the frequency of these six experiences on a 1–5 scale ranging from 1 (”not at all”) to 5 (”extremely”). The 6-item version of the PTSD checklist has previously been validated by Lang and colleagues (2005, 2012), who suggested a score of 14 or above on all six items as a threshold for possible prevalence of clinically significant PTSD. Utilising the14-score cut-off point, previous validation studies of the PCL-6 with other diagnoses tools have found acceptable psychometric rates. Lang and colleagues reported a sensitivity of 0.92 and specificity of 0.72 among a sample of primary care patients in a US Veteran's Affairs healthcare setting (Lang & Stein, 2005), and a sensitivity of 0.92 among a sample of US-based primary care patients in multiple states (Lang et al., 2012). Han and colleagues (2016) found a sensitivity of 0.98 and specificity of 0.43 among primary care patients in a US-based federally qualified healthcare centre setting. A Thai language version of the PCL has been previously used to assess trauma symptomology in multiple contexts, including among residents in the conflict-affected areas of the south (Jatchavala & Vittayanont, 2017), among urban firefighters in Bangkok (Khumtong & Taneepanichskul, 2019), following the Indian Ocean tsunami of 2004 (Udomratn, 2009), and during the COVID-19 pandemic (Srifuengfung et al., 2021). Our scale revealed a Cronbach alpha of 0.759, indicating a suitably high degree of internal consistency.
Questions about overall life satisfaction and happiness were derived from the World Values Survey Wave 6 (WVSA, 2014a). These items were included as confirmatory checks on our enquiry, as we would expect perceptions of subjective wellbeing to correlate negatively with experiencing trauma in conflict. Considerable research conducted in other contexts suggests that exposure to conflict-related trauma negatively affects one's subjective perceptions of wellbeing (Fox & Tang, 2000; Neuner et al., 2004; Shemyakina & Plagnol, 2013; Welsch, 2008). The life satisfaction item asked respondents to rate on a scale of 1–10 “All things considered, how satisfied are you with your life as a whole these days?”, with 1 being low satisfaction and 10 being high satisfaction. The happiness item asked individuals to rate their degree of happiness on a level of 1 (”very happy”) to 4 (”not at all happy”). These World Values Survey questions had previously been validated and used in Thailand in the Thai language (WVSA, 2014b).
Finally, we also employed another question from the World Values Survey to measure perceptions of neighbourhood security. This question asked respondents to indicate “How secure do you feel these days in your neighbourhood?” on a scale of 1 (”very secure”) to 4 (”not at all secure”) (WVSA, 2014b). We also asked this item as a confirmatory check, as we would expect there to be associations between perceptions of neighbourhood security and experiences of conflict-related trauma, with negative perceptions of neighbourhood security correlating positively with increased experiences of trauma.
Data Collection and Consent
School administrators provided permission to the researchers to administer a paper version of the survey. The survey included the trauma checklist, abbreviated PCL-6, the life satisfaction and happiness items, and neighbourhood security item. As permission was obtained from lead administrators at the participating schools, Fatoni University's administration approved the project. The surveys were anonymous and students had the option to opt out of the survey if they wished. In Thailand, the age of consent for participation in research is 18 (Ministry of Public Health, Thailand, 2021). The survey was sent home with all students for parents to review and return written consent. The students then completed the survey at home or during class times, and the completed surveys were then anonymously collected by school teachers. The survey response rate was relatively high, with 4274 adequately completed surveys collected from a total of 450 distributed (95%), most likely due to its in-class administration.
Survey ResultsGender and Socioeconomic Class
It should be noted that there was a pronounced gender imbalance within our sample, with 79% of respondents being girls and 21% being boys. The gender disparity among our sample is to some extent reflected in wider national trends, as in many areas of Thailand girls are significantly more likely to attend and complete upper secondary levels of school (Kantachote, 2013; World Bank, 2009). This disparity is particularly heightened in rural, poor areas of Thailand, as male dropout rates in secondary school are likely due to young men entering the workforce to generate income for families in manual-labour intensive jobs. Respondents were also asked to self-identify in terms of socioeconomic class. The majority of respondents indicated that they were lower income, with over 74% indicating they were “lower middle class”, 10% were “working class” and 10% were “lower class”, whereas only 3% indicated they were “upper middle class” and 2% identified as “upper class”. The average age of respondents was 18 (minimum to maximum range 17–20).
Types of Traumatic Experiences
Of the 427 respondents, a total of 419 indicated that they had experienced at least one traumatic experience (98%) in the 25-item Southern Thailand Trauma Checklist. Among the 419 students who reported traumatic experiences, the least frequently experienced item was “Have you ever been shot at?” (Yes = 1%) and the most frequent was “Have you ever heard a bombing?” (Yes = 88%). Chi-square tests for independence indicate that traumatic experiences significantly differed by gender at the P < 0.05 level for several items. Male students were more likely to have experienced the items: “Have you ever been injured by a shooting, bomb blast or physical attack?” (females yes = 3% vs. males = yes 6%), “Have any of your friends, neighbours or relatives been killed?” (females yes = 32% vs. males yes = 44%), “Have any of your friends, neighbours or relatives been injured?” (females yes = 23% vs. males yes = 44%), “Have you ever been arrested by authorities?” (females yes = 2% vs. males yes = 8%), “Have you ever been humiliated by authorities?” (females yes = 10% vs. males yes = 23%) and “Have you ever felt harassed by others because of your identity?” (females yes = 15% vs. males yes = 25%). Female students were more likely to have experienced only one item at a higher rate than males: “Have you ever felt very frightened for your safety?” (females yes = 71% vs. males yes = 45%).
A total of 419 out of the 427 respondents (98%) indicated that they experienced anywhere from one to 25 types of trauma. The mean number of different trauma types experienced was 6.12 [standard deviation (SD) = 3.459]. Among male students (n = 88), the mean number of different types of traumas experienced was 6.69 (SD = 3.873). Among female students (n = 330), there was a slightly lower mean in the number of different types of traumas experienced (m = 5.97, SD = 3.336). Over half of all students had experienced at least five types of trauma. Shapiro–Wilk tests for assessing normality were conducted for total, male and female students, and indicated non-normal distributions (P = 0.000 for all three samples). Although male students experienced a slightly higher average number of different traumas than female students (m = 6.69 vs. m = 5.97), the Mann–Whitney U test for non-normal distributions indicated that it was not at a significantly different level (U = 12803.0, P = 0.087).
Threshold PTSD Symptomology
Scores for the 6-item abbreviated version of the PCL-6 were summed to assess PTSD symptomology among our sample. Frequency analysis indicated that 18.6% (n = 78) of our sample of 419 students had a score of 14 or above on all six items, the suggested threshold indicating the possible prevalence of clinically significant PTSD (Lang et al., 2012). Among the sample that obtained threshold levels of PTSD, 21% were boys and 79% were girls, roughly corresponding to the gender representation of the overall sample.
Repeated Trauma Exposure
As noted above, respondents typically experienced an average of 5–6 types of different traumas on the Southern Thailand Trauma Checklist. The frequency of experiencing each different types of trauma is summarised in [Table 1], with the most frequently experienced event being “having heard a bombing”, whereas the least frequent was “having been injured by a bomb blast, shooting or physical attack”.
Types of Trauma and Correlations to PTSD Symptomology, Life Satisfaction, Happiness, and Perceptions of Neighbourhood Security
To examine the extent with which different types of traumatic experiences were related to PTSD symptomology and our other variables of interest, we first developed a model classifying types of traumas experienced into separate latent factors, and conducted a confirmatory factor analysis to assess goodness of fit. We utilised a four-factor model in which the trauma items were categorised into four latent categories: “Direct trauma” (personally experiencing a physically violent traumatic event), “Witnessed trauma” (witnessing or being exposed to a physically violent traumatic event), “Indirect trauma” (having a family member or friend exposed to a physically violent traumatic event), and “Intrusion/indignity” (experiencing an intrusive or humiliating event, or active displays of disrespect or ethnic/religious prejudice). Confirmatory factor analysis was completed using the MPlus application (Muthén & Muthén, 2019) and weighted least square mean and variance adjusted. After adjusting for unique variances, residuals and modification indexes, items with low factor loadings or significant overlap were eliminated. “Have you ever heard a bombing?” and “Have you ever heard a shooting?” were highly correlated and combined. Three commonly used fit indexes suggested by scholars were employed to assess goodness of fit, including the comparative fit index (CFI; Bentler, 1990), Tucker–Lewis index (TLI; Tucker & Lewis, 1973) and root means square error of approximation (RMSEA; Steiger, 2000). For categorical data, scholars have suggested thresholds of CFI > 0.95, TLI > 0.95, RMSEA < 0.06 (Hu & Bentler, 1995) to assess global fit. Our model showed a CFI of 0.96, TLI of 0.95, and the RMSEA estimate was 0.019, with a probability of being less than 0.05 being 100%, indicating a good fit for our sample data. A diagram of the model with the four latent factors, correlations and standard error; and the factor loadings and standard error parameter estimates for the trauma variables is presented in [Figure 1]. Two-tailed P-values indicated a significant relationship between all observed variables and latent categories at P < 0.01, with the exception of, “Have heard a shooting or bombing?” at P = 0.158. It should be noted that two items had low factor loadings: “heard shooting or bombing” (0.232) and “had a friend, neighbour or relative killed” (0.366). These items should be treated with caution. The four latent factors, factor loadings and standard error for parameter estimate and two-tailed P-value for each variable are presented in [Table 2].
We used bivariate correlation analysis to identify any associations between these four categories of trauma with PTSD symptomology via the PCL results, the self-ratings of overall satisfaction with life and happiness, and perceptions of neighbourhood security. Correlation results for these categories and variables are presented in [Table 3]. As we would expect, all four categories of trauma were highly correlated with each other and positively correlated with the PCL scores at significant levels (P < 0.05). The intrusion/indignity category had the strongest positive correlation with PCL scores (0.278), notably higher than the other three categories of trauma (0.153, 0.171 and 0.172). This seems to suggest that although intrusion/indignity events do not include physical injury or contact with a person, they are still a distressing form of trauma that possibly capture the identity-based grievances behind the conflict (e.g., harassed because of identity, religious beliefs are disrespected).
Table 3: Correlations between Trauma Types, PTSD Symptomology and Perceptions of Life Satisfaction, Happiness, and Neighbourhood SecurityAll four types of trauma were also correlated negatively with overall life satisfaction at significant levels, indicating that greater experiences of trauma were associated with lower overall life satisfaction. Three types of trauma (witnessed trauma, indirect trauma, and intrusion/indignity) were also correlated with happiness at significant levels, such that greater experiences of trauma were associated with reduced happiness. Finally, all four types of trauma were significantly correlated with security in one's neighbourhood, such that greater experiences of trauma were associated with reduced feelings of security. There were also moderate to strong levels of correlation between perceptions of security in one's neighbourhood with happiness (-0.525) and life satisfaction (0.329), suggesting that feelings of security and safety are salient to one's emotional wellbeing.
Although the categories of traumatic experiences were all correlated at small to moderate ranges with our variables of interest, among all four trauma categories the intrusion/indignity category had the highest levels of correlation with life satisfaction, happiness, and perceptions of neighbourhood security. Similar to its relevance to the PCL scores, this suggests that the intrusion/indignity category of traumas may have had a particularly salient impact among the youth in our sample. [Table 4] (available in supplementary materials) summarises the correlations between all the items.
DiscussionOverall Results
Among our study sample, we found that 18.5% of respondents exceeded the PCL-6 threshold level for symptomology, indicating potential prevalence of PTSD, an amount comparable to that found among students by Panyayong and Juntalasena (2009). In almost all categories of our Southern Thailand Trauma Checklist, male students reported experiencing more forms of trauma than females. All students experienced an average of about six different types of traumatic events at some point. By far, the most frequently reported types of trauma were those experienced indirectly in the community (e.g. hearing a bomb blast or shooting), whereas being directly and physically harmed was rare.
Not surprisingly, our model indicates that experiencing all four categories of trauma (direct, witnessed, indirect, and intrusion/indignity) significantly and negatively affect either life satisfaction, happiness, or both. Experiencing these traumas is correlated with increased PTSD symptomology and decreased overall perceptions of security in the neighbourhood. Although correlations should not be interpreted as directly causal, the consistency in relationships suggest that more frequent exposure to trauma is related to a general sense of insecurity and negative emotional wellbeing.
It is again worth noting that the intrusion/indignity category of traumas – despite not involving physical contact or injury – had the highest correlations with our variables of interest. This suggests the possibility that encounters with security forces perceived as intrusive, degrading, or disrespectful to one's identity or religion are as salient as those involving physical risk or injury. Such encounters likely reinforce ethnic and religious divisions that influence or heighten the Southern Thailand conflict. Connections between religiously affiliated violence and perceived injustices have been discussed at length in theoretical and research contexts (Doosje et al., 2013; Moghaddam, 2005; Tausch et al., 2011). Studies conducted in Muslim minority/non-Muslim majority contexts have examined and found that perceptions of threat or unfair treatment play a significant role in consolidating group identity, political and social attitudes and behaviour, and potentially antisocial or violent activity (Lyons-Padilla et al., 2015; Kunst et al., 2012; Pew Research Center, 2006), as commonly found in many multicultural societies regardless of ethnic, racial, or religious character.
Relevance to Interventions
Although we did not investigate how conflict-related trauma among our sample translated to political attitudes towards the Southern Thailand conflict, or participation and/or support for antisocial or violent activity, we may assume that the experience of identity-based trauma contributes to the conflict's ethnic and religious tensions. Thus, one clear policy implication of our findings is to not only continue support for a formal peace process between the Thai government and antigovernment actors, but also encourage all parties and civil society actors to address ethnoreligious grievances. There is an important role for multifaith leaders and community institutions to promote multiculturalism and nonviolence, particularly among youth. Many grassroots and internationally supported civil society organisations, conflict survivor groups, media organisations, educational institutions, Buddhist and Muslim religious leaders, and Thai government and military initiatives have and continue to promote interfaith and interethnic relations, peaceful dialogues, and human development projects throughout the conflict-affected provinces (Minority Rights Group International, 2018; Panjor, 2018). These efforts should continue on a community level. Likewise, both formal and informal activities to promote community connections and address wider human security issues are keys. Studies have begun outlining human and community factors within the conflict-affected areas that may promote resiliency among residents (Chamratrithiron et al., 2020; Ford et al., 2021).
To more directly implement public mental health interventions, it is likely that a variety of approaches are needed to address trauma arising from acute conflict-related events, as well as the daily stressors of living in a conflict-affected area (Miller & Rasmussen, 2010). Both the news media and studies indicate that area residents commonly face a variety of threats to human security and wide-ranging behavioural health concerns (Abdel-Monem et al., 2020; Chamratrithiron et al., 2020; Jayuphan et al., 2020; Panyayong & Juntalasena, 2009; Sateemae et al., 2015). Culturally relevant and community-based services are recommended for addressing trauma in conflict or postconflict environments using public health approaches (de Jong, 2002; Jegannathan et al., 2015). Thai governmental authorities should increase funding for Islamic private schools to engage as hubs in these efforts, as well as improve overall educational and infrastructure resources for registered schools in ways that may promote safety and security for students and provide globally competitive education. For example, increasing meal programmes and overnight boarding opportunities for students at school would directly provide shelter for students, many of whom are lower socioeconomic status and may reside or travel through high violence areas on public roads, where attacks and bombings are likely. Increased government support for Islamic private school physical and cyber infrastructure (high-speed internet, information technology, etc.), student financial aid, staff salary, training and other resources would greatly enhance educational outcomes. The implementation of such large-scale structural approaches and policies deserve extended discussion elsewhere.
Islamic private schools provide an ideal setting to provide forms of preventative, proresiliency activities, as well as to monitor and screen students for behavioural health issues. Many Islamic private schools have a formal student care and support system5 in place that integrates individual, family and community-based interventions to promote wellbeing. This model has been widely adopted in Thai secondary educational environments (Aiemcharoen & Dechhome, 2021; Chomkhunthod & Eungpuang, 2015). In Islamic educational settings, this approach includes the use of faith-based content and processes such as using halaqah – Islamic study circles in which teachers and students discuss Islamic principles and application to everyday life. Islamic private school staff also monitor student behaviour informally, communicate with parents regularly about behavioural concerns, and employ the Strengths and Difficulties Questionnaire (Goodman, 1997) – which has been translated and used in Thailand (Woerner et al., 2011) – to screen and potentially refer students with high needs for community-based services. However, studies indicate that there are deficiencies with training and resources in Islamic school environments, and opportunities for improvement (Buanak & Chookamnerd, 2020; Laeheem, 2018). Previous research has documented how Islamic school administrators and staff integrate discussion of student wellbeing into halaqah study circles (Laeheem, 2013, 2020). These interventions can be strengthened with continued funding from government authorities to train Islamic private school staff with behavioural health triage curriculums and evidence-based practices. Additionally, there are opportunities for more engagement and education with parents – particularly because stigma related to mental and behavioural health issues remains a significant barrier. It is critical that such efforts are directed and led by affected communities on the ground, using Islamic private school networks to engage communities.
It is worth noting that multiple intervention approaches may be successful. Dorsey and colleagues (2017) conducted a meta-review of studies testing psychosocial treatment to address trauma symptoms for children and youth. They concluded that the most efficacious interventions included some element of cognitive behavioural therapy (CBT). Their review included studies focusing on war/conflict trauma and excluded studies that did not screen or measure mental health symptoms. Some have argued that most CBT work focuses on treatment of single past traumas and not ongoing or systemic traumas (Kira et al., 2015). Conflict zone trauma that is ongoing challenges a sense of safety which is a prerequisite for most trauma therapies. One study tested modified group CBT (two hours on two days) with Thai youth following the 2004 Tsunami. Youth receiving the intervention had some immediate improved symptoms, but avoidance as a symptom lingered. Researchers noted this was likely due to the perceived ongoing threat of danger post-tsunami. The focus of this two-day approach was on psychoeducation and building coping skills (Pityaratstian et al., 2007). Similar interventions directed at reducing distress and decreasing the impact of trauma for children and youth can be delivered in school environments. For example, Layne and colleagues (2008) tested two tiers of intervention with Bosnian Muslim youth with high exposure to conflict. The intervention included classroom psychoeducation and skill building around coping with grief and trauma (tier 1), and school-based, manualised group therapy for youth screened for PTSD symptoms (tier 2). Their results were promising and led the Federal government to sponsor and expand the programme. Trauma interventions in zones with ongoing conflict may be most effective if they begin with skill building directed at increasing sense of safety and providing accessible coping skills to youth who will likely incur multiple trauma exposures.
Limitations
We must acknowledge study limitations. Our sample was not a generalisable representation of the community, but a convenience sample restricted to Muslim high school youth in Islamic private schools. Although our findings on PTSD symptomology generally align with previous cross-sectional studies conducted among different populations in the conflict-affected area, it is not a scientifically representative sample of students in Islamic private schools, or the general youth population in Southern Thailand. As we noted, the gender imbalance among our sample was also pronounced. We may expect gender differences in some traumas (such as sexual assault) but a gender imbalance may result in low statistical power to sufficiently identify significant differences. It is important to recognise that the nature of the sample may have also influenced our findings regarding the intrusion/indignity/identity-based traumas. We may expect that students in religious schools would have heightened sensitivity to traumas associated with religious identity. Religious populations could be more receptive to perceptions of unfairness towards them, a dynamic that may encourage shifts towards more religious fundamentalism or ideologically radical thinking (Muluk et al., 2013).
There are two methodological limitations to note. First, the Southern Thailand Trauma Checklist we produced should be retested in other environments. Our Cronbach alpha score showed a sufficiently high level. However, conflict and contextual dynamics change over time, as do the social memes in which they are experienced, expressed and documented. Secondly, we employed CFA to conceptualise and structure factors according to our own experiences and processing of information from students and peers. Although we were satisfied with the input and development process of our model and consistency of results with a suitably large number of respondents, the CFA approach in general and its inherent limitations should be noted.
Further Research
We recommend at least several directions for further research in this area. However, we must also acknowledge that current conditions in Southern Thailand are not supportive of research in general, a major challenge in all conflict settings. Targeted and random acts of violence and general fear and suspicion among community members are challenging barriers to obtaining data and implementing more robust research designs.
With these barriers in mind, we first recommend the use of a scientifically representative sample of Islamic private school students for future studies in this area. Islamic private schools serve a large segment of the conflict-affected area's population, with well over 150,000 youth enrolled in Islamic private schools (Office of the Private Education Commission, n.d.). A scientifically representative survey would produce generalisable findings about this population's experiences with trauma and conflict, and address the limitations of a convenience sample noted above. Secondly and related to the first recommendation, researchers should aspire to conduct more robust research designs among Southern Thailand's youth population to further investigate experiences of trauma, identify mediating or aggravating factors, and longer-term impacts. These investigations should be expanded to research designs with comparative populations, such as Muslim youth in government educational settings, or those not receiving education. This may better advance our understanding of the experience of trauma in general among different groups, and whether the outcomes we found – that identity-based traumas may be of particular salience – applies more generally to Muslim youth. Similarly, longitudinal research could assess long-term impacts of conflict-related traumatic experiences, identify protective factors within this environment that promote resiliency, and identify other environmental factors that may be relevant. This could be expanded to include broader enquiries relevant to trauma – such as the salience of family functioning, the existence of wider social support, and socioeconomic wellbeing.
A third recommendation for further research should be to examine other, more culturally relevant manifestations of trauma in Southern Thailand and investigate more applicable or detailed measures or variables of interest. We employed the PCL-6 as our primary scale for assessing PTSD symptomology, which aligns with the Diagnostic and Statistical Manual of Mental Disorders – the primary guidance used for diagnosis of mental conditions in the USA. Although the DSM is generally regarded as a credible and authoritative source, many researchers have discussed or investigated the applicability of using western constructs to measure trauma in nonwestern environments (Bracken et al., 1995; Jayawickreme et al., 2012; Kienzler, 2008). All trauma is experienced within unique cultural and social settings, and adequately measuring trauma symptomology in personally and culturally salient ways can optimally measure its impacts on wellbeing (Mollica et al., 1992; Shoeb et al., 2007). Caution should be used with applying a culturally disconnected research instrument or paradigm that may not reliably measure trauma symptomology as experienced by the community of interest (Pedersen, 2002). If a normative, policy goal of measuring PTSD symptomology among traumatised populations is to ultimately address the human impacts of conflict and violence, then the science of measuring trauma must incorporate important cultural factors and caution with applying inappropriate paradigms and researcher bias (Breslau, 2004). This is a broad issue deserving prolonged discussion elsewhere.
Fourth, further research must be carried out to measure the efficacy of interventions with this and other populations residing within the conflict zone. Such research must be performed with full consideration of the complex dynamics of the overall conflict and environment, and how it impacts wellbeing. The overall context will evolve as Thailand continues to experience political unrest on a national level, as well as major impacts from COVID-19. Successfully deploying community-wide interventions in this context will be challenging, and necessary to document and evaluate.
Finally and more broadly, our findings show that our respondents have experienced multiple exposures to many different types of trauma – accumulating over a period of many years. It is likely that individual manifestations of PTSD symptomology are a reaction to this aggregated and chronic experience of violence and conflict. But as recognised in studies from other conflicts, large questions remain about the extent to which direct exposures to conflict-related trauma impact PTSD and mental health status, versus other potentially related life stressors that typically occur in conflict-affected environments, such as financial hardships, household dynamics and violence, disruptions in social services, the ongoing coronavirus disease 2019 pandemic, and so on (Al-Krenawi et al., 2007; Farhood et al., 1993). Identifying the relevance of these items to wellbeing and PTSD in Southern Thailand will require much additional research.
Acknowledgements
The authors thank those students and teachers who supported or participated in this project. They also thank Anna Jaffe for her technical consulting and Ellie Woody for her copyediting. All of the opinions offered in this article belong to the authors alone and not their respective institutions.
Financial support and sponsorship
Data collection and research were supported by a grant from the Strengthening Human Rights and Peace Research and Education in ASEAN/Southeast Asia programme (SHAPE-SEA), funded by the Swedish International Development Agency (SIDA) and Norwegian Centre for Human Rights (NCHR).
Conflicts of interest
There are no conflicts of interest.
1Formal negotiations have started and stopped multiple times between Thai governments and actors representing a fragmented insurgency (see generally, Ganjanakhundee, 2020; Yeo Yaoren & Singh, 2020).
2It should be noted that ponok is the traditional Islamic boarding school institution. Starting in the 1960s, the Thai government moved to a policy in which ponok could voluntarily register with the government to receive public funding in exchange for incorporating required Thai national curriculum components. Upon registration, these ponok are technically called “Islamic Private Schools”. However, in general vernacular, ponok is still used to describe government registered Islamic private schools. For a discussion on the history of ponok schools' integration into the national educational structure, see Liow (2009, pp. 24–62).
3It is unknown how many Islamic schools exist in total within Thailand, as some are very small and do not receive government funding.
4427 completed surveys were collected from across the four schools (school 1, n = 99; school 2, n = 82, school 3, n = 147, school 4, n = 99).
5Known as “
” which translates to “support system”.
References
Comments (0)