Abstract
Using the Broaden-and-Build Theory of Positive Emotions, the relationships among stress, positivity, and depressive symptoms were examined in a sample of military spouses during deployment (N = 367). Over one-third of the spouses reported moderately severe levels of depressive symptoms. After controlling for demographic and deployment variables, stress had a positive association with depressive symptoms (β = .59, p < .001), while positivity had a negative association (β = −.39, p < .001). Positivity was also found to play a moderating role on the relationship between stress and depressive symptoms (β = −.29, p < .001). Spouses with lower positivity reported more depressive symptoms at both low and high levels of stress compared to those with higher positivity. The final model, including both direct and moderating variables, accounted for 69 percent of the total variance in depressive symptoms. Practical implications are discussed in terms of the importance of developing positivity in military spouses.
Recent developments in the field of positive psychology suggest that positive emotions have significant adaptive value, as higher levels of positive emotions have been shown to foster adaptation to adversity, promote health and well-being, and serve as a protective factor against stress. 1 Further, the benefits of positive emotions have been shown to enhance the psychological adjustment of military personnel to traumatic war-related stressors. 2 Consequently, as part of an initiative to enhance soldier readiness, the US Army has incorporated positive emotions into its Comprehensive Soldier Fitness program using the Broaden-and-Build Theory of Positive Emotions. 3
According to this theory, positive and negative emotions have distinct yet complementary adaptive functions. 4 The adaptive value of negative emotions (e.g., anger and fear) lies in their ability to narrow an individual’s thought–action repertoire in order to quickly and decisively react to an adverse situation in a particular manner (e.g., escape when afraid). 5 Negative emotions are viewed as having an evolutionary advantage, as they aid survival during immediate life-threatening situations. Conversely, positive emotions (e.g., contentment and joy) elicit a variety of enduring benefits, including the production of pleasant sensations and increases in cognition and social connectedness. These positive experiences broaden one’s thought–action repertoire, thereby expanding an individual’s cognitive processes, widening the potential responses that come to mind, and ultimately resulting in more thoughtful decision making and improved adaptation to adversity. In turn, this broadened cognition builds personal resilience resources, including social, intellectual, and physical resources, which enhance one’s adaptation to stressful conditions and overall well-being. 6 Further, positive emotions are also thought to have an undoing effect on negative emotions, whereby the broadened cognition that positive emotions elicit liberates individuals from the lingering narrowed mind set due to negative emotions. 7 Empirical evidence supports the adaptive function of positive emotions, as studies have found that anxious individuals recover more quickly following exposure to activities that evoke positive emotions. 8 In addition, positive emotions have been shown to increase one’s likelihood of bouncing back after a stressful experience, 9 accelerate the time needed to recover, 10 and serve as a resource for people coping with adversity. 11 Finally, positive emotions have displayed the ability to directly impact physical health, 12 psychological functioning, 13 and emotional well-being, 14 as well as moderate the effect of stress on coping 15 and depression. 16
Fredrickson and Losada 17 measured positivity as the ratio between one’s reported frequency of experienced positive emotions to the frequency of negative emotions, known in the Broaden-and-Build Theory of Positive Emotions as the positivity ratio. The implication is that positive emotions are necessary to overcome the toxicity of negative emotions, 18 and well-being is associated with high ratios of positive to negative emotions. 19 They determined that a positivity ratio at or above 2.9 is sufficient to promote human flourishing, a state in which individuals operate at an optimal level of functioning. 20 By contrast, a ratio below 2.9 is suggestive of languishing individuals who are “stuck in a rut” and “yearning for more.” 21 Further, a positivity ratio of less than 1.0 suggests a pathological level of functioning, often reported by individuals who are clinically depressed. 22 These associations are quite robust and are evident in a number of populations. For example, individuals, couples, and organizations that are flourishing report positivity levels above 2.9. Those with positivity levels below 2.9, while not diagnosable with a clinical disorder, tend to report only moderate mental health and experience as many illnesses and lost workdays as those who are depressed. 23 On the other hand, individuals being treated for clinical depression, couples with troubled marriages, and unprofitable businesses scored below 1.0. 24
Although improving soldiers’ positivity has been incorporated into the Comprehensive Soldier Fitness program, the Broaden-and-Build Theory of Positive Emotions has not been applied to promote wellness among spouses of those serving in the military; however, the well-being of those married to military personnel is vital, not only to maintain a high level of functioning within the individual family unit but also the military unit. Spouses who perceive military life as stressful display reduced psychological well-being, 25 and soldiers with dissatisfied spouses have higher attrition rates relative to those satisfied with military life. 26 Conversely, spouses who cope well with the demands of military life are more supportive of their soldiers’ career, 27 and as a result, these soldiers become more committed to the military. 28 Therefore, the spouses’ ability to adapt to the emotional stressors of military life plays a critical role in soldier readiness as well as the retention of an experienced military force.
The stressors that military spouses face are unmatched in the civilian world and include frequent relocations, lengthy deployments followed by abrupt reunions, and normative constraints instilled by the military. 29 Of these unique stressors, military spouses cite deployments as their major dissatisfaction with military life. 30 Current conflicts in the Middle East have resulted in American soldiers facing frequent and lengthy combat deployments. Although most soldiers appreciate the opportunity to deploy and use their training in meaningful, real-world combat operations, 31 spouses of deployed soldiers not only fear for their loved ones’ safety but also struggle with disruption of routine, becoming the sole decision maker, coping as a single parent, and loneliness and isolation. 32 Repeated deployments, in combination with the aforementioned stressors of being a military spouse and the conventional stressors of everyday life, place military spouses at serious risk of developing depression.
The inability to successfully manage the demands of military life may be a potential mechanism by which psychopathologies are developed, as the association between stressful life events and the onset of mental disorders is well documented. 33 Depression, among other psychopathologies, has been shown to disproportionately impact the spouses of those serving in the military, with military spouses reporting a prevalence rate of 19 percent, 34 which is over three times that of their civilian peers. 35 This discrepancy is even greater when soldiers are deployed, as wives are 1.24 times more likely to be depressed while their husbands are deployed relative to when their husbands are home; nearly 25 percent of wives are diagnosed with depressive disorder during the deployment of their husbands. 36 Similarly, the use of mental health services also increases during deployments. 37
While previous studies have examined the effects of deployment on a variety of stress-related psychosocial outcomes in military spouses, no studies were found that assessed the ability of positivity to serve as a protective factor against depression in this population. Therefore, the purpose of the present study was to explore the direct and interactive effects of perceived stress and positivity on depressive symptoms among wives of deployed active-duty Army personnel. Based on the Broaden-and-Build Theory of Positive Emotions, it was hypothesized that perceived stress would have a positive association with depressive symptoms, while positivity would have a negative association with depressive symptoms. Further, supporting the undoing hypothesis, it was expected that positivity would moderate the association between stress and depressive symptoms, such that high levels of positivity would lessen the influence of perceived stress on depressive symptoms (see Figure 1).
Hypothesized relationships among perceived stress, positivity, and depressive symptoms among military spouses during deployment.
Method
Participants and Procedures
Participants were obtained from two independent convenience samples of spouses whose husbands were stationed at Fort Hood, Texas and deployed in support of Operation Iraqi Freedom (Iraq; n = 77) or Operation Enduring Freedom (Afghanistan; n = 290). Participants whose husbands were deployed to Iraq with the 4th Infantry Division voluntarily completed a survey during Spouse Appreciation Day at Fort Hood. This day, hosted once per month by the Family Readiness Group (FRG), is designed to provide spouses with information and support. As the spouses gathered, they were given the opportunity to complete the survey. Further, they were assured that their responses were anonymous and their decision to participate or not would have no effect on their relationship with the Army or the university conducting the study.
Participants whose husbands were deployed to Afghanistan with the 1st Infantry Division were granted access to an online version of the survey via flyers posted on the FRG’s password-protected Web site. This Web site is visited frequently, as it is a primary source of information for spouses of deployed soldiers. The survey was posted on Survey Monkey, where participants could anonymously respond. To encourage participation, those who completed the survey could choose to enter their e-mail address into a random drawing to win a gift certificate to a local business (valued at $20). One prize was awarded for every ten surveys completed.
These two independent samples were collapsed into a single sample (N = 367) for the statistical analyses. Doing so not only increases statistical power but also provides an overall sample that is a more accurate representation of the Army population. A variable was included in all analyses to control for whether participants were married to soldiers deployed to Iraq or Afghanistan.
Participants ranged in age from nineteen to fifty-four, with a mean age of twenty-seven (± 5.75) years. The majority were Caucasian (66 percent), followed by Hispanic/Latino (18 percent), African American (11 percent), Asian/Pacific Islander (3 percent), Native American (1 percent), and other (1 percent). With regard to education, 4 percent had a post-graduate degree, 26 percent had a Bachelor’s degree, 7 percent had an Associate’s degree, 51 percent had some college, 11 percent had a high school diploma/general equivalency diploma, and 1 percent had some high school. Participants were married an average of five years, 83 percent had children, with approximately two children per household, and 22 percent maintained employment outside the home. The majority of participants were married to junior enlisted soldiers (53 percent), followed by noncommissioned officers (34 percent) and commissioned officers (13 percent). The participants’ husbands had been deployed for an average of eleven months in a location perceived as dangerous. Participants had experienced an average of two deployments (see Table 1).
Correlations, Means, Standard Deviations, and Range Values for All Variables
Note: Correlations between continuous variables are Pearson, while those including dichotomized variables are Point-biserial.
aLocation of deployment (0 = Iraq; 1 = Afghanistan), Spousal rank (0 = junior enlisted; 1 = noncommissioned and commissioned officers), Minority (0 = Caucasian; 1 = minority), Education (0 = no college degree; 1 = college degree obtained), Employed (0 = unemployed; 1 = employed), and Children (0 = no children; 1 = have children).
* p < .05.
** p < .01 (two-tailed).
Based on recent demographic data from Defense Manpower Data Center, 38 the current sample appears to be representative of the target population with respect to age and number of children. With regard to spousal rank, 46 percent of the active duty US Army is junior enlisted, while 37 percent are noncommissioned officers, and 17 percent are commissioned officers. Therefore, the current sample is comparable in rank to the overall Army. In terms of ethnicity, the active duty US Army is composed of 62 percent Caucasian, 20 percent African American, 11 percent Hispanic, 4 percent Asian, and 3 percent other; thus, our sample was similar to the total Army population with regard to proportion of Caucasian respondents, but slightly overrepresented Hispanics and underrepresented African Americans.
Measures
The survey instruments assessed participant demographic characteristics, deployment characteristics, perceived stress, positivity, and depressive symptoms. Each of these measures is detailed below, and a copy of the entire instrument may be obtained from the corresponding author.
Demographics
Participants were asked to report personal characteristics, including age, ethnicity, education level, employment status, length of marriage, and number of children.
Deployment characteristics
Participants were asked to report characteristics related to the present deployment, including husband’s rank, length of present deployment, perceived danger of husband’s present location, and total number of deployments ever experienced. Perceived danger of husband’s location was measured on a 10-point Likert-type scale from 1 (not dangerous) to 10 (very dangerous).
Perceived stress
Participants’ perceptions of stress were measured using the 10-item Perceived Stress scale (PSS). 39 Designed to measure how unpredictable, uncontrollable, and overloaded participants perceive their lives, the PSS assesses how often individuals have felt or thought a certain way during the past month. Sample items included “In the last month, how often have you felt that you were unable to control important things in your life?” and “In the last month, how often have you felt difficulties were piling up so high that you could not overcome them?” Responses were measured on a Likert-type scale from 0 (never) to 4 (very often) and were summed for a total score. Reliability of the PSS was strong (α = .84), which is similar to the reliabilities found during the development of the measure (α = .84 to .86). 40
Positivity
Positive and negative emotions were measured using the twenty-item Positive and Negative Affect Schedule (PANAS), 41 with positivity representing the ratio of positive emotions to negative emotions. 42 On a Likert-type scale from 1 (not at all) to 5 (very much so), participants indicated to what extent they generally experienced various positive emotions, such as “excited,” “enthusiastic,” “inspired,” and “interested,” and negative emotions, such as “irritable,” “upset,” “distressed,” and “afraid.” The number of positive emotions experienced at least “moderately” (≥3) and the number of negative emotions experienced at least “a little” (≥ 2) were tallied, with the different thresholds in place to account for negativity bias and positive offset. Negativity bias reflects the phenomenon that individuals give more weight to negative rather than positive emotions, 43 while positivity offset reflects the phenomenon that people tend to feel at least mild positive emotions most of the time. 44 A positivity score was then calculated by dividing the frequency of the positive emotion items by the frequency of the negative emotion items. Reliability was strong for both the positive emotion items (α = .91) and negative emotion items (α = .84) and corresponded well to those found in the initial psychometric work for both positive (α = .88) and negative (α = .87) emotion items. 45
Depressive symptoms
The 20-item Center for Epidemiological Studies Depression scale (CES-D) was used to measure depressive symptoms experienced during the past week, such as depressed mood, feelings of guilt, worthlessness, helplessness, and restless sleep. 46 Sample items included “I was bothered by things that usually don’t bother me” and “I felt everything I did was an effort.” Responses ranged from 0 (rarely or none of the time; less than one day) to 3 (most or all of the time; five to seven days), and were summed for a total score. A CES-D score of sixteen or greater is considered a moderately severe level of depressive symptoms. 47 Of the overall sample in the present study, 39 percent reported a score of sixteen or higher. Reliability of the CES-D was strong (α = .81) and comparable to the estimates of reliability in general populations used during the development of the measure (α = .85). 48
Statistical Analyses
Hierarchical linear regression analyses were conducted using the Statistical Package for the Social Sciences version 18. This statistical analysis allows for the examination of the direct effects of stress and positivity on depressive symptoms, as well as the interactive effect of stress and positivity, after controlling for the variance associated with deployment location (Iraq or Afghanistan), as well as other continuous and categorical demographic and deployment variables that were significantly correlated with depressive symptoms. 49 Predictor variables were added as three sequential models in order to determine the contributions of each set of variables based on our theoretical assumptions. Post hoc moderation analyses were also conducted according to the guidelines of Aiken and West. 50 Preliminary analyses indicated that the statistical assumptions of power, normality, linearity, multicollinearity, independence, and homoscedasticity were satisfied. All continuous predictors were mean-centered to minimize multicollinearity.
Length of marriage, number of deployments, length of deployment, and perceived level of danger were retained as continuous control variables. Due to the high correlation (r = .86) between age and length of marriage, age was excluded from analyses. Categorical demographic variables were dichotomized based on conceptual relevance: location of deployment (0 = Iraq; 1 = Afghanistan), spousal rank (0 = junior enlisted; 1 = noncommissioned and commissioned officers), minority (0 = Caucasian; 1 = minority), education (0 = no college degree; 1 = college degree obtained), employed (0 = unemployed; 1 = employed), and children (0 = no children; 1 = have children).
Results
Descriptive Statistics
Table 1 displays the correlations, means, standard deviations, and range values for all variables. Stress had a significant positive correlation with depressive symptoms and a significant negative correlation with positivity, while positivity had a significant negative correlation with depressive symptoms. With regard to the control variables, wives of commissioned officers and those who had experienced more deployments reported lower levels of stress and depressive symptoms and higher levels of positivity. Associations among the control variables indicated that spouses of soldiers deployed to Afghanistan were more likely to be married to enlisted soldiers, had been married for shorter periods of time, and their spouses had been deployed for longer durations in a location perceived to be more dangerous as compared to spouses with husbands in Iraq.
Regression
Deployment location and other demographic and deployment variables significantly correlated with depressive symptoms (length of marriage, number of deployments, spousal rank, and education) were entered into model 1, and explained 12 percent of the variance in depressive symptoms (see Table 2). Following the entry of perceived stress and positivity in model 2, the total variance explained by the model was 61 percent (F 7, 344 = 79.91, p < .001). The two focal predictors explained an additional 49 percent of the variance in depressive symptoms, after controlling for the demographic and deployment variables (F change2, 344 = 220.97, p < .001). In the final model, both perceived stress (β = .59, p < .001) and positivity (β = −.39, p < .001) were associated with depressive symptoms, while all demographic and deployment control variables became nonsignificant. To examine the ability of positivity to moderate the effect of stress on depressive symptoms, the interaction term (Perceived stress × Positivity) was tested in model 3. This moderation variable was significant (β = −.29, p < .001), resulting in the explanation of an additional 8 percent of the variance (F change1, 343 = 88.67, p < .001). The final model explained 69 percent of the variance in depressive symptoms.
Summary of Hierarchical Regression Analysis for Variables Predicting Depressive Symptoms (N = 355)
* p < 0.05.
** p < 0.01.
*** p < 0.001.
Post hoc probing of the significant interaction term was performed according to the guidelines of Aiken and West. 51 Plotting the interaction was done to express the regression of stress on depressive symptoms for various levels of positivity. Aiken and West 52 recommend the use of three values for the moderating variable (i.e., positivity) that represent the mean, 1 SD above the mean, and 1 SD below the mean; however, when theory and previous research suggest values of the moderating variable that may be of interest, those values may be used. Based on the work of Fredrickson and Losada, 53 three positivity cutoff scores were identified and used to plot the interaction: flourishing (positivity ratio ≥ 2.9), languishing (positivity ratio = 1.0 to < 2.9), and depressed (positivity ratio < 1.0). Low and high values of stress were calculated as 1 SD below and above the mean, respectively.
Figure 2 depicts the pattern of depressive symptoms for the three levels of positivity. The less positivity a spouse reported the stronger the relationship between stress and depressive symptoms. But regardless of level of positivity, spouses did not meet the criterion (≥16 CES-D score) for a moderately severe level of depressive symptoms at low stress levels. As stress levels rose, all wives reported an increase in depressive symptoms; however, those who were flourishing remained well below the aforementioned CES-D criterion even at high levels of stress. Conversely, both wives who reported languishing and depressed levels of positivity exceeded the criterion at high levels of stress; further, those with depressed levels of positivity reported nearly twice as many depressive symptoms as flourishing wives.
Positivity as a moderator of the relationship between perceived stress and depressive symptoms. Note: Depressed = positivity ratio < 1.0; Languishing = positivity ratio 1.0 to < 2.9; Flourishing = positivity ratio ≥ 2.9.
Discussion
This exploratory study used Fredrickson’s Broaden-and-Build Theory of Positive Emotions 54 as a guide to examine the relationships among stress, positivity, and depressive symptoms in military spouses during deployment after controlling for a number of demographic and deployment variables. As hypothesized, there was a positive association between stress and depressive symptoms and a negative association between positivity and depressive symptoms. Positivity moderated the relationship between stress and depressive symptoms, supporting Fredrickson’s undoing hypothesis that positive emotions downregulate the negative effects of stress. 55 This interaction between stress and positivity revealed that higher levels of positivity protect military spouses from developing depressive symptoms at both low and high levels of stress.
According to Fredrickson and Losada, 56 a positivity ratio of 2.9 and higher is indicative of individuals who are flourishing, while a ratio below 2.9 is reflective of languishing individuals; further, a positivity ratio of less than 1.0 may be a marker for clinical depression. 57 Of US adults, approximately 17 percent are characterized as flourishing, in that they report optimal social and psychological well-being 58 ;58 comparatively, 33 percent of participants in the present study reported positivity ratios of at least 2.9, signifying that a large portion of military spouses are functioning at a high level. However, 23 percent reported a ratio below 1.0, which suggests clinical depression, a percentage nearly four times that of the national sample (4.7 percent). 59 Further, 39 percent of the present sample reported moderately severe levels of depressive symptoms. Future research should examine whether the higher proportion of military spouses at both extremes—flourishing and depressed—is representative of the true military spouse population. In particular, the high proportion of military spouses at the depressed extreme is problematic, as positivity ratios less than 2.9 not only impact the mental health of the military spouse but also the health of the marriage, 60 threatening the spouses’ satisfaction with military life and increasing soldier attrition. 61
In addition to the current programs available to the military soldier, programs are also needed to increase the positivity of military spouses. Traditionally, military families were expected to adapt to the needs of the military; however, with the adoption of an all-volunteer force and an increase in the percentage of married military personnel, recognition of military family concerns has become a necessity. 62 Furthermore, the increased reliance on the military since September 11 has placed high demands on military families, and this demand is increasingly being met with intolerance and dissatisfaction. 63 In response, Drummet and colleagues 64 have suggested that the military adopt Family Life Educators (FLEs) to “strengthen and enrich individual and family well-being.” 65 Trained to focus on prevention instead of postcrisis management, FLEs emphasize education and growth rather than pathology. Through both individualized counseling and the development of informal support groups, 66 FLEs could provide military spouses with the tools to increase their positivity ratio, either by increasing positive and/or decreasing negative emotions. Using positive psychology to teach military spouses to dispute negative thinking, prevent rumination, avoid negative circumstances and people, find positive meaning, relish in good experiences, keep an open mind, and connect with others can arm them with the skills to combat languishing and depression.
The findings of the present study should be considered in light of several limitations. First, this study utilized a convenience sample. The spouses of those deployed to Iraq were not randomly sampled from the population of interest. Rather, they self-selected into the study through their attendance at the Spouse Appreciation Day. It may be that spouses not in attendance at this FRG-sponsored event are suffering from depressive symptoms to a lesser or greater extent; however, the sample of spouses whose soldiers were deployed to Afghanistan was comprised of spouses that both attended (31 percent) and did not attend (69 percent) FRG events. Therefore, the combination of these two samples aided in the alleviation of this limitation. Second, both samples were comprised of spouses of infantrymen stationed at Fort Hood, Texas, and these results may not generalize to spouses at other military installations or branches. Third, the use of self-report surveys has inherent limitations, such as the potential for inaccurate responses due to lack of self-awareness and common method variance. Although the anonymity of the surveys enhanced the strength of the study, future research should include objective measures of stress and verifiable measures of depression. Fourth, the response rates of these surveys were unavailable, as the number of spouses in attendance at the Spouse Appreciation Day and who are members of the FRG Web site were unknown. Because the study was conducted using cross-sectional data, causality cannot be determined. Furthermore, unmeasured variables, such as resilience, may account for some of the observed relationships. Future research should adopt a prospective design to further examine the nature of the predictive and interactive relationships among these constructs.
Despite these limitations, the implications of this study can be beneficial for the spouses of our military personnel. Military spouses are consistently exposed to numerous stressors that may contribute to the development of psychopathology; however, consistent with the Broaden-and-Build Theory of Positive Emotions, 67 positive emotions were found to undo or downregulate the negative effects of stress on depressive symptoms. Programs can be implemented that educate spouses with regard to effective ways to increase positivity in order to promote successful adaptation and achieve optimal functioning during times of stress, thereby enhancing not only the well-being of military families but also soldier readiness and the retention of an experienced military force.
Footnotes
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
The authors received no financial support for the research, authorship, and/or publication of this article.