Comparison of Social and Economic Stress in Military and Civilian Families: A Rapid Review of the Evidence ()
1. Introduction
Military families are embedded within a broader military context and culture which may influence many aspects of family life, including socioeconomic status (SES) and social participation (Blakely et al., 2012; Clever & Segal, 2013; Thandi et al., 2018). Research pertaining to military families’ SES and social participation has considerable policy and practice implications as they may impact retention. However, whilst military retention is related to multiple inter-related factors at various levels including community, organisation, family and individual (Hawkins et al., 2018), most studies have focused on the individual context, and less on the SES and/or social participation of military families (Blakely et al., 2012; Clever & Segal, 2013; Thandi et al., 2018).
Previous research categorises both current serving members and ex-serving members as a single combined group; limiting the generalisability of findings to current-serving members and families (Mailey et al., 2018; Wang & Pullman, 2019). In addition, ex-serving members may experience different circumstances and/or present with additional specific needs compared with current serving members (Stevelink et al., 2019; Maguire et al., 2022). For example, after transiting to civilian environments, ex-serving members experience multiple losses, including culture, healthcare, employment, housing, income, military identity and social networks (Sayer et al., 2011). Studies focusing on currently deployed or serving members are limited, and those with such a focus rarely use civilian families as a control group.
2. Literature Review
The Australian Defence Census 2019 Public Report indicate that amongst permenant Australian Defence Force (ADF) members, 60% had some form of post-school education compared to 40% with some form of secondary education (Roy, 2020). The Army had the higher proportion of members with completed Year 12 or less (48%), compared with the Navy (37%) and Air Force (25%). The Air Force had the highest percentage of members with a bachelor’s degree or higher (32%), compared to Navy (23%) and Army (20%). In terms of employment, around 21% of the ADF permanent members with a partner indicated that their partner was not employed and 25% of the members indicated that their partner was employed by the ADF. The members of ADF had strong participation in community activities, as indicated by 25% reporting engagement with sporting clubs or groups. Among the branches of the defence force, Air Force members had the higher (45%) involvement in community or social groups than members in the Navy (41%) or Army (36%).
Instability caused by frequent movement and absence of military members is likely to have an impact on the education and employment of military spouses. Several US based studies have found differences in employment, income, and education status between military and civilian spouses (Harrell et al., 2004; Hisnanick & Little, 2010; Hosek & Wadsworth, 2013; Meadows et al., 2015; Wang & Pullman, 2019). Military spouses appear to report lower rates of employment and lower weekly earning than their civilian spouses; these differences remain after adjusting for age, geography, education and number of children (Wang & Pullman, 2019). Studies suggest that constant anticipation of relocation may discourage military spouses from seeking employment or enrolling in formal education (Burke & Miller, 2016). Military spouses are also more likely to engage in careers which they are over-qualified for because they prioritise flexibility (Gribble et al., 2019).
In addition, the varied and unpredictable work schedule of military members may place more of the family/child caring responsibilities on the spouse, further limiting their options for employment and education (Burke & Miller, 2016). Other studies suggest that military spouses are more likely to be young, better educated and have dependent children compared to their civilian counterparts, affecting their likelihood of being employed (Lim et al., 2007). The studies above did not separate current and ex-serving servicing members within military study samples which may bias the results as the characteristics of ex-serving members are likely to be different to currently serving military members.
There is increasing attention on outcomes for military families’ dependants in addition to military spouses, as the mobile nature of the military service life may have adverse impacts on children’s education (Anderson, 1997). In the US, childcare and children education functions, parenting and family functions, finance and spouse employment, marital problems, military life experience and rank of military position level were found to be significantly associated with the retention of the military services (Hawkins et al., 2018). However, these findings were based on military samples with no comparison with civilian families.
Social economic status and social participation may then impact risk of adverse experiences. Several studies indicate that variables such as age, education and marital status are associated with intimate partner violence (IPV) within military families (Smith-Marek et al., 2016; Kwan et al., 2020). Previous systematic reviews (Smith-Marek et al., 2016; Kwan et al., 2020) on the risk factors of IPV in military families, highlighted that military lifestyle stresses including frequent relocation or family separation may impact relationship satisfaction and may lead to higher rates of IPV within the military family. Research indicates that protective factors such as social support and social participation may decrease the risk and/or impact of IPV among military families (Hawkins et al., 2018). Whilst social, economic, and military stressors are likely to contribute to adverse experiences, there is a need to further explore and compare the SES and social participation of current serving military sample with civilian populations.
Considering these gaps in the literature, this paper utilises a rapid review methodology to address the following two questions:
1) How do social and economic outcomes (social participation, education, legal/justice, housing, family income, spousal employment, and social participation) differ for current serving military families, relative to families in the general population?
2) What are key modifiable risk factors (prevention targets) to support military family social and economic wellbeing?
3. Methods
This review utilised the Cochrane Rapid Reviews guidelines (Garritty et al., 2021) with the following phases: 1) deciding on a research question and appropriate search terms; 2) utilising the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines (Moher et al., 2009) to systematically search relevant databases; 3) using inclusion and exclusion criteria to systematically screen retrieved studies; 4) undertaking data extraction; 5) assess the quality of the included studies; and 6) synthesising the findings using a narrative approach. The review protocol was registered and is available on PROSPERO (CRD42022323317).
Search strategy
As a rapid review, electronic databases of Medline, PsycINFO, CINAHL and ProQuest Central, were searched for published records from January 2000 to February 2022. Search term selection was guided by the research questions using the PICO strategy (da Costa Santos, de Mattos Pimenta, & Nobre, 2007).
P: Current serving members who have family/spousal partners of current serving members
I: Serving in a defence force
C: Never serving in a defence force
O: Social and economic outcomes/social participation
A comprehensive list of search terms was developed. Outcomes pertained to one or both members of the partnership and included: community participation (e.g., involvement in clubs, societies, or community activities; social support; civic engagement), education, family income, household income, partner income, partner employment, housing conditions, disadvantage or socioeconomic disadvantage, legal issues and violence.
Inclusion and exclusion criteria
Studies were published in English since the year 2000, inclusive. Study designs included any of experimental, RCT, systematic review, mixed methods, or qualitative studies (where a general population or never-served comparison group is used). Studies excluded were case studies, studies containing data collected post-service, even if it concerned the period of service retroactively, and military member(s) on mandatory service.
Screening and data extraction (selection and coding)
A pilot exercise was conducted using 30 - 50 abstracts for the screening team to calibrate and test the review form. Two reviewers independently screened 20% of abstracts, with conflict resolution through discussion. One reviewer screened the remaining abstracts and a second screened all excluded abstracts, with resolution of conflicts by a third screener. Using a standardized full-text form, a pilot exercise used the same 10 full-text articles for the entire screening team to calibrate and test the review form. A single reviewer extracted the data using a piloted data extraction form within the Covidence webapp. A second reviewer then verified that the data extraction has been conducted correctly.
For all included studies, we extracted information on the year of publication, name of the first author, country, study design, study population, research participants, age of participants, comparison or control group, attrition, point of service, indicator variables, outcome variables, outcome measure(s), covariates, relevant findings, and effect size.
A narrative approach was used to synthesise the findings. A meta-analytic approach was not appropriate, given the small pool of studies with inconsistent outcomes and various measurements. In addition, the JBI Critical Appraisal Checklist for Case Control Studies was used to assess the quality of included studies (Joanna Briggs Institute, 2016). A total score of 0 - 3 was considered as high risk of bias, 4 - 6 was considered as moderate, and 7 - 10 was assigned to low risk of bias.
4. Results
The search identified 3057 records. Following the removal of 901 duplicates, 2156 records were screened for title and abstract screening using systematic review software (Covidence systematic review software). Forty-five studies were chosen for full-text screening, from which five were included in the final review (Figure 1).
![]()
Figure 1. PRISMA chart of study identification.
Characteristics of included studies
Of the five included studies (Table 1), two were from Nigeria and three from the US. Four studies utilised a comparative cross-sectional research design, and one study used a cohort prospective design. The quality of the included papers ranged from low (3) to moderate (6) (please see Table A1. Quality assessment for the studies included in the review in Appendix).
Overall, there was considerable heterogeneity across the studies, including differences in terms of sample size, participants and comparison groups. One study focused on employment and income gaps between military wives and civilian
![]()
![]()
Table 1. Description and characteristics of included studies (N = 5).
Note: a = N of comparison group was reported in a range (from 220 to 128,990 in the match group), b = a neighbourhood in Abuja, Nigeria containing 16 streets, c = incomplete data at baseline and/or 2-year follow up.
wives (Meadows et al., 2015). Four studies focused on the relationship between socioeconomic indicators or participation and use of violence, assaults or intimate partner violence in military and civilian families (Merrill et al., 2005; Chimah et al., 2015; Patton et al., 2017; Adejimi et al., 2022). Only one study [e.g., Merrill et al. (2005)] had both male and female participants, others had study participants of a single gender only (either male or female). These domains are synthesised in the following sections.
Study findings
Table 2 presents a summary of key findings. The findings are synthesised within broad domains such as employment and income, socioeconomic predictors of violence in military families.
Employment and income
Using comparative cross-sectional survey analysis, Meadows et al. (2015) examined the gaps of employment and income between military spouses and matched civilian spouses with similar characteristics (e.g., age groups, presence of children under 6 years, highest education level achieved). The study found wives with military partners worked around 15 hours less per week than their civilian counterparts. Military wives also reported they earned approximately USD$17,000/year less than the matched civilians. The authors found that socioeconomic factors, such as age, education, minority status, and children under 6 years were not associated with the working hours and employment gaps between military spouses and their civilian counterparts. Instead, characteristics of the military service-person’s service, such as Navy vs Army, rank of level/grade and frequency or times of move may predict employment gap. In contrast, there was a significant negative association of age and education with the earning or
![]()
![]()
![]()
Table 2. Summary of outcomes, covariates and effect size of reviewed studies (N = 5).
income gap between military spouses and their civilian counterparts. Characteristics of the military service-person’s service, rank of level/grade of their partners in the military and frequency or times of move were also significantly associated with the military spouses’ earning gap.
Prevalence of violent behaviour
A higher prevalence of violence or assault perpetration within the family was found among male military members compared with male civilians in the US and Nigeria (Patton et al., 2017; Adejimi et al., 2022). Furthermore, Chimah et al. (2015) found that women with a male military partner reported significantly higher prevalence of experiencing controlling behaviour, physical abuse and emotional abuse from their partner compared with women who has male civilian partner in Nigeria.
Using a pre/post analysis method, Merrill et al. (2005) compared rates of severe IPV perpetration during the year before enlistment and the second year of service in a US Navy personnel sample. Merrill and colleagues found that 11% of respondents self-reported severe IPV within the past year (premilitary). The percentage of perpetrating IPV was higher among women than men [20% vs 4%, X2(1) = 43.99, p < 0.001]. In contrast, respondents reported a slightly higher rate of (14%) perpetrating severe IPV during their second year of service, but the difference with premilitary rate is not statistically significant. The prevlanece rates of selfreported severe IPV perpetration did not vary by gender during the second year in the military (12% vs 16%). Furthermore, the prevalence rates of severe IPV perpetration among men were significantly increased from 4% to 16% [X2(1) = 32.51, p < 0.001] after join the military. While the prevalence rates of severe IPV perpetration among women were significantly decreased from 20% to 12% [X2(1) = 4.92, p < 0.05].
Socio-economic factors associated with increased risk for violent behaviour
Two studies compared the social and economic differences of male violence or homicide perpetrators between military members and civilians (Patton et al., 2017; Adejimi et al., 2022). Using US military sample, Patton et al. (2017) found that age, levels of education, race, marital status, and primary motive were significantly associated with perpetrator of violence being military members compared with civilians. However, after controlling for various covariates, only age, decline in health status, and physical health problems predicted perpetration by male military members compared with the male civilian control group. Using a Nigerian sample, Adejimi et al. (2022) found that age, education, marital status and current use of alcohol significantly predicted male civilians who perpetrated IPV. But these predictors were not associated with perpetrating IPV by male military members. Adejimi et al. (2022) also found that male military members who were senior level officers were more likely to perpetrate IPV than junior level officers [OR: 2.561 (95% CI: 1.264 - 5.190), p < 0.01]. This result is opposite to the results of the civilian sample where senior level officers were less likely to perpetrate IPV than junior level officers [OR: 0.485 (95% CI: 0.318 - 0.739), p < 0.001].
In Nigeria, Chimah et al. (2015) found that a lower level of education and younger age were significantly correlated with increased risk for the experiencing controlling behaviour, physical abuse and emotional abuse from their partner. The authors concluded that the respondents who have male civilian partners were better educated than the respondents who had male military partners (proportions of tertiary education were 62.0% vs 23%, p < 0.001), and mean ages for male civilian partners were higher than male military partners (44.9 + 9.6 and 37.75 + 5.90).
Using a pre/post analysis method and the US sample, Merrill et al. (2005) found that younger age was a significant factor that predicted premilitary IPV for men (r = 0.12, n = 421, p < 0.05) but not for women (r = 0.03, n = 542, p > 0.05). They also found ethnicity was a significant factor that predicted IPV for both men (c2 (2) = 9.51, n = 359, p < 0.01) and women (c2 (2) = 20.01, n = 476, p < 0.001). In contrast, respondents’ family income, level of education, and marital status were not statistically associated with the experience of past-year IPV. At the second year of military service, the proportion perpetrating IPV increased among men, however there was no significant difference between IPV perpetration among men and women. Interestingly, for those in the second year of military service, age, ethnicity family income, education level, or marital status were found not associated with past year IPV.
5. Discussion
This rapid review synthesised the limited available studies (n = 5) pertaining to social and economic status between current military serving families and civilian families. Most studies focused on social and economic factors as predictors of IPV and/or assault in military and civilian families. The findings highlight a higher risk of experiencing IPV within military families compared to civilian families. Some socio-economic predictors of IPV or assaults (i.e., education level and marital status) were significant for civilian populations but not for current military families. Age, decline in health status, and physical health were significant predictors of use of domestic violence in current military families. Overall, this rapid review highlight considerable gaps in this underresearched area and recommend future research, in particular Australian-based research with military families.
Socioeconomic outcomes, deployment and relocation
This rapid review found compared with civilian spouses, maintaining and developing a career may be difficult for military spouses due to partners’ relocation or frequent moves (Meadows et al., 2015). Relocation, gender, education, and service member pay grade or level appear to impact on a military spouse’s ability to receive and maintain employment and higher income. This may lead to family related concerns or stress, which may affect the retention of the military members.
The findings also suggest that education is likely a significant predictor that affects the income gap between military and civilian spouses. Therefore, policies or programs aiming to improve educational opportunities, career choices, and job search for spouses may help to reduce employment and income gaps between military and civilian spouses (Hisnanick & Little, 2015). Lack of stability may be a barrier to using the opportunities available to the military spouses. Providing access to online education and supporting meaningful connection to social networks may help military spouses with completion of study.
Two of the reviewed studies stated that compared with civilian spouses, military spouses were more likely to live in metropolitan areas (Harrell et al., 2004; Wang & Pullman, 2019). Furthermore, military wives living in remote areas appear to earn less than their metropolitan counterparts (Harrell et al., 2004). With this, military spouses may reap the benefits of living among a larger population density with a high degree of economic and social integration. This may create more employment opportunities for military spouses and opportunities to earn relatively more compared with those who live in remote areas. This finding suggests that providing housing supports and locating military spouses in metropolitan areas may help to increase rate of employment and increase earnings, which may result in improved military retention.
Social support and participation in military families
This review found no studies published in the past 20 years comparing social participation between current serving military and civilian families, thus evidence in this area continues to rely on within-group designs. Previous studies of military samples found that social support or participation including formal and informal supports may improve mental and physical health and family relationship for both serving members and their spouses (Crouch et al., 2017). Social supports from parents, friends or communities were significantly associated with improved health and well-being of active serving members and their military spouse (Hawkins et al., 2018). A positive link between social support and mental health was found in Australia, where military partners who reported more social support from family or non-family were less likely to screen positive for post-traumatic stress disorder (Australian Centre for Posttraumatic Mental Health, 2011). These findings suggest that involving military family in social interactions with family, friends and communities may help to improve health and well-being of military families and improve the retention rate of current serving members (Rossiter & Ling, 2022).
Studies also found frequent relocation and deployment may influence social connection to the wider society and impact the social support of military families (Smith-Marek et al., 2016). Social connection barries in military life have been identified (Mailey et al., 2018), through the transient lifestyle of military families who often move every 2 to 3 years, leaving some military spouses to report feeling discouraged about getting to know others since they would likely have to start over again in the near future. A military community service program supporting military spouses or families to build networks and maintain social relationships across relocations may strengthen mental and physical health within military families. Alternatively, policy may consider the evidence for less transition for military personnel with dependent families, to enable time to build and maintain critical social networks.
Socioeconomic outcomes and IPV in military families
Risks for use of intimate partner violence (IPV) within military families may share some characteristics with IPV in civilians. Younger age and lower level of education were found to be significant risk factors for the use of IPV (Evans et al., 1980). Military families were generally younger compared to civilian families within the studies, contributing to higher variance rate of IPV in military families, compared to civilian families. Hoyt et al. (2014) used the matched sample in the USA to include the civilian group with same age, ethnicity, education, or marital status with military groups (worth noting that the proportions of currently serving members were not distinguished from veterans in the military group), and the authors still found that male military members reported higher prevalence of using IPV compared with their civilian counterparts.
Many military spouses are required to relocate with their partners and are unable to achieve career advancement with the majority working in part-time or low-wage roles. Over time, some are placed in a position of social isolation and economic dependence due to their partner’s service, and further experienced financial abuse (Adejimi, Sekoni, & Fawole, 2022). Additionally, family stressors arise with the challenges of redefining power relationships within the family after long periods of separation during deployments (Asbury & Martin, 2011; Pollard & Ferguson, 2020). Providing more social support and education to assist with military member-family reintegration following periods of separation may help to improve family relationship and stress, and reduce the IPV rate in military families.
Limitations
The findings of this rapid review are to be interpreted within the context of study limitations. A rapid review methodology represents an abbreviated version of systematic review methods to synthesise and generate research evidence in a short period of time; hence this search was confined to four databases, and precluded searching of grey and ancillary literature. Our inclusion criteria excluded research outputs such as doctoral theses or reports. The criteria for sample inclusion in this study were restricted for current-serving members and a general population comparison group and publication date of studies was limited to the past 20 years. This may have led to omission of some relevant research on socioeconomic stress for civilian and military groups.
6. Conclusion
This rapid review highlights key socio-economic stressors for military spouses, including considerably lower employment rates and weekly earnings compared with civilian spouses. Young age of the serving member and decline in health status were both associated with higher prevalence of intimate partner violence in military families compared with civilian families. These factors were strongly correlated with low retention of military members and compromised mental and health well-being in military families. Furthermore, future studies on the comparison of military and civilian’s social and family network quality are warranted.
Prevention strategies, including those targeting educational opportunities, career choices, and employment support for military spouses, social and housing support, and social networks may contribute to the health and wellbeing, as well as economic status of military families, which may consequently support retention of military members.
Funding Sources/Sponsors
This review is funded by the Australian Government Department of Defence (SON3352211). RD is currently supported by a Research Training Program Scholarship funded by the Australian Federal Government. HJ and RD’s research is funded by ARC Discovery Project Grant (DP200101781).
Data Availability Statement (DAS)
The authors confirm that the data supporting the findings of this study are available within the articles reviewed in this study and its supplementary materials.
Appendix
![]()
Table A1. Quality assessment for the studies included in the review
Please note: 1 = Yes; .05 = Unclear; 0 = No; N/A = Not applicable. JBI Critical Appraisal Checklist for Case Control Studies was used to assess the quality of the reviewed studies. Risk of bias are low (7 - 10), moderate (4 - 6) and high (0 - 3).
Detailed search results in four key databases
Medline
Search Strategy:
PsycINFO
Search Strategy:
CINAHL
Search Strategy:
ProQuest Central
Search Strategy:
((noft (milita*) OR noft (soldier*) OR noft (officer*) OR noft (infantry) OR noft (defen?e) OR noft (arm*) OR noft (navy) OR noft (air force*) OR noft (armed service*) OR noft (marine)) AND (noft (combat*) OR noft (armed force*) OR mainsubject (military personnel))) AND (noft (spouse) OR noft (husband) OR noft (wife) OR noft (de facto) OR noft (partner) OR noft (spousal partner) OR noft (accompained) OR mainsubject (Spouse)) AND (noft (serv*) OR noft (deploy*) OR noft (enlist*) OR noft (duty) OR noft (post*) OR noft (station*) OR noft (assign*) OR noft (combat) OR noft (armed service)) AND ((noft (education) OR noft (child protection) OR noft (family court) OR noft (crim*) OR noft (violence) OR noft (disadvantage*) OR noft (houshold income) OR noft (family income) OR noft (spous* employment) OR noft (occupation)) OR (noft (job) OR noft (community participat*) OR noft (civic involve*) OR noft (civic work) OR noft (social support) OR noft (social quality) OR noft (community support) OR noft (community engagement) OR noft (community club))); Limit 2000-01-01-2022-02-28; scholarly journals