CC BY
2
UDC 314; DOI 10.18551/rjoas.2023-03.03
VIOLENCE IN BUDUBURAM REFUGEE CAMP IN GHANA: FORMS, SOURCES, AND CONSEQUENCES
Acquah Joyce De-Graft*, Tuffour Kwadwo Addo
Department of Peace Studies, School for Development Studies, University of Cape Coast,
Cape Coast, Ghana *E-mail: joyce.acquah@ucc.edu.gh
ABSTRACT
The study applies descriptive statistics and logistic regression analysis to investigate the forms, sources, consequences, and socio-economic factors influencing violence in the Buduburam Refugee Camp in Ghana. The study was conducted in the Buduburam refugee camp in Ghana with a sample of 100 refugee households selected randomly. Findings showed that most refugees had experienced violence in the camp. Physical abuse of women and children by adult men, fear and intimidation from armed robbers and violent groups, and beating, kicking, slapping, and striking with an object were identified as the major forms of violence. The major source of violence was disagreement over camp leadership. The majority of the refugees suggested that violence resulted in police arrest. The study revealed that establishing more United National High Commissioner for Refugee branch offices around the camps, increasing police presence and patrol, training refugees in negotiation and conflict management, handling issues of sexual violence with more seriousness, and encouraging understanding, tolerance, and commitment to peaceful coexistence were proposed measures that could reduce violence in the camp. Results from the logistic regression estimation indicated that age, gender, household income, household size, household head, years of formal education, and occupation significantly predicted the chances of experiencing violence in the camp.
KEY WORDS
Violence, refugees, refugee camp, logistic regression, Buduburam, Ghana.
According to UNHCR (2022), there were 26.6 million refugees as of mid-2021. This represents a growth of almost 16% from 2020, when the United Nations High Commissioner for Refugees was protecting 20.7 million refugees globally (UNHCR complete ref). By 2020, the number of refugee camps had grown in tandem with the global refugee population. About 6.6 million individuals or around 22 percent of the world's refugees reside in camps. Of this number, 4.5 million live in planned and overseen camps, and over 2 million are housed in independently erected camps run by UNHCR. 85% of the world's refugees and Venezuelans who have been uprooted overseas are housed in developing nations. 27% of the total is given asylum by the least developed nations (UNHCR, 2022).
Refugee camps, according to the UNHCR, are temporary structures built to provide emergency security and assistance to people affected by violent conflict and persecution. They are not intended to provide a long-term solution to refugees' needs, but rather to provide a safe haven and basic necessities. Notwithstanding, there is an increase in the recording of violent activities in refugee camps, indicating that violence constitutes a particular challenge (UNHCR, 1999). Forms and sources of violence, especially sexual and gender-based violence, and other structural and cultural violence are known to constitute great challenges for refugees and internally displaced people (Crisp, 1999).
Research investigating the relationship between the arrival of refugees and conflict incidence in host countries mostly aggregates data at the national level (e.g., Salehyan and Gleditsch, 2006; Salehyan, 2008; Böhmelt et al., 2019; Rüegger, 2019). Furthermore, the literature on refugee arrival and conflict incidence in host countries has concentrated on the conflict between surrounding host communities and refugees (Anomat Ali and Ocha 2018, Kumssa, Jones, & Herbert, 2009). For instance Codjoe et al (2013) noted that violence
between Buduburam refugees and local people is a result of pressure on the use of facilities, increased cost of goods and services, increased social vices, and environmental degradation. These studies do not show the dynamics of conflict and violence in refugee camps. To fill this void, this research analyses the forms, sources, and consequences of violence in a refugee camp. The current research is meant to inform the academic discourse on our understanding of the insecurities refugees face in their own settlement, which will in turn inform the design and mechanisms for ensuring safety in camps.
The broad objective of the paper is to examine the forms and sources of violence and strategies aimed at reducing violence in the Buduburam refugee camp. Specifically, the study aims to: (1) describe the socio-economic characteristics of refugees; (2) examine the forms of violence; (3) identify the sources of violence; (4) analyse the consequences of violence; (5) investigate the measures to be adopted for reducing violence; and (6) determine the socio-economic factors that influence violence in the camp. The rest of the paper is structured as follows: review of related literature on the subject under investigation presentation of the methodology, discussion of the results, and conclusions and recommendations of the study.
International refugee flow, according to some academics, can lead to conflict and other security implications in both sending and receiving countries (Loescher and Milner, 2005; Rosenblum and Salehyan, 2004; Loescher, 1993). As a result, refugee camps can frequently serve as a haven for rebel groups, a base of operations, and a rich source of potential recruits (Salehyan, 2005; Zolberg, Suhrke, and Aguayo, 1989). Anomat Ali and Ocha (2018) found that tension and conflict between refugees and host communities in the Kakuma refugee camp in Kenya were a result of competition over resources such as wood and water. Similar research by Kirui and Mwaruvie (2012) on physical insecurity in Dadaab refugee camps in northeastern Kenya in the 1990s discovered that disputes over governance structures and other diverse interests frequently arose between refugees and the host community. They also found that host communities' and refugees' tensions and conflicts are largely caused by limited natural resources as well as unequal treatment of refugees and host communities by governments and host communities. Likewise, Fisk (2018) discovered that locations that hosted larger refugee populations were more likely to experience intentional attacks on civilians. Salehyan and Gleditsch (2006) found that the presence of refugees increased the likelihood of conflict in host nations, though the vast majority of refugees were never directly involved in conflict events. Using the UNHCR Camp Mapping Database of 36 African countries, Coniglio, Peragine, and Vurchio (2022) found that the location of refugee camps increased the probability of protest only in the first two years after the arrival of refugees, while there was no such effect in the succeeding years. As a result, there was no evidence that camps had a significant impact on subsequent social unrest or organised violence events.
Johnson (2011) presented a quantitative model on the African continent that investigated refugee camp attacks by armed groups and discovered that the demographic make-up of camps affected the likelihood of an attack. As the percentage of males within the camps increased, the likelihood of an attack decreased. Also, the likelihood of an attack decreased when the age distribution of refugees within the camp changed to a lower level, and lastly, as the likelihood of an attack increased, the size of camps increased.
Salehyan (2008) studied whether refugee flow between states increases the probability of military interstate disputes (MIDs). A quantitative analysis was used from 1955 to 2000 to test the preposition of the relationship between MID initiation and refugee. It was found that refugees significantly raised the likelihood of international conflict. Corroborating academic literature suggests that the refugees' movement is sometimes linked to factors that contribute to the regional clustering of war. Similarly, Salehyan and Gleditsch (2006) conducted an empirical analysis of the link between refugees and civil conflict since the mid-twentieth century and found that the presence of refugees from neighbouring countries led to an increased probability of violence in host countries.
Contrary to the above studies, Bohmelt et al (2019) constructed a time series cross-section dataset containing all countries between 1989 and 2015 to establish the link between
refugees and non-state actor violence. They discovered that there is no link between the arrival of refugees and non-state violence in a country. In instances where refugees can be linked with non-state forms of violence, this is mainly because state institutions are weak. There was no clear effect of refugee populations on civil war, suggesting that the link only depends on existing conflict and weak state institutions. Savun and Gineste (2019) investigated violence against refugees, and they found that refugees are more likely to be exposed to violence by the agents of the host country in the wake of security threats. They argue that in the aftermath of terrorist attacks in host countries, refugees are especially vulnerable to violations of their physical safety (see also Loescher and Milner, 2005). Comparably, Ruegger, (2019) explored the linkage between forced migration, transnational connections, and ethnic civil conflict in the country of asylum. He contended that ethnic power politics were a determinant of the onset of intrastate conflict in an asylum state following a refugee influx. He concluded that the presence of refugees alone is not a predictor of violence but only in combination with the political tension in the receiving country. Using global data from 2,536 provinces from 1990 to 2018 on locations of refugee communities and civil conflict at the subnational level, Zhou and Shaver (2021) found no proof that the presence of refugees in host countries increased the probability of new conflict or sustained existing conflict.
The studies cited above have shown that early research discovered that refugee populations exacerbated host countries' internal and external vulnerabilities, leading to violent conflict. However, other findings suggest that, while the presence of refugees is likely to increase the likelihood of conflict, this is not always the case. To ensure that refugees enjoy stability in their new homes, studies on the sources of violence and mechanisms of violence prevention cannot be overemphasized.
The sample for the study comprises 100 refugee households. The population for the study was all refugees in need of indigenous microcredit support services in Buduburam. A simple random sampling technique was used to select the respondents.
An interview schedule was the main tool for data collection, while descriptive statistics and logistic regression analysis were the main analytical techniques. The data were analysed using STATA version 13.0. Results of the study were presented in tables and bar charts, as well as as frequencies and percentages.
Logistic regression is used to model a response variable that has a binomial distribution. With the logistic regression, we specify a model that describes the relationship between predictors and probability. Thus, we define a link function that links probabilities (p) to linear predictors of the model. The following is the link function:
METHODS OF RESEARCH
logito?) = log, =4 + .+ +
(1)
Where:
P is the probability that an event occurs (success);
? is the probability that an event does not occur (failure);
P
^ ? is the odds of success. The estimated probability is:
P =
In specifying the model for the study, the dependent variable is modelled as violence measured as 1 = if respondents experienced violence and 0 = otherwise. Therefore, the model for this data is specified as:
Where:
Age of respondent measured in years; Sex: 1 = male, 0 = female; Household head: 1= female, 0= male; Years of formal education of respondents; Number of people in respondents' household; Occupation: 1= unemployed, 0 = employed; Household income.
RESULTS AND DISCUSSION
In this section, we report our findings about the socioeconomic characteristics of respondents. The data shows that 30 percent were between the ages of 18 and 25, 30 percent were between the ages of 26 and 33, 19 percent were between the ages of 34 and 41, 10 percent were between the ages of 42 and 49, and 6 percent were between the ages of 50 and 58, with 2 percent being age 59 or older. Also, with the gender distribution of the respondents interviewed, 57 percent were females while 43 percent were males. The marital status of the respondents shows that 54 percent of the respondents were single, 40 percent were married, 3 percent were divorced, and 3 percent were widowed. Respondents' level of education shows that 42 percent have had secondary education, 20 percent have had primary education, 20 percent have had no formal education, 10 percent have had tertiary education, and 8 percent have had vocational training. Most of the refugees (91 percent) were from Liberia, 6 percent from Cote d'lvoire, 2 percent from Mali, and 1 percent from Togo.
With regards to the household size of respondents, 47 percent had a household size between 4-6 persons, 36 percent had 1-3 persons, 12 percent had 7-9 persons, and 2 percent had 10 persons or above. In addition, 64 percent had dependents aged 0 to 2, 29 percent had dependents aged 3 to 5, and 7 percent had dependents aged 6 to 8.Respondents interviewed indicated that 23 percent had lived at the camp for 10-12 years, 16 percent had lived at the camp for 13-15 years, 13 percent have spent between 4-6 years at the camp, 12 percent have spent between 16-18 years at the camp, 11 percent have spent 22 years or more at the camp, 9 percent have spent between 1-3 and 7-9 years at the camp, respectively, while only 7 percent have stayed at the camp for between 19-21 years.
Table 1 - Monthly income of respondents (in Ghana cedis)
Monthly Income_
0-99 100-199 200-299 300-399
400 and above_
Source: Survey Data, 2014.
Table 1 presents the monthly income of the respondents interviewed in the survey. From Table 1, the results indicated that 50 percent receive a monthly income of up to GH
50 26 12 8 4
50.0 26.0 12.0 8.0 4.0
99.00, 26 percent receive between the amounts of GH 100 and 199 monthly, 12 percent receive between the amounts of GH 200 and 299, and 8 percent receive between 300 and 399, while 4 percent receive a monthly income of GH 400 and above. Table 1 presents the monthly income of the respondents interviewed in the survey.
Forms of Violence in the Buduburam Refugee Camp. In order to investigate the forms of violence experienced by the refugees at the camp, the respondents were asked if they had ever experienced violence at the camp and in what form the violence was. From Figure 1, 86 percent said physical abuse of women and children by adult men is one of the major forms of violence in the refugee camp, and 37 percent indicated that violence in the camp takes the form of groups in the camp that are forced to fight on behalf of a particular group or government within their country of origin. Also, 76 percent identified rape of women and girls as a form of violence in the camp, and 4 percent of the respondents said female genital mutilation is one of the forms of violence in the camp. 27 percent identified women and girls forced into marriage as one of the forms of violence. Gang rape was noticed by 47 percent of the respondents as a form of violence experienced by refugees in the camp. Fear and intimidation from armed robbers were also identified by 93 percent of the respondents as forms of violence, and 86 percent of the respondents agreed that fear and intimidation by violent groups was a form of violence in the camp. 91 percent of the respondents agreed that beating was also a form of violence in the camp, and 78 percent, 87 percent, and 85 percent say violence in the camp comes in the form of kicking, slapping, and striking with an object, respectively. Head butting, electric shock, biting, and eye-poking were also identified as forms of violence in the camp by 57 percent, 84 percent, 54 percent, and 33 percent of respondents, respectively. Figure 1 shows the forms of violence in refugee camps.
Note: PAWC = Physical abuse of women and children by adult men, GFGCO = Groups in the camp forced to fight on behalf of a particular group or government within their country of origin, RWG= Rape of women and or girls, FGM=Female genital mutilation, WGFM= Women and girls forced into marriage, GR=Gang rape, FIA=Fear and intimidation from armed robbers, FIV= Fear and intimidation by violent groups, BE= Beating, K= Kicking, S= Slapping, SO= Striking with an object, HB= Head butting, EES= Exposure to electric shock, BI= Biting, EP= Eye poking.
Figure 1 - Forms of Violence in Refugee Camps (Source: Survey Data)
Sources of Violence in Refugee Camp. To understand the sources of clashes or violence experienced by refugees at the refugee camp, respondents were asked to identify some major sources of clashes at the camp. The study found that 23 percent of the respondents agreed that factions within the camp originating from home countries were a source of the violence witnessed at the camp, while 77 percent of the respondents disagreed. Also, 23 percent agree that factions within the camp originating from the camp in question are the source of violence at the camp, while 77 percent of the respondents do not agree. Of the 100 respondents, 13 percent say unresolved conflicts from home countries are a source of violence at the camp, and 87 percent disagree. Only 7 percent of respondents
said political affiliation in their home countries was a source of violence, while 93% said it wasn't. 66 percent identified disagreement over camp leadership as a source of violence at the camp, while 34% disagreed.
Note: FCOHC = Factions within the camp originating from home countries, FC = Factions within the camp originating from camp, URCHC = Unresolved conflicts from home countries, PAOHC = Political affiliation originating from home countries, DCL = Disagreement over camp leadership.
Figure 2 - Sources of Violence in Refugee camps Source: Survey Data
Consequences of Violence in Refugee Camp. In order to ascertain the consequences of violence in refugee camps, respondents were asked to identify some of the consequences of violence or clashes at the camp. Figure 3 shows the consequences of violence in refugee camps. It is clear that the most common result or consequence of violence is police arrest. Most (89 percent) of the respondents agreed that violence results in police arrest. 31 percent of the respondents consent that a change in camp leadership results in violence, while 69 percent of the respondents disagree. Also, 49 percent agree that police brutality results in violence, while 51 percent of the respondents do not agree. Of the respondents interviewed, 57 percent agreed that violence does not lead to the establishment of by-laws to prevent reoccurrence, while 43 percent disagreed.
Note: CCL = Change in camp leadership, PB = Police brutality, EBL = Establishment of by-laws to prevent reoccurrence, PA = Police arrest.
Figure 3 - Consequences of Violence in Refugee Camp Source: Survey Data
Measures aimed at Reducing insecurity in Refugee Camp. To find out the measures to be adopted for reducing insecurity in the camps, respondents were asked to list measures they think are likely to reduce insecurity in the camps. Figure 4 shows measures for reducing
insecurity in the refugee camps. It was known that 77 percent of the respondents agreed that more UNHCR branch offices around the camp should be established, while 23 percent did not. 99 percent of the respondents agree that police presence must be increased. Also, 99 percent agreed that police should regularly embark on night-time patrols. All the respondents (100 percent) think that refugees must be trained in negotiation and conflict management. Of the respondents interviewed, 98 percent agreed that watchdog committees should be encouraged among refugees, while 43 percent disagreed. Also, 81 percent believe that camp fencing can help reduce insecurity, while 19 percent do not agree. All respondents (100 percent) interviewed think that issues of sexual violence must be handled with more seriousness and that refugees should encourage understanding, tolerance, and commitment to peaceful coexistence among themselves.
120
Note: UNHCR= Establishing more UNHCR branch offices around the camp, IPP = Increasing the presence of the police, NTP = Night time patrol by the police, TRNCM = Training refugees in negotiation and conflict management, WDC = Watch-dog committees must be encouraged among refugees, CF = Fencing of camp, ISVHS = Issues of sexual violence must be handled with more seriousness, EUTCPF = Encourage understanding, tolerance and commitment to peaceful-coexistence among refugees.
Figure 4 - Measures aimed at reducing insecurity in refugee camp Source: Survey Data
Model Estimation Results of the Logistic Regression Analysis. This model was employed to analyse the socio-economic factors that influence violence at the refugee camp. The normal regression models that can be used in the case of the analysis of variables such as production or consumption, which is an interval variable in which each unit of measurement carries equal weight, are inappropriate when modelling involves the use of a qualitative response for the dependent variable. When there is a single decision, there are only two possible outcomes; the logistic model is used. The logistic regression model is appropriate in this analysis because of the binary qualitative responses of the dependent variable. The dependent variable is whether a refugee has ever experienced violence at the camp or not at the time of this study.
The Maximum Likelihood Estimation (MLE) technique was used to estimate the logit model. From the results, the Wald chi-square value of 22.46 with 7 degrees of freedom and a p-value of 0.002 less than 0.05 shows that the model has a good fit for the data. Besides, the Hosmer-Lemershaw model fitness test shows that we cannot reject our model, which also means that our model fits reasonably well. The link test also reveals no problems with the specification of our model. The results in Table 2 show the estimated coefficients of the variables for whether respondents experienced violence or not.
Six explanatory variables were found to be significant. These variables are therefore interpreted and explained as indicated in Table 2. As refugees grow older, the probability that they will experience violence at the camp reduces by 0.7 percent. The probability of men experiencing violence decreases by 15.5% compared to females. When compared to a male-
headed household, a female-headed household is 12.8% more likely to experience violence.An increase in the household size of respondents by one person increases the probability of the respondents experiencing violence at the camp by 4.9% at a 1 percent significant level, holding other variables constant. This finding is consistent with Yamano and Deininger (2005). With respect to the occupation of refugees, the probability of the unemployed experiencing violence increases by 24.9% compared to the employed. An increase in household income of $1.00 decreases the probability of the household experiencing violence by 8.6% at a 5 percent significant level, holding other variables constant.
Table 2 - Parameter estimates of the Logit Model
Dependent Variable: Violence
Explanatory Variables Coefficients Marginal Effects
Age -0.063 -0.007
(1.78)*
Gender(Female=0) -1.385 -.155
(1.87)*
Household head (Male=0) 1.209 0.128
(1.73)*
Years of formal education -0.153 -0.016
(-1.28)
Household size 0.462 0.049
(2.65)***
Occupation (Employed=0) 2.374 0.249
(2.16)**
Household income -0.822 -.086
(2.02)**
Constant 2.464
(2.13)**
P-value of Link test ( hat sq.) N Wald chi2(7) (P-value) Hosmer-Lemershaw test for goodness-of-fit (P-value) 0.551 100 22.46(0.002) 82.42(0.753)
t statistics in parentheses; * p<0.1; ** p<0.05; *** p<0.01. Source: Survey data, 2014.
CONCLUSION
According to studies, refugees can have a negative impact on national and international security. However, the presence of refugees does not always present a security threat. When refugees arrive in host countries, UNHCR and other aid agencies rely on camps as a temporary solution to help them settle into their new homes present a security threat. When refugees arrive in host countries, UNHCR and other aid agencies rely on camps as a temporary solution to help them settle into their new homes. It is therefore imperative that violence in camps be investigated to ensure safety for both refugees and receiving countries.
A survey of 100 randomly sampled refugee households was conducted using a standard questionnaire. Data collection was done using an interview schedule. Descriptive statistics and logistic regression analysis were the main analytical techniques. The objectives of the study were (1) to describe the socio-economic characteristics of refugees; (2) to examine the forms of violence; (3) to identify the sources of violence; (4) to analyse the consequences of violence; (5) to investigate the measures to be adopted in order to reduce violence; (6) to determine the socio-economic factors that influence violence in the camp. Findings show that most refugees have experienced violence in the camp. Physical abuse of women and children by adult men, fear and intimidation from armed robbers and violent groups, beating, kicking, slapping, and striking with an object were identified as the major forms of violence. Respondents indicated that the major source of violence was disagreement over camp leadership. Regarding the consequences, the majority of the
refugees suggested that violence results in police arrest. The study revealed that establishing more UNHCR branch offices around the camp, increasing police presence, regularly embarking on night-time patrols, training refugees in negotiation and conflict management, encouraging watchdog committees, fencing the camp, handling issues of sexual violence with more seriousness, and encouraging understanding, tolerance, and commitment to peaceful coexistence would reduce violence in the camp. Results from the logistic regression estimation imply age, gender, years of formal education, and household income as negative predictors of the probability of experiencing violence. Household head, household size, and occupation are positive predictors of the probability of experiencing violence. The implications for policy are that authorities should take the socioeconomic characteristics of refugees into account when dealing with violent clashes in the camp.
REFERENCES
1. Alexander M (1999) Refugee status determination conducted by UNHCR. International Journal of Refugee Law, 11(2), 251-289.
2. Ali JA and Ocha W (2018) East African refugee crisis: Causes of tensions and conflicts between the local community and refugees in Kakuma refugee camp, Kenya. Journal of Social Science Studies, 5(1), 298.
3. Bohmelt T, Bove V and Gleditsch KS (2019) Blame the victims? Refugees, state capacity, and non-state actor violence. Journal of Peace Research, 56(1), 73-87.
4. Codjoe SNA, Quartey P, Tagoe CA and Reed HE (2013) Perceptions of the impact of refugees on host communities: the case of Liberian refugees in Ghana. Journal of International Migration and Integration, 14(3), 439-456.
5. Crisp J (1999) A state of insecurity: the political economy of violence in refugee populated areas of Kenya. In: New Issues in Refugee Research, No. 16. http://www.refworld.org/docid/4ff58a7012.html
6. Coniglio N D, Peragine V and Vurchio D (2022) The geography of displacement, refugees' camps and social conflicts.
7. Fisk K (2018) One-sided violence in refugee-hosting areas. Journal of Conflict Resolution, 62(3), 529-556.
8. Johnson R A (2011) Refugee camp security: decreasing vulnerability through demographic controls. Journal of Refugee Studies, 24(1), 23-46.
9. Kirui P and Mwaruvie J (2012) The dilemma of hosting refugees: a focus on the insecurity in north-eastern Kenya. International Journal of Business and Social Science, 3(8), 161171.
10. Kumssa A, Jones J F and Herbert Williams J (2009) Conflict and human security in the North Rift and North Eastern Kenya. International Journal of social economics, 36(10), 1008-1020.
11. Loescher G and Milner J (2005) Case studies: contemporary protracted refugee populations in Africa and Asia. The Adelphi Papers, 45(375), 35-65.
12. Rosenblum MR and Salehyan I (2004) Norms and interests in US asylum enforcement. Journal of Peace Research, 41(6), 677-697.
13. Ruegger S (2019) Refugees, ethnic power relations, and civil conflict in the country of asylum. Journal of Peace Research, 56(1), 42-57.
14. Salehyan I (2008) The externalities of civil strife: Refugees as a source of international conflict. American Journal of Political Science, 52(4), 787-801.
15. Salehyan I and Gleditsch KS (2006) Refugees and the spread of civil war. International organization, 60(2), 335-366.
16. Savun B and Gineste C (2019) From protection to persecution: Threat environment and refugee scapegoating. Journal of Peace Research, 56(1), 88-102.
17. Savun B and Gineste C (2019) From protection to persecution: Threat environment and refugee scapegoating. Journal of Peace Research, 56(1), 88-102.
18. Zhou YY and Shaver A (2021) Reexamining the effect of refugees on civil conflict: a global subnational analysis. American Political Science Review, 115(4), 1175-1196.
