Digging Deeper: Gender-Related Stressors and Coping Strategies Among Farmers (N=1,394)

doi:10.21203/rs.3.rs-4027396/v1

Background

Farmer mental health - including stress levels, stressors, and coping mechanisms - has rarely been investigated in quantitative work. In addition, gender differences among farmers are seldom studied.

Aims:

The aim is to uncover gender-based disparities in stress experiences and coping strategies among agricultural workers.

Methods:

The study included 1,394 farmers (67.7% male, 32.3% female) from Georgia, USA, who completed an online questionnaire on various aspects of mental well-being.

Results

The findings highlight distinct differences in characteristics between male and female farmers, with a notable proportion of female farmers being first-generation, suggesting unique challenges for this subgroup. While Perceived Stress Scores were similar between genders, male farmers reported more unhappiness in their farm roles, while a significantly higher proportion of female farmers experienced depression 5-12 times per year. Additionally, females exhibited a slightly higher frequency of suicidal ideation.

Stressors were generally consistent among farmers, yet females expressed greater concern about work-life balance and financial issues, while males were more worried about sexual issues and alcohol use. Coping mechanisms varied, with male farmers employing a wider array of strategies, including alcohol consumption as a prominent coping method.

A stratified propensity score matching analysis was conducted to explore the gender effect on suicidal ideation. After controlling for relevant covariates, the gender difference in suicidal ideation disappeared. This suggests that gender-based stress responses may be influenced more by external stressors and environmental factors, potentially mediated by gender roles, rather than inherent sex differences.

Conclusion:

These findings underscore the significance of considering job-related variables when studying mental health outcomes in farmers, as these factors significantly influence suicidal ideation patterns. Identifying and addressing gender-specific challenges can inform targeted interventions to support the mental health and well-being of agricultural workers.

Purpose

Literature on the mental health of farmers has been scarce. The most common studies focus on the potential outcome of poor mental health, such as depression or suicide. Although some work is emerging on farmer’s stressors, there has not been a cross-sectional survey that compares male and female farmers’ stressors and coping mechanisms. Throughout this study, the term "farmers" will encompass agricultural workers of various roles, including farm workers, managers, and owners.

Georgia (GA) is one of the nation’s leaders in agriculture, boasting a robust industry comprising over 40,000 farms with approximately 68,000 producers and 49,000 hired farm workers (USDA National Agricultural Statistics Service, 2017). Agricultural commodities play a significant role in the state's economy, contributing a substantial $12.2 billion out of the $1.1 trillion total (USDA National Agricultural Statistics Service, 2017). Consequently, prioritizing the well-being of agricultural workers, and addressing their physical and mental health, is of utmost importance.

While research attention on Georgia’s farming population is a relatively recent phenomenon, the existing literature predominantly focuses on suicide and comprises qualitative studies and secondary data analysis. Consistent with a broader national trend, it has been observed that farmers in Georgia face higher suicide rates than their non-farming counterparts (Kennedy, Cerel, Kheibari, Leske, & Watts, 2021). A myriad of stressors, such as weather-related uncertainties, fluctuating commodity prices, and rising fuel costs, can weigh heavily on agricultural workers. These stressors have the potential to elevate levels of anxiety and depression (Brown & Harris, 1978), which could lead to a generally negative outlook on life– especially when persistent over time, coupled with limited coping mechanisms, coping mechanisms are limited (Bjornestad, Cuthbertson, & Hendricks, 2021), and exacerbated by challenges accessing healthcare (Reilly, 2021).

Gender Roles and Societal Expectations in Farming

Biological sex refers to the physical and physiological characteristics that distinguish males and females, typically categorized as male or female based on reproductive anatomy and functions. In contrast, gender roles are societal expectations, behaviors, and norms associated with being male or female (Bem, 1981). Sandra Bem pioneered the Gender Schema Theory, which suggests that individuals develop gender roles through the internalization of societal norms and expectations (Bem, 1981), a theory still heavily applied to this day in various academic domains (Starr & Zurbriggen, 2017). This paper focuses on the socially constructed nature of gender, emphasizing that it explores gender roles shaped by cultural and social influences rather than biological factors.

The separation of responsibilities by gender, commonly referred to as “gender roles,” has origins in the Neolithic Revolution. Initially designed to ensure safety and household surplus amid challenges like food scarcity, maternal mortality, and physical conflicts (Hansen, Jensen, & Skovsgaard, 2015), these roles have persisted through modern society despite industrialization eliminating many historical adversarial forces. They remain a significant social challenge at home and in the workplace.

This challenge is frequently document in current research which highlights fairness issues, such as financial and time contributions as a major source of controversy within married couples’ households (Peetz, Meloff, & Royle, 2023). The United States Department of Labor reports a substantial increase in women’s presence in the workforce, rising from approximately 28% in 1950 to 47% in 2020 (US Department of Labor, 2022). However, this rise in the workspace has not been met by a reduction of domestic responsibilities. In 2020, women were estimated to spend nearly triple the amount of time on domestic duties as their male counterparts worldwide, averaging 4.2 hours daily versus 1.7, respectively (United Nations, 2020). The gap is even wider in less developed regions.

Gender roles are believed to be more pronounced in societies with long agricultural histories (Hansen, Jensen, & Skovsgaard, 2015). Previous studies on this topic primarily focus on sustenance farmers in developing nations, such as rural regions of India and African countries. A 2016 study found that today 30% of US farmers are female, a significant increase from the 11% observed in 2005 (Sachs, Barbercheck, Brasier, Kiernan, & Terman, 2016). However, the expectations for a woman’s role within the home have not shifted with their increased participation in the workforce (Cooke & Rousseau, 1984). Previous work has found that the burden of work affects a female’s quality of family life more than males. This dilemma can quickly lead to role conflict for female farmers, particularly in agricultural-based communities with traditional gender role expectations. (Duxbury & Higgins, 1991).

Examining stress response and coping mechanisms reveals differences between men and women related to role conflict. While stress response typically varies by culture, it is widely believed that men use more active coping behaviors and emotional inhibition, whereas women tend to use more passive and emotion-focused behaviors (Pearlin & Schooler, 1978; Ptacek, Smith, & Zanas, 1992; Rogier, 2019). Stressors also differ, with men reporting more work, finance, or friend-related stressors, while women are more likely to report family conflicts such as health issues or children (Matud, 2004). The same author also suggests that women with children were more likely to report dissatisfaction in the workplace than men with children. These results apply broadly to males and females, and are closely related to education status and age. While research has demonstrated there may be significant differences between reported stressors amongst men and women, if one accounts for age and education status, these differences become less apparent (Galanakis, Stalikas, Kallia, Karagianni, & Karela, 2009).

Review of Studies on Stress Levels and Stressors in Farmers

The inherent life disparities between men and women pose challenges in accurately comparing their stress responses. Family duties and financial responsibilities significantly influence individuals' reported experiences, complicating direct comparisons of male versus female stress responses to the same stressor. To mitigate these differences, college serves as a useful setting for studying stress responses, as students have autonomy and responsibilities, while typically lacking the family and financial obligations that emerge later in life. A study conducted in Florida analyzed stress levels and coping mechanisms among 448 college students. As the semester progressed, female students were more likely to report higher scores on the Perceived Stress Scale and moderate stress loads compared to their male counterparts (Graves, Hall, Dias-Karch, Haischer, & Apter, 2021). Notably, no differences were observed between male and female students in perceptions of light or severe stress loads. Differences in coping mechanisms also emerged. There were no coping categories where male students had increased rates over females, while females reported higher rates of self-distraction, venting, emotional support, and instrumental support aligning with previous research suggesting that women tend to employ passive and emotion-based coping strategies (Peetz, Meloff, & Royle, 2023; Ptacek, Smith, & Zanas, 1992).

To delve into stress response disparities among male and female farmers, it's crucial to understand the unique stressors inherent in agriculture. Farmers often experience heightened stress compared to their non-farming counterparts, reflected in an elevated suicide rate compared to the general population (Klingelschmidt, et al., 2018; Kennedy, Cerel, Kheibari, Leske, & Watts, 2021). Factors contributing to this stress include increased mechanization of agriculture coupled with a decreasing workforce, which places greater responsibility and burden on individual farmers (Kennedy, Cerel, Kheibari, Leske, & Watts, 2021). Long working hours without breaks for vacation or weekends are common, compounded by unpredictable weather patterns such as droughts, floods, fires, and temperature fluctuations that can devastate crops without warning. Additionally, farmers face occupational hazards resulting in injuries such as trauma, hearing loss, skin cancer, and pulmonary diseases. Financial uncertainties add to the stress, as farmers navigate fluctuating global markets that dictate the price of their products (Kennedy, Cerel, Kheibari, Leske, & Watts, 2021).

Coping Mechanisms among Farmers

While farming presents its unique set of stressors, farmers' coping mechanisms align closely with those of the general population. One distinctive coping mechanism reported by farmers is detachment from the farm, whether mentally or physically. However, this detachment poses challenges, particularly for farmers who reside on their operations and involve family members in their work. Unlike other occupations where leaving work behind is feasible, the intertwined nature of farming and family life complicates the ability to separate the two realms, highlighting yet another challenge faced by this population (Vayro, Brownlow, Ireland, & March, 2019).

Despite its significance, limited research exists on coping mechanisms among farmers, and gender differences in coping strategies remain largely unexplored, emphasizing the need for this study.

The purpose of this cross-sectional study is threefold:

Examine the variance in negative feelings, perceived stress levels, and stressors experienced by female and male farmers.

Compare the coping strategies employed by female and male farmers to manage stress, identifying potential similarities and disparities in coping mechanisms.

Investigate the influence of gender on suicidal ideation.

Our main hypotheses are as follows:

Women farmers are likely to report higher levels of stress, experience different types of stressors, and utilize coping mechanisms more frequently than their male counterparts.

Women farmers are also more inclined to contemplate suicide.

Through this study, we aim to shed light on the gender-specific stressors and coping mechanisms within the farming community, providing valuable insights for addressing mental health challenges in this population.

Ethical Considerations

This observational cross-sectional study was approved by the IRB of Mercer University under #H2007187, minimal risk research. Participants signed an electronic consent form at the start, despite information being anonymized. All individuals could at any point abandon the study, and their data would get deleted after two weeks. No identifying details were gathered.

Participants

From the original sample of 1,424 farmers, inclusion criteria for this study required participants to have self-reported their gender, resulting in 1,422 respondents. Due to low numbers of gender-diverse respondents (N = 3) and a small number who preferred not to answer (N = 25), these two response categories were excluded from the sample. The final sample consisted of 1,394 farmers, with 67.7% self-identifying as male and 32.3% as female.

The median age of the sample was 42 years old, with women being slightly younger than men (p = 0.003). Notably, a higher proportion of female farmers belonged to Generation Y (born 1982–1994). One-fifth of female farmers identified as Black or African American, a significantly higher proportion compared to their male counterparts (p = 0.009). In terms of citizenship status, 69% of farmers were US citizens, with an additional 19% being legal permanent residents. Interestingly, being an undocumented resident was more prevalent among female farmers, with 12% indicating this status compared to 5% of male farmers (p < 0.001).

Materials

The survey was conducted online using Qualtrics and started with informed consent and a Captcha assignment to discourage multiple responses and bots. The Mercer University's Institutional Review Board for Human Subjects Research approved the study for expedited review (H2007187). We used literature to select suitable participant demographics, stressors, time spent worrying, coping mechanisms for stress, access to healthcare, and the Perceived Stress Scale. The phrase “mental well-being” was used in our study’s descriptions as literature showed that rural residents from the Southeastern USA prefer it over the term “mental health” (Crowe, Averett, Harris, Crumb, & Littlewood, 2019). Feedback was requested and received from various stakeholders, including farmers and farmer groups representing diverse commodities or subsets of farmers (e.g., Southeastern African American Farmers), as well as fellow academics within related sectors. This collaborative approach ensured the survey's relevance and sensitivity to the experiences of the target population.

Perceived Stress Scale

The Perceived Stress Scale (PSS) is a validated scale containing 10 items (Cohen, Kamarck, & Mermelstein, 1983) on stress beyond the respondent’s ability to cope. Higher scores on the scale indicate higher stress levels, with scores ranging from 0 to 40. Participants were asked to indicate how frequently they experienced each item (e.g., hopelessness) over the past 3 months. Principal components analysis revealed that the ten items loaded onto a single component, demonstrating adequate internal consistency with a Cronbach’s alpha of 65.8%, although slightly lower than reported in other literature.

Stressors

Participants were presented with a list of 27 stressors and asked to rate the frequency of their worries about each issue. Examples of stressors included "Non-Farmer Neighbors," "Unexpected Financial Burdens," and "Difficulty Hiring or Managing Employees." Responses were recorded on a 5-point Likert scale ranging from "1. Not at all" to "5. Extremely." For data analysis, responses were simplified into a dichotomous variable, combining "Not at all" and "A little bit" into one category, and "Moderately" through "Extremely" into another. Additionally, aside from identifying the most common stressors that participants worried about moderately, we also calculated the total number of stressors they worried about moderately, which is summarized as "Number of Stressors" in the Findings section.

Negative Feelings

Participants were asked to rate the frequency of experiencing specific feelings or thoughts within the past 3 months using a 5-point Likert scale, ranging from "1. Never" to "5. Daily." The feelings assessed included unhappiness related to their role on the farm, fear of job loss, fear of losing a significant amount of farming income, fear of losing their farm (for farm owners only), feelings of loneliness, nervousness or worry about the future, feelings of sadness or depression, feelings of hopelessness or a persistent lack of happiness, and thoughts of wanting to die by suicide. For questions related to depression and suicidal ideation, we provided two active phone numbers for suicide prevention (Georgia crisis and access line; National Suicide and Crisis Lifeline) both during the question and if participants responded with anything other than "never."

Coping Mechanisms

Participants were presented with a multiple-choice list of potential coping strategies derived from existing literature and insights from farming contacts. They were asked, "Which of the following activities do you currently engage in to alleviate your stress?" Assurances of anonymity and confidentiality were provided. Coping mechanisms included exercising, talking to family or friends, engaging in hobbies, consuming alcohol, watching TV or reading, sleeping, praying or engaging in religious activities, spending time alone or with pets, drawing or painting, meditating, venting anger physically, using cannabis, self-harming, journaling, using illicit drugs, seeking counseling, and using over-the-counter drugs (such as cough medicine). Frequency tables were generated for each coping mechanism, and the total number of different coping mechanisms used by each person to manage stress was calculated to compare the variety of strategies employed.

Procedure

The statewide online survey was conducted from January 1st, 2022, to April 30th, 2022, with participation from individuals in 157 out of 159 counties in Georgia. Recruitment efforts were facilitated by the Georgia Foundation for Agriculture's extensive connections with commodity and farming groups. Marketing materials, available in both English and Spanish, prominently featured the logos of the Georgia Foundation for Agriculture and the Georgia Rural Health Innovation Center, along with a QR code and website link leading to the survey. Requests for participation were targeted at Georgia farm owners, farm workers, or spouses of farmers. Although an incentive of a $10 Visa gift card was offered, fewer than 10% of participants opted to receive the gift card. Advertisements were disseminated through the Georgia Associated Press Network, agricultural magazines, and the Market Bulletin. Additionally, direct mail postcards were distributed to Georgia Farm Bureau members by its president, Tom McCall. Social media platforms were also utilized to target specific farm groups for advertising.

Data analysis was conducted using JMP Pro 17.0.0. All statistical tests employed an alpha threshold of 5%. Pearson's Chi-Square tests were used for categorical variables, independent t-tests for parametric continuous data (Mean, SD), and Mann-Whitney tests for non-parametric continuous data (Median, IQR).

For the last section, a stratified propensity score matching technique was employed to estimate the role of gender in suicidal ideation. Propensity scores, representing the probability of belonging to each gender group, were estimated using logistic regression models separately for male and female groups. The dependent variable was the 'treatment' variable (male or female), and the independent variables included relevant covariates that could potentially influence both the treatment assignment (gender) and the outcome variable (suicidal ideation). Covariates included age, education, number of children, farm role, first-generation status, farming experience, Perceived Stress Scale score, feelings of sadness or depression, unhappiness with one's role, concerns about home and work-life balance, parenting, partner problems, financial issues (e.g., financial burdens, retirement savings, succession planning), experiences of microaggressions, obtaining training certificates, interactions with non-farmer neighbors, sexual issues, experiences of racism, alcohol use, and the number of coping mechanisms employed. Propensity scores were categorized into four groups using Jenks Natural Breaks cut points. Within each group, propensity score matching was performed using the nearest neighbor matching algorithm with a caliper of 0.5. Balance was assessed using standardized differences, with independent t-tests (or Mann-Whitney tests) and Pearson Chi-Square tests conducted for the pre-matching portion of the table, and paired t-tests (or Wilcoxon’s tests) and McNemar Chi-Square tests for the post-matching portion of the table.

The findings contain results on the basic demographics of the sample, work characteristics, emotions and stress, suicidal ideation, and coping mechanisms for stress.

Demographics

Table 1 presents the main demographics of the entire farmer sample but is also split up by gender. Female farmers were slightly younger, with a higher proportion belonging to Generation Y (p = 0.003 and p = 0.047, respectively). Education differed slightly between male and female farmers, with a higher proportion of females having some college experience (48% compared to 41% in males) and lower proportions of females who did not graduate high school (6% compared to 9% in males), as well as those who received a high school diploma or equivalent (p = 0.006), suggesting a slightly higher level of education among female farmers. Marriage or cohabitation rates were similar between genders (roughly 88%, p = 0.470), as were rates of parenthood (73% have children, p = 0.102). However, male farmers had slightly more children on average, with a slightly higher Wilcoxon Rank Sum despite similar median and interquartile range values (p = 0.006).

[Table 1 ABOUT HERE}

Work Characteristics

"Work characteristics" refer to the occupation of the farmer and the attributes of their farms, detailed in Table 2. Female farmers were less likely to be farm owners (40% compared to 46% in men) and more likely to be farm managers (28% compared to 20% in men), indicating differences in managerial roles by gender (p = 0.001).

In addition, a higher proportion of female farmers identified as first-generation farmers (68%) compared to male farmers (54%, p < 0.001), The genders also differ when it comes to how long they have been in their current role: women are more likely to have been in their role less than five years (52% compared to 27% of male farmers), while male farmers are more likely to have five or more years of experience (73% of male farmers compared to 48% of female farmers), p < 0.001. About a quarter of female farmers reported farming over 50 hours a week, which was significantly lower than the 36% of male farmers who farm 50 hours or more per week, p < 0.001.

There were no gender differences in the number of farms operated (p = 0.608) or in the rural-urban composition of farms (p = 0.692). However, female farmers had a marginally lower proportion of smaller farms (0-100 acres) compared to male farmers (23% compared to 28%, p = 0.038). For commodity type, we split up the commodities into Row Crops (such as corn), Livestock (such as hogs), and Fibers (cotton and wool). As farmers often have several commodity types on the same farm or spread across different farms, the numbers do not add up to 100%. Female farmers had a lower proportion involved in row crops (63% compared to 72% of male farmers, p = 0.002) but were equally likely to farm livestock (68%, p = 0.128). Conversely, male farmers were more likely to farm fiber crops (p < 0.001) and were overall more likely to farm all three commodity types (9% compared to 3% of female farmers, p < 0.001).

[Table 2 ABOUT HERE]

Emotions and Stress

Emotions and stress were assessed using the Perceived Stress Scale (PSS) and self-reported measures of negative feelings, as detailed in Table 3. Stress was measured using the PSS scale, with higher scores meaning higher stress levels. The median PSS score was 19, indicating moderately high-stress levels, with no significant gender difference observed (p = 0.435). Gender comparisons revealed no differences in feelings of fear related to job loss, significant income loss, farm loss, loneliness, nervousness about the future, hopelessness, or lack of happiness (all p > 0.05). However, male farmers were more likely to often (once a week or more) feel unhappy with their role on the farm compared to female farmers (24% vs. 18%, p = 0.009). A higher proportion of female farmers reported feeling sad or depressed 5–12 times per year compared to male farmers (36% vs. 26%, p < 0.001).

[Table 3 ABOUT HERE]

Suicidal Ideation

Respondents were asked about the frequency of thoughts about dying by suicide; detailed in Table 4. Approximately 43% reported having at least one thought of wanting to die by suicide in the past year. Female farmers were more likely than male farmers to report thinking about it at least once a month (19% vs. 12%) or daily (8% vs. 6%, p = 0.002).

[Table 4 ABOUT HERE]

Stressors

Respondents were asked how intensely they worried about a list of 27 stressors; the full list can be found in Table 5. There are three types of effects to compare: the number of stressors causing at least a moderate amount of stress, the top 5 stressors, and the proportion that each stressor is experienced as moderately high or higher. For the first, there was no significant gender difference observed in the number of stressors. For the second, the top stressors for the entire sample included balancing home and work life, weather effects on income, COVID-19 impact on income, saving for retirement, and unexpected financial burdens. The only observed differences in top stressors between male and female farmers included female farmers not reporting weather effects on income unlike their male counterparts. For the third effect, weather effects on income, lack of opportunities, managing debt, community discrimination, and lack of role models showed no gender differences in proportion of who experienced it as at least moderate stress (all p > 0.05). Female farmers more often expressed concern about home and work-life balance (including parenting and partner problems), financial issues, micro-aggressions, obtaining training certificates, and non-farmer neighbors (all p < 0.05). Conversely, male farmers reported greater worry about sexual issues, experiencing racism, and alcohol use (all p < 0.05). The other non-significant stressors can be found in Table 5.

[Table 5 ABOUT HERE]

Coping Mechanisms for Stress

We also inquired about how the respondents cope with stress; the results are in Table 6. Male farmers exhibited a higher variety of coping mechanisms compared to females, with a mean of 3 for males versus 2 for females (p < 0.001).

The top 5 coping mechanisms were consistent across the entire sample, with exercising and walking ranking as the most prevalent. However, differences emerged between genders in the secondary coping strategies. For females, talking to family or friends ranked second, while engaging in a hobby was second among males. Watching TV or reading was the third most common coping mechanism for females, whereas, for males, it was drinking alcohol.

A greater proportion of females relied on exercising, walking, and journaling as coping mechanisms (all p’s < 0.010), whereas more males engaged in hobbies, alcohol consumption, sleep, and solitary activities to manage stress (all p’s < 0.010).

[Table 6 ABOUT HERE]

THE ROLE OF GENDER IN SUICIDAL IDEATION OF FARMERS

In previous sections, we noted a slight disparity between male and female farmers in thoughts of suicide. However, given the observational nature of our study and the lack of controls, we employed a stratified (gender-specific) propensity score matching technique to control for confounding variables. Detailed analysis is provided in the Methods section.

Figure 1 illustrates the distribution of propensity scores among males and females across four bins, showcasing comparable distributions post-matching. Following matching, we assessed the balance of covariates between the treated and control group within each gender stratum to ensure that the matching has successfully created comparable comparison groups in terms of the covariates. Table 7 demonstrates successful balance, as most post-matching results are not significant.

[FIGURE 1 HERE]

[Table 7 ABOUT HERE]

With covariates balanced post-matching, a McNemar test examined the relationship between gender and suicidal ideation. A McNemar-Bowker Chi-Square analysis shows that after matching, there is no difference in suicidal ideation in the past year between males and females, X2(1) = 0.778, p = 0.378. On average, 53% of the matched sample had thought about dying by suicide at least once in the past year.

In this study, we presented the findings of a cross-sectional survey involving 1,424 farmers, approximately one-third of whom self-identified as female. The aim was to investigate differences between male and female farmers. This sample revealed several noteworthy findings.

Concerning demographics and work characteristics, female farmers were slightly more educated than their male counterparts. They were also more likely to be first-generation farmers, less likely to be farm owners, and more inclined towards farm management roles. Additionally, female farmers reported less experience in their current farm roles and worked slightly fewer hours compared to male farmers. The variance in experience levels and distinct farm roles could manifest as elevated stress levels or a unique array of stressors (Kallioniemi, Simola, Kaseva, & Kymalainen, 2016).

In the current study, the Perceived Stress Score of 19 was statistically similar between genders, yet notably higher than the established means from the US general population (12.1 and 13.7 for males and females respectively) (Cohen & Williamson, 1988). Male farmers more frequently expressed unhappiness in their farm role, while a substantially higher number of female farmers (36%) reported feeling depressed 5–12 times per year, in contrast to 26% of male farmers, suggesting that unhappiness in male farmers does not always manifest to a depressive mood. Moreover, females exhibited a marginally higher frequency of suicidal ideation, potentially influenced by job-related characteristics (Loerbroks, et al., 2016).

Although the stressors were generally similar across genders, female farmers were more worried about life-work balance and financial issues while males were more worried about sexual issues and their own alcohol use. The top five coping mechanisms were similar between male and female farmers, with some exceptions. Male farmers displayed a wider variety of coping mechanisms, with alcohol consumption in the top three. This aligns with the literature that shows that males often use emotional inhibition and alcohol over emotion-focused behaviors (Pearlin & Schooler, 1978; Ptacek, Smith, & Zanas, 1992; Cooper, Frone, Russell, & Mudar, 1995; Dvorak, et al., 2014).

To understand the potential gender effect on suicidal ideation, a stratified propensity score matching analysis was conducted. Matching male and female farmers based on important covariates (such as job characteristics, stress levels, etc.) revealed that the gender difference in suicidal ideation disappeared. This implies that observed gender differences in suicidal ideation may stem from differences in job characteristics, rather than being a result of individuals identifying as male or female. These job characteristics can be heavily influenced by society’s gender roles, implying gender roles functioning as a potential mediator. Notably, the higher prevalence of first-generation farmers among females in our sample may significantly contribute to this gender-specific outcome, consistent with prior research indicating elevated suicide risk among first-generation farmers. This observation is in line with the 2023 study conducted by Montgomery et al (Montgomery, Basey, Baucom, & Scoggins, 2024), which demonstrated that first-generation farmers are 4.6 times more likely to contemplate suicide compared to generational farmers.

Limitations

Cross-sectional studies cannot determine causality, prompting the need for longitudinal research to follow male and female farmers over time. Such studies could examine how stress levels and stressors evolve, and whether gender differences persist or change with these dynamics. Longitudinal investigations could offer valuable insights into the long-term impact of stress on farmers' mental health and well-being.

This study lacked detailed information on the frequency and context of coping mechanism use. Conducting qualitative investigations into coping strategies among male and female farmers could provide a deeper understanding of coping patterns, potentially revealing gender-specific approaches to stress management. Furthermore, it is important to note that this study did not directly assess gender roles as a latent concept. Instead, the study focused on gathering data related to an extensive list of demographics, job characteristics, and other relevant measures. To gain a more nuanced understanding of the potential pressure farmers might experience from societal gender roles, it could be beneficial to conduct in-depth qualitative research.

A last caveat is that the success of propensity score matching relies on the availability of suitable covariates and a sufficient sample size to achieve an adequate balance between the treated and control groups within each gender stratum. Although we had a sufficient sample size, it is not possible to determine if we have all the suitable covariates included in our study. Additionally, the choice of specific covariates can impact the validity of the results.

Despite the significance of first-generation farmers, particularly females, there is a notable lack of literature on their experiences. Further research could delve into the specific challenges encountered by first-generation female farmers, building upon insights from prior studies like Montgomery et al.'s work (Montgomery, Basey, Baucom, & Scoggins, 2024). Understanding these factors could help design support systems and resources to enhance the success and mental well-being of female farmers in the early stages of their new farming careers.

Funding: Funded by Georgia Rural Health Innovation Center, with support by Georgia Foundation for Agriculture, The Pittuloch Foundation, and Georgia Farm Bureau.

Disclosures: All authors working for Mercer University’s Georgia Rural Health Innovation Center (a state-grant funded organization focused on improving rural health in GA) received salaries from the Center, but were freely able to collect, analyze, and interpret the data without interference from the employer.

No financial or non-financial competing interests are declared.

Acknowledgements:

We would like to thank all the farmers who wanted to share their information with us. Lily Baucom was crucial in helping us connect to the farmers. In addition, Mercer University School of Medicine provided us with summer scholar student Taylore Brooks who helped design an initial part of the survey, and doctoral research assistant Dahlia Al-Haleem helped revise the instrument. We would also like to thank the medical students Claire Belay, Michal Kshatri, and Pedro Diaz for their translation of the collection instrument into Spanish. Dr. Jean Sumner and Glenda Grant have both been invaluable in their support of this project.

Author Contribution

A.M. wrote the main manuscript text, analyzed the data, and interpreted the data. J.D. contributed significantly to the literature review and discussion. J.D., S.B., and C.S. revised the work substantively. All authors reviewed the manuscript.

Bem, S. (1981). Gender Schema Theory: A Cognitive Account of Sex Typing. Psychological Review, 88(4), 354-364. doi:10.1037/0033-295X.88.4.354
Bjornestad, A., Cuthbertson, C., & Hendricks, J. (2021). An Analysis of Suicide Risk Factors Among Farmers in the Midwestern United States. Int J Environ Res Public Health, 18(7), 3563. doi:10.3390/ijerph18073563
Brown, G., & Harris, T. (1978). Social Origins of Depression. London: Free Press.
Cohen, S., & Williamson, G. (1988). Perceived stres in a probability sample of the United States. In S. Spacapan, & S. Oskamp, The Social Psychology of Health (pp. 31-67). Newbury Park: Sage.
Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress. J Health Soc Behav., 24, 386-396. doi:10.2307/2136404
Cooke, R., & Rousseau, D. (1984). Stress and strain from family roles and work-role expectations. Applied Psychology, 69(2), 252-260. doi:https://psycnet.apa.org/doi/10.1037/0021-9010.69.2.252
Cooper, M., Frone, M., Russell, M., & Mudar, P. (1995). Drinking to Regulate Positive and Negative Emotions: A Motivational Model of Alcohol Use. Journal of Personality and Social Psychology, 69(5), 990-1005. doi:10.1037/0022-3514.69.5.990
Crowe, A., Averett, A., Harris, J., Crumb, L., & Littlewood, K. (2019). In my own words: Exploring Definitions of mental health in the rural Southeastern United States. Professional Counselor, 9(4), 381-395. doi:https://psycnet.apa.org/doi/10.15241/ac.9.4.381
Duxbury, L., & Higgins, C. (1991). Gender differences in work-family conflict. Journal of Applied Psychology, 76(1), 60-74. doi:https://psycnet.apa.org/doi/10.1037/0021-9010.76.1.60
Dvorak, R., Sargent, E., Kilwein, T., Stevenson, B., Kuvaas, N., & Williams, T. (2014). Alcohol use and alcohol-related consequences: associations with emotion regulation difficulties. The American Journal of Drug and Alcohol Abuse, 40(2), 125-130. doi:10.3109/00952990.2013.877920
Galanakis, M., Stalikas, A., Kallia, H., Karagianni, C., & Karela, C. (2009). Gender differences in experiencing occupational stress: the role of age, education, and marital status. Stress and Health, 25, 397-404. doi:10.1002/smi.1248
Graves, S., Hall, M., Dias-Karch, C., Haischer, M., & Apter, C. (2021). Gender differences in perceived stress and coping among college students. Plos One, 16(8), e0255634. doi:10.1371/journal.pone.0255634
Hansen, C., Jensen, P., & Skovsgaard, C. (2015). Modern Gender Roles and Agricultural History: The Neolithic Inheritance. Journal of Economic Growth, 20, 365-404. doi:https://doi.org/10.1007/s10887-015-9119-y
Kallioniemi, M., Simola, A., Kaseva, J., & Kymalainen, H.-R. (2016). Stress and Burnout Among Finish Dairy Farmers. Journal of Agromedicine, 21(3), 259-267. doi:10.1080/1059924X.2016.1178611
Kennedy, A., Cerel, J., Kheibari, A., Leske, S., & Watts, J. (2021). A comparison of farming- and non-farming-related suicides from the United States’ National Violent Deaths Reporting System, 2003–2016. Suicide and life-threatening behavior, 51(3), 504-514. doi:https://doi.org/10.1111/sltb.12725
Klingelschmidt, J., Milner, A., Khireddine-Medouni, I., Witt, K., Alexopoulos, E., Toivanen, S., . . . Niedhammer, I. (2018). Suicide among agricultural, forestry, and fishery workers: a systematic literature review and meta-analysis. Scand J Work Environ Health., 44(1), 3-15. doi:10.5271/sjweh.3682
Loerbroks, A., Cho, S., Dollard, M., Zou, J., Fischer, J., Jiang, Y., . . . Li, J. (2016). Associations between work stress and suicidal ideation: Individual-participant data from six cross-sectional studies. Journal of Psychomatic Research, 90, 62-69. doi:10.1016/j.jpsychores.2016.09.008
Matud, M. (2004). Gender differences in stress and coping styles. Personality and individual differences, 37(7), 1401-1415. doi:https://doi.org/10.1016/j.paid.2004.01.010
Montgomery, A., Basey, S., Baucom, L., & Scoggins, C. (2024). Stress and suicidal ideation among first-generation farmers: A cross-sectional study with 1,288 farmers in Georgia. The Journal of Rural Health, 40, 75-86. doi:10.1111/jrh.12764
Pearlin, L., & Schooler, C. (1978). The Structure of Coping. Journal of health and social behavior, 2-21. Retrieved from https://www.jstor.org/stable/2136319
Peetz, J., Meloff, Z., & Royle, C. (2023). When couples fight about money, what do they fight about? Journal of Social and Personal Relationships, 40(11), 3723-3751. doi:https://doi.org/10.1177/02654075231187897
Ptacek, J., Smith, R., & Zanas, J. (1992). Gender, appraisal, and coping: a longitudinal analysis. Journal of Personality, 60(2), 747-770. doi:https://doi.org/10.1111/j.1467-6494.1992.tb00272.x
Reilly, M. (2021). Health disparities and access to healthcare in rural vs urban areas. Theory in Action, 14(2), 6-27. doi:10.3798/tia.1937-0237.2109
Rogier, G. G. (2019). Is Emotional Suppression Always Bad? A Matter of Flexibility and Gender Differences. Current Psychology, 38, 411-420. doi:10.1007/s12144-017-9623-7
Sachs, C., Barbercheck, M., Brasier, K., Kiernan, N., & Terman, A. (2016). The Rise of Women Farmers and Sustainable Agriculture. Iowa City, Iowa, United Sstates of America: University of Iowa Press. doi:https://doi.org/10.2307/j.ctt20p57gr
Starr, C., & Zurbriggen, E. (2017). Sandra Bem's Gender Schema Theory After 34 Years: A Review of its Reach and Impact. Sex Roles, 76, 566-578. doi:doi.org/10.1007/s11199-016-0591-4
United Nations. (2020, October 20). The World's Women 2020: Trends and Statistics. Retrieved 01 19, 2024, from https://worlds-women-2020-data-undesa.hub.arcgis.com/
US Department of Labor. (2022). Civilian Labor Force by Sex. Retrieved January 19, 2024, from https://www.dol.gov/agencies/wb/data/lfp/civilianlfbysex
USDA National Agricultural Statistics Service. (2017). The 2017 Census of Agriculture - State data on Georgia. Retrieved 12 13, 2023, from nass.usda.gov
Vayro, C., Brownlow, C., Ireland, M., & March, S. (2019). 'Farming is not Just an Occupation [but] a Whole Lifestyle’: A Qualitative Examination of Lifestyle and Cultural Factors Affecting Mental Health Help-Seeking in Australian Farmers. Farming and Rural Life, 60(1), 151-173. doi:https://doi.org/10.1111/soru.12274

Table 1: Demographics of Entire Sample, Female Farmers, and Male Farmers.

DEMOGRAPHICS

Entire Sample

N=1,394

(100%)

Female Farmers
N=450

(32.3%)

Male Farmers
N=944

(67.7%)

p-value

Age years

Median (IQR)

Gen Z

Gen Y

Gen X

Boomers

Silent Generation

42 (35 - 49)

30 (2.3%)

549 (44.8%)

567 (44.2%)

108 (8.4%)

30 (2.3%)

41 (35-49)

14 (3.3%)

200 (47.2%) ^a

173 (40.8%)

30 (7.1%)

7 (1.7%)

43 (36-50)

16 (1.9%)

349 (40.6%) ^a

394 (45.8%)

78 (9.1%)

23 (2.7%)

0.003

0.047

Race

Black/African American

White

Other

227 (16.3%)

1017 (72.9%)

150 (10.8%)

90 (20.0%)

312 (69.3%)

48 (10.7%)

137 (14.5%)

705 (74.7%)

102 (10.8%)

0.009

0.044

0.978

Education Level

Did not graduate High School

High school or GED

Vocational Technical Trade

Some college

College or higher

112 (8.1%)

240 (17.2%)

193 (13.9%)

601 (43.2%)

246 (17.7%)

27 (6.0%) ^a

58 (13.0%) ^b

68 (15.2%)

213 (47.5%) ^c

82 (18.3%)

85 (9.0%) ^a

182 (19.3%) ^b

125 (13.2%)

388 (41.1%) ^c

164 (17.4%)

0.006

Marital Status

Single, divorced, separated

Married, living together

161 (11.6%)

1230 (88.4%)

56 (12.5%)

393 (87.5%)

105 (11.2%)

837 (88.9%)

0.470

Has Children

Yes N (%)

970 (73.4%)

333 (76.2%)

637 (72.0%)

0.102

Number of Children

Median (IQR)

1 (1-2)

1 (1-2)⁺

0.006

Bolded p-values are significant at a threshold of 5%. Categorical variables used Pearson’s Chi Square Test or Fischer’s Exact Test where applicable. Continuous variables used an independent t-test (reported with means) or Mann-Whitney (reported with medians). The same superscript letters show statistically different answer categories. Superscript plus indicates that value is higher based on mean score for Wilcoxon Rank Sum. IQR = interquartile range; GED = General Educational Development test.

Table 2: Work Characteristics of Entire Sample, Female Farmers, and Male Farmers.

WORK CHARACTERISTICS

Entire Sample

N=1,394

(100%)

Female Farmers
N=450

(32.3%)

Male Farmers
N=944

(67.7%)

p-value

Role

Farm worker

Farm manager/supervisor

Farm owner

467 (33.5%)

310 (22.2%)

617 (44.3%)

145 (32.2%)

126 (28.0%)^a

179 (39.8%) ^b

322 (34.1%)

184 (19.5%) ^a
438 (46.4%) ^b

0.001

Generational Status

First-Generation Farmers

Generational Farmer

738 (58.4%)

525 (41.6%)

266 (67.9%)

126 (32.1%)

472 (54.2%)

399 (45.8%)

<0.001

How long in current role?

<1 year

1-4

5 – 10

11 – 20

21+

15 (1.2%)

439 (33.7%)

424 (32.6%)

259 (19.9%)

165 (12.7%)

8 (2.0%) ^a

199 (49.6%) ^b

103 (25.7%) ^c

60 (15.0%) ^d

31 (7.7%) ^e

7 (0.8%) ^a

240 (26.6%) ^b

321 (35.6%) ^c

199 (22.1%) ^d

134 (14.9%) ^e

<0.001

Hours of Farming per Week

Less than 50 hours

Equal or more than 50 hours

858 (63.6%)

417 (36.4%)

290 (75.9%)

92 (24.1%)

568 (63.6%)

325 (36.4%)

<0.001

Number of Farms

Median (IQR)

1 (1 – 2)

2 (1 – 2)

0.608

Rurality of Farms

Only rural farms

Only urban farms

Mix of rural and urban farms

889 (70.5%)

314 (24.9%)

58 (4.6%)

270 (72.2%)

88 (23.5%)

16 (4.3%)

619 (69.8%)

226 (25.5%)

42 (4.7%)

0.692

Farm Operation Size

0-100 acres

101-500 acres

501+ acres

345 (26.5%)

738 (56.7%)

219 (16.8%)

91 (22.7%)^a

248 (61.9%)

62 (15.5%)

254 (28.2%)^a

490 (54.4%)

157 (17.4%)

0.038

Commodity Type

Row Crop

Livestock

Fiber

873 (69.0%)

858 (67.8%)

240 (21.5%)

245 (62.8%)

114 (70.8%)

51 (16.0%)

628 (71.7%)

582 (66.4%)

189 (23.7%)

0.002

0.128

0.005

Different Commodity Types

One Commodity Type

Two Commodity Types

Three Commodity types

649 (51.3%)

529 (41.8%)

88 (7.0%)

219 (56.2%)

160 (41.0%)

11 (2.8%) ^a

430 (49.1%)

369 (42.1%)

77 (8.8%) ^a

<0.001

Bolded p-values are significant at a threshold of 5%. Categorical variables used Pearson’s Chi Square Test or Fischer’s Exact Test where applicable. Continuous variables used an independent t-test (reported with means) or Mann-Whitney (reported with medians). The same superscript letters show statistically different answer categories. IQR = interquartile range.

Table 3: Emotions, Stress, and Coping Mechanisms of the Entire Sample, Female Farmers, and Male Farmers.

EMOTIONAL CHARACTERISTICS

Entire Sample

N=1,394

(100%)

Female Farmers
N=450

(32.3%)

Male Farmers
N=944

(67.7%)

p-value

PSS

Median (IQR)

19 (16 – 22)

19 (16-22)

0.435

Feeling unhappy with role

Rarely (0–4 times/year)

Sometimes (5–12 times /year)

Often (1/ week or more)

714 (51.6%)

369 (26.7%)

301 (21.8%)

228 (51.0%)

139 (31.1%)^a

80 (17.9%)^b

486 (51.9%)

230 (24.6%)^a

221 (23.6%)^b

0.009

Scared to lose their job

Rarely (0–4 times/year)

Sometimes (5–12 times /year)

Often (1/ week or more)

674 (52.3%)

314 (24.3%)

302 (23.4%)

215 (54.4%)

86 (21.8%)

94 (23.8%)

459 (52.3%)

228 (25.5%)

208 (23.2%)

0.350

Scared to lose significant income

Rarely (0–4 times/year)

Sometimes (5–12 times /year)

Often (1/ week or more)

681 (49.3%)

355 (25.7%)

345 (25.0%)

206 (46.4%)

126 (28.4%)

112 (25.2%)

475 (50.7%)

229 (24.4%)

233 (24.9%)

0.227

Scared to lose farm

Rarely (0–4 times/year)

Sometimes (5–12 times /year)

Often (1/ week or more)

380 (61.1%)

104 (16.7%)

138 (22.2%)

104 (57.1%)

34 (18.7%)

44 (24.2%)

276 (62.7%)

70 (15.9%)

94 (21.4%)

0.426

Feeling lonely

Rarely (0–4 times/year)

Sometimes (5–12 times /year)

Often (1/ week or more)

714 (52.0%)

338 (24.6%)

322 (23.4%)

225 (51.1%)

120 (27.3%)

95 (21.6%)

489 (52.4%)

218 (23.3%)

227 (24.3%)

0.234

Feeling nervous or worried about the future

Rarely (0–4 times/year)

Sometimes (5–12 times /year)

Often (1/ week or more)

619 (45.0%)

362 (26.3%)

396 (28.8%)

187 (42.4%)

131 (29.7%)

123 (27.9%)

432 (46.2%)

231 (24.7%)

273 (29.2%)

0.136

Feeling sad or depressed

Rarely (0–4 times/year)

Sometimes (5–12 times/year)

Often (1/week or more)

665 (48.4%)

404 (29.4%)

306 (22.3%)

194 (44.1%)

159 (36.1%) ^a

87 (19.8%)

471 (50.4%)

245 (26.2%) ^a

219 (23.4%)

<0.001

Feeling hopeless or like you will never be happy

Rarely (0–4 times/year)

Sometimes (5–12 times/year)

Often (1/week or more)

832 (60.6%)

277 (20.2%)

265 (19.3%)

273 (62.2%)

91 (20.7%)

75 (17.1%)

559 (59.8%)

186 (19.9%)

190 (20.3%)

0.366

Bolded p-values are significant at a threshold of 5%. Categorical variables used Pearson’s Chi Square Test or Fischer’s Exact Test where applicable. Continuous variables used an independent t-test (reported with means) or Mann-Whitney (reported with medians). The same superscript letters show statistically different answer categories. PSS = Perceived Stress Scale; IQR = interquartile range.

Table 4: Suicidal Ideation in Entire Sample, Female Farmers, and Male Farmers.

SUICIDAL IDEATION

Entire Sample

N=1,394

(100%)

Female Farmers
N=450

(32.3%)

Male Farmers
N=944

(67.7%)

p-value

Thoughts of wanting to die by suicide

Never

1–4 times/year

At least once a month

At least once a week

Daily

795 (57.2%)

167 (12.1%)

193 (14.0%)

149 (10.8%)

76 (5.5%)

235 (53.2%)

50 (11.3%)

83 (18.8%) ^a

39 (8.8%)

35 (7.9%) ^b

560 (59.7%)

117 (12.1%)

110 (11.7%) ^a

110 (11.7%)

41 (5.5%) ^b

0.002

Bolded p-values are significant at a threshold of 5%. Categorical variables used Pearson’s Chi Square Test. Same superscript letters show statistically different answer categories.

Table 5: Stressors of Female Farmers and Male Farmers.: % Moderately Stressed

STRESSORS (% Moderately Stressed)	Entire Sample N=1,394 (100%)	Female Farmers N=450 (32.3%)	Male Farmers N=944 (67.7%)	p-value
Nr of Stressors Mod. Stressed Median (IQR)	14 (8 – 18)	14 (10-18)	14 (6-19)	0.498
Balancing Home-Work Life N (%)	823 (62.4%)	298 (71.6%)	525 (58.2%)	<0.001
Weather and effect on income N (%)	821 (62.2%)	252 (60.1%)	569 (63.2%)	0.283
COVID-19 Affecting Income N (%)	779 (58.7%)	285 (67.5%)	494 (54.6%)	<0.001
Saving for Retirement N (%)	770 (58.4%)	273 (64.5%)	497 (55.5%)	0.002
Unexpected Financial Burdens N (%)	754 (58.1%)	250 (63.0%)	504 (56.0%)	0.019
Succession N (%)	717 (55.2%)	246 (61.4%)	471 (52.4%)	0.003
Health Issues N (%)	721 (54.6%)	246 (57.9%)	475 (53.1%)	0.101
Changes in Ag Laws Policies N (%)	714 (53.8%)	241 (56.4%)	473 (52.5%)	0.178
Catching COVID-19 N (%)	709 (53.5%)	237 (56.0%)	472 (52.3%)	0.201
Microaggressions N (%)	680 (52.2%)	236 (56.9%)	444 (50.0%)	0.021
Operation Certifications N (%)	674 (51.9%)	218 (53.3%)	456 (51.2%)	0.489
Obtaining Training Certificates N (%)	658 (51.9%)	241 (58.9%)	417 (48.5%)	<0.001
Lack of Opportunities for Ppl like me N (%)	656 (51.1%)	205 (51.0%)	451 (51.1%)	0.963
Problems w Partner N (%)	666 (50.9%)	235 (56.6%)	431 (48.2%)	0.005
Problems Parenting N (%)	638 (50.4%)	225 (58.0%)	413 (47.1%)	<0.001
Hiring or Managing Ppl N (%)	647 (50.2%)	186 (47.0%)	461 (51.6%)	0.128
Managing Debt N (%)	653 (49.9%)	220 (52.4%)	433 (48.7%)	0.208
Non-Farmer Neighbors N (%)	638 (49.3%)	215 (54.4%)	423 (47.1%)	0.015
Not Able to Provide N (%)	646 (49.2%)	218 (52.0%)	428 (47.8%)	0.155
Discrimination in Community N (%)	631 (49.1%)	196 (50.9%)	435 (48.4%)	0.408
Problems w Friends N (%)	624 (48.3%)	203 (49.8%)	421 (47.7%)	0.488
Sexual Issues N (%)	614 (47.8%)	178 (43.7%)	436 (49.7%)	0.046
Injuries on the Job N (%)	628 (47.8%)	201 (48.6%)	427 (47.4%)	0.709
Lack of Role Models N (%)	626 (47.6%)	209 (50.0%)	417 (46.5%)	0.242
Racism in Community N (%)	617 (47.1%)	191 (45.3%)	426 (47.9%)	0.368
Directly Experiencing Racism N (%)	610 (46.5%)	175 (41.8%)	435 (48.8%)	0.018
Own Alcohol Use N (%)	590 (45.8%)	172 (42.0%)	418 (47.6%)	0.060

Bolded p-values are significant at a threshold of 5%. Categorical variables used Pearson’s Chi Square Test or Fischer’s Exact Test where applicable. Continuous variables used an independent t-test (reported with means) or Mann-Whitney (reported with medians). IQR = interquartile range.

Table 6: Coping Mechanisms of the Entire Sample, Female Farmers, and Male Farmers.

COPING MECHANISMS	Entire Sample N=1,394 (100%)	Female Farmers N=450 (32.3%)	Male Farmers N=944 (67.7%)	p-value
Nr of Different Coping Mechanisms Median (IQR)	3 (1 – 4)	2 (1 - 4)	3 (1 - 4)	<0.001
Exercise/walk N (%)	567 (40.7%)	205 (45.6%)	362 (38.4%)	0.010
Talk to family or friends N (%)	445 (31.9%)	153 (34.0%)	292 (30.9%)	0.251
Engage in a hobby N (%)	406 (29.1%)	107 (23.8%)	299 (31.7%)	0.002
Drink alcohol N (%)	394 (28.3%)	98 (21.8%)	296 (31.4%)	<0.001
Watch TV or read N (%)	381 (27.3%)	109 (24.2%)	272 (28.8%)	0.072
Sleep N (%)	325 (23.3%)	84 (18.7%)	241 (25.5%)	0.005
Pray or other religious acts N (%)	279 (20.0%)	78 (17.3%)	201 (21.3%)	0.084
Spend time alone N (%)	252 (18.1%)	56 (12.4%)	196 (20.8%)	<0.001
Spend time with pets N (%)	234 (16.8%)	67 (14.9%)	167 (17.7%)	0.191
Draw or paint N (%)	170 (12.2%)	52 (11.6%)	118 (12.5%)	0.614
Meditate N (%)	131 (10.2%)	35 (7.8%)	96 (10.2%)	0.153
Hit or kick things N (%)	89 (6.4%)	21 (4.7%)	68 (7.2%)	0.070
Cannabis N (%)	80 (5.7%)	28 (6.2%)	52 (5.5%)	0.592
Hit or injure myself N (%)	75 (5.4%)	21 (4.7%)	54 (5.7%)	0.415
Write in journal N (%)	73 (5.2%)	44 (9.8%)	29 (3.1%)	<0.001
Talk to counselor N (%)	59 (4.2%)	16 (3.6%)	43 (4.6%)	0.386
Other illicit drugs N (%)	54 (3.9%)	15 (3.3%)	39 (4.1%)	0.470
OTC drugs N (%)	31 (2.2%)	10 (2.2%)	21 (2.2%)	0.998

Bolded p-values are significant at a threshold of 5%. Categorical variables used Pearson’s Chi Square Test or Fischer’s Exact Test where applicable. Continuous variables used an independent t-test (reported with means) or Mann-Whitney (reported with medians). The same superscript letters show statistically different answer categories. IQR = interquartile range.

Table 7: Characteristics before and after gender-specific propensity score matching

	Before Propensity Matching			After Propensity Matching
CONFOUNDING VARIABLES	MALE FARMERS N=944	FEMALE FARMERS N=450	p-value	MALE FARMERS N=150	FEMALE FARMERS N=150	p-value
Age Median (IQR)	43 (36 – 50)	41 (35 – 49)	0.003	42 (35.8 – 47)	43 (36.8 – 49)	0.241
Education Did not graduate HS HS graduate Some college+	85 (8.9%) 309 (32.2%) 565 (58.9%)	27 (5.8%) 129 (27.8%) 308 (66.4%)	0.014	10 (6.7%) 41 (27.3%) 99 (66.0%)	12 (8.0%) 45 (30.0%) 93 (62.0%)	0.758
Nr of Children Median (IQR)	1 (1-2)	1 (1-2)	0.018	1 (1-2)	1 (1-2)	0.689
First Generation Yes N (%) No N (%)	472 (54.2%) 399 (45.8%)	266 (32.1%) 126 (67.9%)	<0.001	115 (76.8%) 35 (23.3%)	11 (74.0%) 39 (26.0%)	0.592
Role Farm worker N (%) Farm manager N (%) Farm owner N (%)	322 (34.1%) 184 (19.5%) 438 (46.4%)	145 (32.2%) 126 (28.0%) 179 (39.8%)	0.001	47 (31.3%) 52 (34.7%) 51 (34.0%)	46 (30.7%) 52 (34.7%) 52 (34.7%)	0.990
Experience Beginner <5y Experienced 5y+	568 (63.0%) 333 (37.0%)	310 (77.3%) 91 (22.7%)	<0.001	120 (80.0%) 30 (20.0%)	116 (77.3%) 34 (22.7%)	0.573
PSS score Median (IQR)	19 (16 – 22)	19 (16 – 22)	0.463	20 (18 – 22)	20 (18 – 22)	0.521
Feeling unhappy with role Rarely (0–4 /y) Sometimes (5–12 /y) Often (13+/y)	496 (52.2%) 234 (24.6%) 221 (23.2%)	236 (51.2%) 141 (30.6%) 84 (18.2%)	0.020	79 (52.7%) 33 (22.0%) 38 (25.3%)	69 (46.0%) 41 (27.3%) 40 (26.7%)	0.451
Feeling sad or depressed Rarely (0–4 /y) Sometimes (5–12 /y) Often (13+/y)	482 (50.7%) 249 (26.2%) 219 (23.1%)	204 (44.8%) 159 (35.0%) 92 (20.2%)	0.003	64 (42.7%) 49 (32.7%) 37 (24.7%)	68 (45.3%) 48 (32.0%) 34 (22.7%)	0.879
Nr of Coping Mech. Median (IQR)	3 (1 – 4)	2 (1 – 4)	<0.001	1 (1 – 2)	1 (1 – 3)	0.029
Alcohol Use	418 (47.6%)	172 (42.0%)	0.060	77 (51.3%)	79 (52.7%)	0.817
Problems Parenting	413 (47.1%)	225 (58.0%)	<0.001	89 (59.3%)	96 (64.0%)	0.406
Sexual Issues	436 (49.7%)	178 (43.7%)	0.046	72 (48.0%)	74 (49.3%)	0.817
Problems Partner	431 (48.2%)	235 (56.6%)	0.005	103 (68.7%)	96 (64.0%)	0.392
Balancing Home and Work Life	525 (58.2%)	298 (71.6%)	<0.001	100 (66.7%)	100 (66.7%)	0.999
Non-Farmer Neighbors	423 (47.1%)	215 (54.4%)	0.015	83 (55.3%)	92 (61.3%)	0.292
Training Certificates	417 (48.5%)	241 (58.9%)	<0.001	93 (62.0%)	94 (62.7%)	0.905
Directly Exper Racism	435 (48.8%)	175 (41.8%)	0.018	77 (51.3%)	83 (55.3%)	0.488
Succession Stress	471 (52.4%)	246 (61.4%)	0.003	104 (69.3%)	99 (66.0%)	0.537
Microaggressions	444 (50.0%)	236 (56.9%)	0.021	107 (71.3%)	99 (66.0%)	0.319
Saving Retirement	497 (55.5%)	273 (64.5%)	0.002	109 (72.7%)	101 (67.3%)	0.314
Unexpected Financial Burdens	504 (56.0%)	250 (63.0%)	0.019	102 (68.0%)	105 (70.0%)	0.708

Bolded p-values are significant at a threshold of 5%. Categorical variables used Pearson’s Chi Square Test or Fischer’s Exact Test where applicable. Continuous variables used an independent t-test (reported with means) or Mann-Whitney (reported with medians). The same superscript letters show statistically different answer categories. IQR = interquartile range. Matched sample used McNemar Chi Square Tests and Wilcoxon tests. IQR = Interquartile Range; HS = High School; Yr = Year; Nr = Number; Exper = Experienced.

No competing interests reported.

Digging Deeper: Gender-Related Stressors and Coping Strategies Among Farmers (N=1,394)

Status:

Version 1

Abstract

Figures

Background

Gender Roles and Societal Expectations in Farming

Review of Studies on Stress Levels and Stressors in Farmers

Coping Mechanisms among Farmers

Methods

Perceived Stress Scale

Stressors

Negative Feelings

Coping Mechanisms

Findings

Conclusions

Limitations

Declarations

Author Contribution

References

Tables

Additional Declarations

Status:

Version 1