Experiences of Cultural Stress among US Citizen Migrants: A Mixed Method Approach

Abstract

On September 20, 2017, Hurricane Maria made landfall in Puerto Rico, devastating the archipelago and forcing thousands of Puerto Ricans to migrate to the United States (U.S.) mainland. Guided by a cultural stress theory framework, the present mixed method study examined how various cultural stressors impact participants’ daily interactions and mental health outcomes. A total of 319 adult Hurricane Maria survivors residing on the U.S. mainland participated in the study. A mixed method sequential explanatory design was used. First, we used latent profile analysis (LPA) and multinomial logistic regression to identify the varied cultural stress experiences that participants had. Then we used data from semi-structured interviews to better understand the experiences of participants classified into the different LPA profiles. Four profiles were identified: “Moderate Cultural Stress” (35%), “Overall Low” (29%), “High Cultural Stress” (26%), and “Low Language Stress” (10%). Multinomial regression indicated that members of the Moderate Cultural Stress, High Cultural Stress, and Low Language Stress profiles all reported significantly higher depressive and anxiety symptoms compared to members of the Overall Low profile. Qualitative data exemplified the daily experiences of participants placed into each profile, demonstrating that participants have rich and varied experiences that can contribute to their mental health symptoms. The present study documents the contributions of hurricane trauma and cultural stress vis-à-vis current mental health symptoms. Clinicians working with Hurricane Maria survivors should be cognizant of and inquire about migration-related cultural stressors including discrimination, feelings of being unwelcomed, and difficulty communicating in the destination community’s dominant language.

On September 20, 2017, Hurricane Maria made landfall in Puerto Rico with maximum sustained winds of 155 miles per hour. The storm devastated the archipelago and left many residents without power for months, and many Puerto Ricans were unable to access food and water. Maria was one of the worst hurricanes in United States (U.S.) history, and its devastating effects were similar to those caused by Hurricane Katrina on the U.S. Gulf Coast. Indeed, both storms displaced many thousands of people away from their homes, in many cases permanently (Hinojosa, 2018; Sastry & Gregory, 2014). Although both storms led to mass outmigration, Maria is unique among U.S. hurricanes in that it displaced residents to linguistically and culturally dissimilar areas where they faced manifold challenges in communicating with new neighbors, prospective employers, and others residing in their destination communities (Clark-Ginsberg et al., in press).

Indeed, although Puerto Ricans are U.S. citizens at birth, Puerto Rico maintains a distinct language and culture from that of the mainland (Vélez, 2017). Spanish is the lingua franca in Puerto Rico, and many Puerto Ricans (especially those outside the San Juan metropolitan area) do not speak English well (Puerto Rico Report, 2021). Culturally speaking, Puerto Rico is more similar to Latin American countries in terms of food, customs, and value systems than it is to the mainland U.S. (Acosta-Belén & Santiago, 2018). As a result, Puerto Ricans migrating to the U.S. mainland often must go through a process of acculturation – and may experience cultural stressors – similar to immigrants from Latin American countries (Vélez, 2017).

U.S. Citizen Crisis Migration: Hurricane Maria Survivors on the U.S. Mainland

Climate-related crises, such as major hurricanes, often displace large numbers of people from their homelands toward assumedly safer locations (Martin et al., 2013; Vos et al., 2021). The Hurricane Maria exodus from Puerto Rico is an example of internal migration, where individuals migrate within a country and are not subject to immigration policies. However, many Puerto Ricans report being treated as “second-class citizens” on the U.S. mainland (Capielo Rosario et al., 2023; Lloréns, 2018). Because Puerto Rico is not a state, and because Puerto Ricans do not enjoy the full benefits of U.S. citizenship (e.g., voting in presidential elections) while residing on the archipelago (Gonzalez et al., 2020), the ambiguity regarding Puerto Rico’s status creates considerable difficulty for Puerto Ricans who migrate to the U.S. mainland (Pantojas-García, 2005). Given this complex context of Puerto Rican migration, scholars have previously described Puerto Ricans as colonial migrants (Capielo Rosario & Dillon, 2020; Capielo Rosario et al., 2023; Gonzalez et al., 2020). Research continuously shows that Puerto Ricans are often treated as foreigners when they move to the U.S. mainland, and many Americans do not even know that Puerto Ricans are U.S. citizens (Dropp & Nyhan, 2017). Moreover, many people on the mainland view Puerto Ricans as migrating to the mainland primarily to receive government benefits (Dhuman Giron & Vargas, 2020). The receiving context for many Puerto Ricans may therefore be ambivalent at best, underscoring that Puerto Ricans migrating to the U.S. mainland represent a special case of U.S. citizen migration. That is, Puerto Ricans moving to Florida (or elsewhere on the mainland U.S.) likely face more cultural stressors compared to people moving from one U.S. state to another (such as Californians relocating to Texas or New Yorkers relocating to Florida)—but Puerto Ricans are not confronted with the same visa, documentation, or family separation challenges that people moving to the U.S. from foreign countries often are.

Just as Puerto Ricans are a unique group of U.S. citizens, Hurricane Maria survivors seemingly represent a unique group of Puerto Ricans. That is, although most Puerto Ricans who have migrated to the U.S. mainland have done so for economic or family reasons (Acosta-Belén & Santiago, 2018; Capielo Rosario et al., 2020), many Hurricane Maria survivors migrated to the mainland because their houses were destroyed, their jobs were no longer available, their healthcare needs could not be met without electrical power, and they did not see a future for themselves in Puerto Rico (García et al., under review). That is, Hurricane Maria survivors were likely more similar to international crisis migrants fleeing disasters than to prior waves of Puerto Ricans moving to the U.S. mainland in search of work or to rejoin family members (see Ertanir et al., in press; Vos et al., 2021, for reviews of the crisis migration literature). As a result, it is essential to examine the experiences of Hurricane Maria survivors on the U.S. mainland, even though similar work has already been conducted on other groups of Puerto Ricans (e.g., Capielo Rosario et al., 2019; Capielo Rosario & Dillon, 2020).

Cultural Stress Theory

Cultural stress theory was selected as the theoretical framework for this study given its emphasis on adverse migration-related experiences and their links with mental health outcomes among Latin American migrant populations. Cultural stress theory posits that adverse experiences such as discrimination, negative context of reception (NCR), and language stress can result in negative mental health repercussions among foreign-born populations (for an in-depth overview, see Salas-Wright et al., 2021; Salas-Wright & Schwartz, 2019).

Within a cultural stress framework, discrimination refers to specific events that occur in the lives of migrants such as slights or mistreatment on the basis of one’s ethnic and/or migrant identity (Araújo Dawson, 2009; Ayón et al., 2010). NCR refers to a broader, more general sense of unwelcomeness and lack of opportunity for migrants that may be experienced in the absence of specific discriminatory experiences (Portes & Rumbaut, 2014; Salas-Wright et al., 2021). Work by Schwartz et al. (2014) has demonstrated that discrimination and context of reception represent distinct constructs and that NCR, although perhaps more distal in terms of experience, is related to mental health outcomes above and beyond discriminatory experiences.

Cultural stress theorists have also noted that language stressors represent a key component of cultural stress within adult Puerto Rican and other migrant populations. Whereas children and adolescents often learn English fairly quickly in school and in the company of peers, Puerto Rican migrants who arrive as adults may experience greater difficulty mastering English (Capielo Rosario et al., 2015; Schwartz et al., 2006). Indeed, because individuals arriving as adults generally do not participate in formal schooling on the U.S. mainland, they are far less likely to be formally instructed in English. Moreover, there is a robust body of literature indicating that language learning and mastery is increasingly difficult after childhood (Hartshorne, Tenenbaum, & Pinker, 2018). Language stress—fear of speaking English in public, not being able to understand spoken English, and feeling uncomfortable around people who speak only English—may represent a unique component of cultural stress, especially for adult migrants (Cervantes & Bui, 2017; Lorenzo-Blanco et al., 2019; Schwartz et al., 2015).

In applying cultural stress theory to the present study, the theory assumes that various cultural stressors cooccur within the lives of Hurricane Maria survivors, and that these stressors compromise their mental health (Capielo Rosario & Dillon, 2020; Salas-Wright & Schwartz, 2019; Salas-Wright et al., 2021; Zhen-Duan et al., 2018). This assumption is based on prior work by Capielo Rosario et al. (2019). Capielo Rosario and colleagues (2019) identified three profiles of acculturation-acculturative stress among a predominantly U.S.-born Puerto Rican sample (N = 367) and found that each profile was associated with different levels of acculturative stress and depressive symptoms. However, these stressors have not been examined in a sample of Hurricane Maria survivors, who may experience them differently. Some may have experienced high or moderate levels of discrimination when they first arrived, whereas others may have felt welcomed. Some may have arrived with low levels of English and then improved, whereas others may not have improved. Still others may have arrived with high English proficiency. Therefore, identifying subgroups based on the different levels of cultural stress experienced among Hurricane Maria survivors is vital and can help us understand the complex ways in which cultural stress is experienced.

Although cultural stress theory has been examined within a variety of Latino immigrant populations (e.g., Schwartz et al., 2015, 2018), there is a dearth of research investigating cultural stress among Puerto Rican populations generally and Hurricane Maria survivors specifically (see Montero-Zamora et al., 2023; Schwartz, Maldonado-Molina et al., in press, for examples of such work). Critically, most extant cultural stress research, and nearly all of the work conducted with Hurricane Maria survivors, has been exclusively quantitative (e.g., Capielo Rosario & Dillon, 2020; Capielo Rosario et al., 2019). As a result, there is a need for mixed method work to examine the features of cultural stress theory in the lives of these hurricane survivors, and to bring a degree of descriptive or qualitative richness to the fore. In addition to “bringing the theory to life,” such work would allow us to ascertain the extent to which Hurricane Maria survivors’ statements are consistent with the specific stressors included within this theory. For example, consistent with prior theorizing and quantitative research, would participants mention being discriminated against, feeling shut out of opportunities, and/or being unable to communicate effectively with non-Spanish speakers? Would they note the ways in which these experiences may have compromised their family relationships or mental health? These remain open questions and warrant additional research.

Based on the tenets of cultural stress theory, any study testing this theory must evaluate two primary assumptions: (a) that the various cultural stressors cluster together—potentially in distinct configurations—within participants, and (b) that this amalgamation of cultural stressors is linked with impaired mental health (Schwartz, Zamboanga, et al., in press). Further, in crisis migrant samples (or other types of samples that have experienced considerable trauma prior to migration), it is also important to examine how traumatic events might interact with cultural stressors to predict mental health outcomes (Salas-Wright et al., 2021, in press). In the case of Hurricane Maria survivors on the U.S. mainland, such work might involve examining the contributions of hurricane trauma and cultural stress vis-à-vis current mental health symptoms (see Montero-Zamora et al., 2023, for an example of a quantitative study pursuing such an objective).

The Present Study

In our study, we built on the work of Capielo Rosario and colleagues (Capielo Rosario et al., 2019; Capielo Rosario et al., 2020; Capielo Rosario & Dillon 2020) by applying similar questions to a qualitatively different population: Hurricane Maria survivors. In the present study, we aimed to extend the existing literature by providing a mixed method test of cultural stress theory with a sample of Puerto Rican Hurricane Maria survivors on the U.S. mainland using a person-centered approach. Although variable-centered approaches are vital in providing information on how specific cultural stressors relate directly to key mental health outcomes, a person-centered approach allows researchers to model and capture the complex ways in which cultural stressors interrelate and manifest in the lives of individuals and, in turn, relate to mental health.

Our research was conducted in three steps. First, we aimed to identify the ways in which different cultural stress factors, namely, discrimination, negative context of reception, and language stress, would cluster together, representing the ways in which participants experienced different manifestations of cultural stress after migrating to the U.S. mainland. Although finite mixture modeling is, at its core, an exploratory approach and therefore does not necessarily require specific hypotheses, we expected to identify a subset of the sample reporting consistently elevated levels of cultural stress across all domains, as well as a universally low cultural stress class. We also suspected that some Maria survivors may experience cultural stress in some domains and not in others (e.g., low language stress but high negative context of reception).

Second, we sought to understand the ways in which cultural stressors were linked with decreased well-being and with symptoms of depression and anxiety. We hypothesized that classes reporting greater cultural stress in multiple domains would be more likely to experience increased symptoms of depression and anxiety compared to other groups.

Finally, using a sequential mixed method approach (Creswell & Clark, 2017), we endeavored to ascertain the extent to which cultural stressors would emerge from participants’ first-person quotes. No formal hypotheses were articulated, but the expectation was that qualitative research would both cohere with the quantitative findings and reveal new insights beyond what could be learned from closed-response survey items.

Method

Sample

The present sample included 319 adult Hurricane Maria survivors (71% women, mean age 38.7, SD = 12.1) residing on the U.S. mainland. Most participants arrived on the U.S. mainland in 2017 (60.2%) or 2018 (29.6%), and a small number (10.2%) relocated in 2019 or 2020. We used a referral system to recruit participants, where initial seed participants were referred through community partners and these seed participants, in turn, referred other individuals to the study in exchange for additional compensation (Heckathorn, 2002). The research team has worked with these community partners for several years on a variety of funded projects and regularly interacts in-person with both community leaders and lay members of the broader community.

Subsequent participants were also asked to refer additional eligible individuals. Each participant received $100 for completing the study measures, and an additional $30 per successful referral (i.e., referred participant who joined the study). Participants were allowed to refer up to five additional people. Most participants (75.3%) resided in Central Florida at the time of data collection, but other locations of residence included Texas (6.3%), New England (3.8%), Illinois (3.2%), and other U.S. locations (11.4%). All research procedures were approved by the Institutional Review Board at the University of Miami under protocol number IRB0053570.

Recruitment and Eligibility

Data were collected between August 2020 and October 2021, which coincided with the COVID-19 pandemic. As a result, all data were collected online with Qualtrics (survey data) and Zoom (qualitative interviews). Participants had the option of choosing alternative methods of participation (i.e., via phone), but no participants elected to do so. Research team members, along with collaborators working in the local community, distributed flyers and recruited eligible participants.

Inclusion criteria were: (a) being a Maria survivor, (b) having relocated to the U.S. mainland after Hurricane Maria, and (c) being 18 years of age or older. Individuals who agreed to participate were consented, enrolled, and emailed surveys to complete. Whenever participants requested help with troubleshooting technical issues, study staff members were available to assist. All participants completed the quantitative component of the study in Spanish using a survey that was translated and back translated by bilingual professionals with knowledge of Puerto Rican Spanish. The measures that were already available in Spanish were reviewed by bilingual professionals from Puerto Rico in order to ensure the measures would be congruent with Puerto Rican Spanish. The complete Spanish survey was refined in collaboration with Puerto Rican community advisory board members and via cognitive interviews with Puerto Rican Hurricane Maria survivors to ensure interpretability and acceptability.

Survey Measures

Cultural stress.

Cultural stress was measured using the following three indicators: negative context of reception, discrimination, and language stress. For each of these cultural stress constructs, we identified the four items that most accurately —in terms of factor loading and face validity— captured the core construct. The decision to use four items per construct was done a priori (i.e., before conducting the LPA with the current data). Although prior research has not made use of these reduced item lists, our approach allowed for greater parsimony and precision in the mixture modeling analyses and ensured that the same number of indicators was included for each construct examined.

Negative context of reception.

Negative context of reception (NCR) examines the degree to which migrants, in this case Puerto Ricans, perceive that their group is unwelcome or mistreated based on its cultural or national identity (Salas-Wright, Oh, et al., 2021; Schwartz et al., 2014). NCR was measured using four key items of the Perceived Negative Context of Reception Scale, which has been rigorously validated with Puerto Rican and other Latin American migrant populations (Salas-Wright, Oh, et al., 2021; Schwartz et al., 2014; Schwartz, Maldonado-Molina et al., in press). These items, characterized by high factor loading and the clearest conceptual clarity for Puerto Ricans, include “I don’t have the same chances in life here as immigrants from other countries,” “Puerto Ricans are not welcome here,” “It is hard for me to do well at work here because I am Puerto Rican,” and “Employers treat Puerto Ricans differently than they treat people from other countries.” Responses were measured using a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). In recent research, the coefficient alpha (reliability) for the original six-item measure ranged from .76 to .86 (Salas-Wright, Oh, et al., 2021), including .84 with Hurricane Maria survivors (Schwartz, Maldonado-Molina et al., in press). The alpha for the four items used in the present sample was .78.

Discrimination.

Discrimination assesses the frequency of being treated negatively by employers or not being accepted due to national origin or ethnicity. Discrimination was measured using items from Phinney et al. (1998) that loaded cleanly onto the direct injustice discrimination factor, rather than broader unwelcomeness. These items include: “How often do employers treat you unfairly or negatively because you are Puerto Rican?”, “How often do people you don’t know treat you unfairly or negatively because you are Puerto Rican?”, “How often do other people (such as police and shopkeepers) treat you unfairly or negatively?”, and “How often do you feel that others behave in an unfair or negative way toward Puerto Ricans?” Responses were measured using a 5-point Likert scale ranging from 1 (not at all) to 5 (almost every day). In prior research, the original seven-item measure has been shown to have a strong coefficient alpha (.87). The alpha for the four items used in the present sample was .93.

Language stress.

Language-related stress was measured using items from the Hispanic Stress Inventory (Cervantes, Padilla, & Salgado de Snyder, 1991) that had strong factor loading and most closely captured language stress and discrimination (rather than English language ability). These items include: “I feel uncomfortable being around people who only speak English,” “I feel pressure to learn English,” “I have been discriminated against because I have difficulty speaking English,” and “Since I don’t speak English well, people have treated me rudely or unfairly.” Participants responded to each item using a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). Recent research suggests strong reliability with the original, seven-item measure (coefficient alpha = .85), and the alpha for these four items in the present sample was .85.

Mental health.

Mental health was assessed in terms of self-reported depression and anxiety symptoms. For depressive symptoms (coefficient alpha = .89), we used the Boston Form of the Centers for Epidemiologic Studies Depression Scale (Grzywacz et al., 2006). This instrument taps into symptoms such as listlessness, anhedonia, and lack of interest in activities during the week prior to assessment. Consistent with other cross-cultural and non-English language studies (Yu et al., 2013), the two positively phrased items in the 10-item scale (“I was happy” and “I enjoyed life”) were removed due to very low (i.e., .05 and −.04, respectively) item-total correlation values. Participants responded to each item using a 4-point Likert scale ranging from 0 (rarely or none of the time) to 3 (all of the time).

To assess anxiety symptoms (coefficient alpha = .93), we used the 7-item Generalized Anxiety Disorder (GAD) scale (Spitzer et al., 2006). This measure assesses symptoms of anxiety, such as excessive worrying, tension, irritability, and difficulty sleeping, during the two weeks prior to assessment. Sample items include “feeling nervous, anxious or on edge” and “worrying too much about different things”. Participants responded to each item using a 4-point Likert scale ranging from 0 (not at all) to 3 (nearly every day). Then we created a mental health variable in which depression and anxiety item responses were aggregated as one total mental health outcome (coefficient alpha = .88).

Sociodemographic factors.

Sociodemographic factors assessed in the present study include age in years, gender (female = 1, male = 0), income (less than $10,000 = 1; $10,000$−24,999 = 2; $25,000-$34,999 = 3; $25,000-$34,999 = 4), educational level (less than high-school = 1, some college/associate’s degree = 2, college or more = 3), year of arrival to the U.S. (2017 = 1, 2018 = 2, 2019/2020 = 3), and English proficiency (Very Limited = 1, Limited = 2, Good = 3, Very Good = 4).

Quantitative Data Analysis

Data were analyzed using a three-step approach. First, we identified a sequence of latent profile models ranging from one to five profiles using Latent GOLD^® 5.1 software (Vermunt & Magidson, 2016). Across the domains of language stress, negative context of reception, and discrimination, four items per domain were specified as indicator variables, with demographic variables (i.e., age, gender, income, educational level, year of arrival on the U.S. mainland, and English proficiency) specified as indicator covariates. Then, we used five statistical criteria to identify the best fitting latent profile model: the Bayesian Information Criterion (BIC), Akaike Information Criterion (AIC), Consistent Akaike Information Criterion (CAIC), log likelihood (LL), and entropy.

All things being equal, lower BIC, AIC, and CAIC values and higher LL values reflected better model fit. Higher entropy values (approaching 1.00) indicate clear profile delineation (Celeux & Soromenho, 1996). When selecting the best fitting model, analysts should also consider parsimony (preferring solutions with fewer profiles unless additional profiles provide increased explanatory value) and the substantive interpretability of the solution. After settling on a profile solution, we conducted multinomial regression—with the nominal profile solution specified as the dependent variable—using Stata 16 SE to examine migration-related correlates (i.e., level of spoken English and year of arrival) of profile membership and mental health outcomes while controlling for demographic factors.

Qualitative Procedures

A subsample of 47 qualitative interviews were collected from participants who were parents or caregivers of youth ages 10 to 17 at the time of enrollment (110 participants in total). Of these 110 eligible participants, a total of 47 agreed to be interviewed. The qualitative interviews were limited to parents/caregivers given the overall study emphasis on the experiences of crisis migrant parents.

A semi-structured interview guide was developed by the research team and reviewed for appropriateness with the community advisory board. The semi-structured interview included questions about participants’ lives in Puerto Rico prior to Hurricane Maria, their experiences relocating to the mainland, their resettlement process on the mainland, and their experiences with mental health. Sample questions include: “Since relocating to [new destination], how has life been for you?”, “What have been the most important challenges for you and your family?” and “What would be helpful to you and your family as you continue to get settled?” (see supplemental material for interview protocol). All interviews were conducted in Spanish by bilingual interviewers and lasted between 45 and 60 minutes. Interviews were conducted between December 2020 and August 2021. Participants who took part in the interviews received an additional $100 to compensate them for their time.

Qualitative interviews were transcribed verbatim by research assistants (RAs) and imported into NVivo (NVivo, 2002). Two bilingual authors (MFG and BCR) used thematic analysis to identify generic codes around Hurricane Maria survivors’ experiences of relocation (Braun & Clarke, 2021). At the time of the data analysis, MFG was a doctoral student in counseling psychology and BCR was an undergraduate psychology student. MFG received extensive qualitative training throughout the doctoral coursework on thematic analysis. MFG and MPL provided extensive training to BCR on thematic analysis and using NVivo to facilitate the coding process. All authors involved in data analysis (i.e., MPL, MFG, and BCR) were born in Latin American countries or Puerto Rico and their native language is Spanish. One of the authors (BCR) identifies as Puerto Rican and had a rich understanding of the Puerto Rican context and the idioms used, which facilitated and enriched the data analytic process. The qualitative coding team met bi-weekly to discuss participant narratives and identify codes. MFG and BCR then developed a codebook and coded all the interviews. For each code, at least an 80% intercoder agreement was reached. When agreement rates were below 80%, the coders discussed their disagreements until they came to an agreement on their definitions and interpretations. Codes for which initial agreement was less than 80% were then recoded. Following the completion of the codebook and identification of codes, these codes were grouped into themes. A total of four themes were initially identified: early adjustment to the U.S., difficulties learning and speaking the language, perceived context of reception (feeling welcome or unwelcome) and experiencing prejudice and discrimination from Latinos and non-Latinos. MPL served as an auditor during the early stages of the coding process, and CSW served as an auditor during the later stages of the coding process. Having conducted the quantitative data analysis and some of the qualitative interviews, CSW was able to provide important insights regarding the identified themes. Specifically, the auditors assisted in looking for similarities and dissimilarities between the emergent themes, finding convergences and divergences with previous literature, and incorporating theory into theme development. A narrative of each theme explaining the shared reality of participants was developed, and pertinent quotes were identified to exemplify participants’ shared experiences.

Mixed Method Data Integration

For the present study we followed a mixed method sequential explanatory design where quantitative data were collected first, followed by qualitative data (Creswell & Clark, 2017). Through this sequential approach, the qualitative data helped to give meaning and “bring to life” the information obtained from the quantitative results (Creswell & Clark, 2017). Qualitative data were included in the study design to exemplify the experiences that participants had and how they differed across the various profiles (Table 3). The sequential approach for this study consisted of first identifying the cultural stress profiles and then selecting participants assigned to each profile. Qualitative data were coded using thematic analysis, which considered the different constellations of stressors that participants reported. Quotes that exemplified a shared experience among participants assigned to a given profile were selected and included as examples. The qualitative results were used to complement, elucidate, and triangulate the quantitative results (Ivankova, Creswell, & Stick, 2006).

Table 3.

Scale means and qualitative themes for each profile group

Scale and main qualitative theme	#1: Moderate Cultural Stress M (SD)	#2: Overall Low M (SD)	#3: High Cultural Stress M (SD)	#4: Low Language Stress M (SD)
Language Stress	2.60 (.46) Difficulties related to not speaking English	1.74 (.62) English language was not a barrier	3.22 (.52) Struggled speaking English	1.36 (.32) Basic knowledge of English
Negative Context of Reception	2.37 (.70) Feeling unwelcome	1.77 (.70) Feeling welcome by other Puerto Ricans and Latinos	3.15 (.57) Difficulty navigating different U.S. systems (e.g., school)	2.29 (.83) Friends and family provided instrumental help in the relocation process
Discrimination	2.02 (.38) Microaggressions as main type of discrimination	1.09 (.16) Smooth resettlement	3.33 (.58) Feeling of being underemployed and discriminated at work	2.40 (.52) Microaggressions due to Latino ethnicity

Results

Selecting a Profile Solution

The statistical criteria suggested that a four-profile solution provided the best fit to the data. Although the log likelihood, BIC, AIC, and CAIC values for the five-profile solution were slightly superior to that of the four-profile solution, these were minor differences. As shown in Table 1, the accelerated flattening of the fit statistics suggested that the addition of a fifth profile would not be parsimonious. The entropy values for 2–5 profile solutions were all at .90 or above. Beyond fit statistics, the five-profile solution was a variation of another class and did not provide additional, clear, or coherent conceptual interpretability when compared with the four-profile solution, which provided further evidence for excluding the fifth profile (see Figure 1).

Table 1.

Fit Indices for Latent Profiles

# Profile Solution	Log Likelihood | LL	Bayesian Information Criterion | BIC	Akaike’s Information Criterion | AIC	Consistent Akaike’s Information Criterion | CAIC	Entropy R2
1-Profile	−5162.39	10578.44	10412.78	10622.45	--
2-Profile	−4591.72	9558.18	9313.44	9623.18	.90
3-Profile	−4389.60	9275.00	8951.20	9361.00	.91
4-Profile	−4311.23	9239.33	8836.46	9346.33	.91
5-Profile	−4220.00	9177.94	8696.00	9305.94	.92

Figure 1.

Patterns of self-reported cultural stress among Hurricane Maria survivors in the United States, by latent profile

A total of 47 interviews were coded, with 18 interviews being coded as part of Profile 1, 10 interviews being coded as part of Profile 2, 17 interviews being coded as part of Profile 3, and two interviews being coded as part of Profile 4.

Profile 1: Moderate Cultural Stress

Profile 1 (n = 112) was comprised of 35% of the sample. In terms of the survey measures focused on cultural stress, participants in this profile reported moderate levels of language stress, perceived negative context of reception, and discrimination. Members of this profile were not meaningfully distinct from the full sample in terms of demographic or migration-related characteristics—most were under age 50 (80%), female (74%), and resided in households earning less than $50,000 per year (93%; see Table 2 for more detailed information).

Table 2.

Demographic and migration-related characteristics of latent profiles

	#1: Moderate Cultural Stress (n = 112; 35%)		#2: Overall Low (n = 92; 29%)		#3: High Cultural Stress (n = 82; 26%)		#4: Low Language Stress (n = 33; 10%)		Pearson Chi2 (df)
	n	%	n	%	n	%	n	%
DEMOGRAPHIC
Age
18–34	45	40.54	39	42.86	21	25.61	19	57.58	12.39 (6) p = .054
35–49	44	39.64	33	36.26	43	52.44	10	30.30
50 and older	22	19.82	19	20.88	18	21.95	4	12.12
Sex
Male	31	27.68	24	26.09	19	23.17	18	54.55	12.32 (3) p = .006
Female	81	72.32	68	73.91	63	76.83	15	45.45
Education
< High School	39	34.82	22	23.91	27	32.93	16	48.48	8.95 (6) p = .176
Some College	46	41.07	40	43.48	35	42.68	8	24.24
College Graduate	27	24.11	30	32.61	20	24.39	9	27.27
Income
< $15,000	49	43.75	37	40.22	43	52.44	11	33.33	6.33 (6) p = .007
$15,000-$49,999	55	49.11	45	48.91	34	41.46	17	51.52
$50,000 or more	8	7.14	10	10.87	5	6.10	5	15.15
MIGRATION-RELATED
Spoken English
Very Little	34	30.36	22	23.91	34	41.46	5	15.15	56.48 (9) p < .001
Little	44	39.29	24	26.09	29	35.37	4	12.12
Good	25	22.32	27	29.35	15	18.29	7	21.21
Very Good	9	8.04	19	20.65	4	4.88	17	51.52
Year of Arrival
2017	62	55.86	56	61.54	58	73.42	13	39.39	17.66 (6) p = .007
2018	41	36.94	22	24.18	15	18.99	15	45.45
2019/2020	8	7.21	13	14.29	6	7.59	5	15.15

Notes. % = Proportion of latent profile.

In qualitative interviews, many participants in this cluster mentioned that, when they first arrived on the U.S. mainland, they were challenged to adjust to their new environment, which felt very different from Puerto Rico. Furthermore, participants repeatedly described the perception of their new environment as unwelcoming, a trend noted especially among those who were first housed in hotel rooms by the Federal Emergency Management Agency (FEMA) upon their arrival on the mainland (see Clark-Ginsberg et al., in press). For example, a mother explained what she viewed as an inaccurate and biased perception toward Puerto Rican Hurricane Maria survivors temporarily resettled in hotels:

At a given moment they treated Puerto Ricans, well, there was a bit of discrimination. “Oh, you came from FEMA? Well, you are a person who depends on the government and came to take advantage being in those hotels, living the high life in the hotel,” when what we were really doing was suffering and going through a hard time. Because the hotel we were at was not a five-star Hyatt; it was a Standard Day, there was prostitution, drugs, strange things happening, strange people coming in and out, the police… It wasn’t a place we were used to living in. We were in the countryside and then they moved us to the city. It was a pretty drastic change, very different.

Y en cierto momento como que se les trató a los puertorriqueños, hubo como un...un discrimen. “¿Ah tu viniste de [FEMA]? Pues eres una persona dependiente que viniste a aprovecharte y estás en hoteles, en hoteles dándote gran vida en un hotel, cuando en realidad lo que hacíamos era sufriendo y pasando necesidades. Porque el hotel en el que estábamos no era un hotel de Hyatt de cinco estrellas; era un Standard Day, había prostitución, había droga, habían movimientos extraños, gente extraña que entraba y salía, policías… No era un sitio al que estábamos acostumbrados a vivir. Primero que estábamos en el campo y nos pasaron a la ciudad.

However, for many, the perceived context of reception changed when they moved (again) to areas where there was a degree of racial/ethnic diversity, especially areas with more Latinos. Most participants in this cluster eventually settled in areas with other Puerto Ricans or Latinos and expressed feeling welcomed by their community. For instance, a mother mentioned:

Honestly, it is like living in Puerto Rico. I live surrounded by Latinos. Yes, there are a lot of Puerto Ricans, many Mexicans.

Honestamente, es como vivir en Puerto Rico. Yo vivo rodeada de latinos. Sí, mucho puertorriqueño, mucho mexicano.

Notably, a few families who moved to predominantly non-Latino White areas in Central/Northern Florida felt welcomed by their receiving community. In general, participants in this cluster expressed not having faced overt discrimination, but rather experiences more in the realm of slights or microaggressions that not only stemmed from White counterparts, but also from Latinos and even other Puerto Ricans. For instance, participants expressed receiving “bad looks” just because they were Puerto Rican (both by established Puerto Ricans and other Latinos). Maria migrants in this cluster also reported experiences involving difficulties due to not speaking English. For example, a mother described her experience working with clients who she could not understand:

I tell you, there are Americans that maybe you try, they say something, a word you don’t understand and you say ‘Oh my God! What is he trying to tell me?’ But there are some that just say, ‘It’s okay, leave it, I’ll get it.’

Por eso te digo que hay su americano que quizás tu tratas, te dice una palabra y tu no la entiendes y tu como ‘¡ay Dios mío! ¿que será lo que me quiere decir?’ Pero hay algunos que dicen ‘está bien déjalo, yo lo busco.’

Profile 2: Overall Low

Profile 2 accounted for 29% of the sample (n = 92). On the survey measures, participants in this profile did not report language stress, negative context of reception, or discrimination experiences. Similar to Profile 1, members of this profile were not meaningfully distinct from the full sample in terms of demographic or migration-related characteristics—most were under age 50 (79%), female (74%), and resided in households earning less than $50,000 per year (89%).

Qualitative data for participants in this cluster indicated that most moved and settled close to other Puerto Rican or Latino families. This proximity to other Puerto Ricans and Latinos helped them to feel comfortable and welcome as they received help finding employment and support in adjusting to life on the United States mainland. For instance, when a parent was asked whether she liked her work colleagues, she said:

Yes, I like the feel, we are all Hispanic. There are people from Venezuela, from Puerto Rico. There is a Dominican and there are Americans as well, there is a bit of everything, diversity. Everyone gets along well with each other.

Sí, me gusta el ambiente, todos somos hispanos. Habemos de Venezuela, de Puerto Rico. Hay otro dominicano y así y americanos hay de todo, diversidad. Buen compañerismo.

Another parent mentioned that the resettlement was easy for her because she knew English well and could communicate effectively:

It was really easy here [when I arrived]. I speak English perfectly... It was really easy to talk to them [school administrators] and fill out all the paperwork… and I didn’t have to wait for interpreters or anything like that.

Aquí fue que te digo para mí bastante fue bastante fácil. El inglés lo sé perfecto… Y, además pues me hizo fácil no sé cómo hablar con ellos, ir o hacer todos los papeles… y que no tenían que esperar, por intérpretes y por de nada

Profile 3: High Cultural Stress

Participants in Profile 3 made up 26% of the sample (n = 82). On the quantitative measures, participants in this profile scored very high in all domains of cultural stress: language stress, negative context of reception, and perceived discrimination. In terms of demographic characteristics, this profile had the highest proportion of members earning less than $15,000 per year (52% compared to 44% in the full sample). Members of this profile were also the most likely to report speaking “very little” English (41% compared to 30% in the full sample), and to have relocated to the U.S. mainland in 2017 (73% compared to 60% of the full sample).

Qualitative findings indicated that participants in this cluster struggled with the English language, and most felt they did not speak it well. Participants in this cluster described feeling discriminated against at work and many felt they were underemployed, working in jobs that required fewer skills than what they had gone to school for in Puerto Rico. For example, a participant described being discriminated against at work for not speaking English:

When I’m working, I let people know that my ‘English is not good.’ And if I ask if they could repeat what they said, many times they’ve cut me off and just walked away. Other times they’ve insulted me like, ‘You’re here and you don’t know how to speak English.’

A la hora de yo prestar mi servicio, yo les dejo saber que mi ‘English is not good.’ Que si algo que si lo puede repetir y ahí, muchas veces me han dejado con la palabra y siguen caminando, otras veces me insultan, como que ‘Tú estás aquí y no sabes inglés.’

Parents also mentioned experiencing challenges in navigating mainland U.S. systems, particularly enrolling their children in school, because they did not speak or understand English very well. For instance, one of the mothers explained that she felt discriminated against because her daughter’s education documents were not accepted by the school, even though Florida’s governor had created a program to facilitate the transition for Puerto Rican Hurricane Maria survivors:

I was going to the school, talking to the teachers, and they never put her in the gifted track. When she was about to graduate fifth grade, they called for a meeting and OH! they didn’t accept the testing done in Puerto Rico, and they were supposed to. That is something that I did notice, I would say it was the first time I would venture to say with certainty that I felt a bit of racism.

Yo estuve yendo a la escuela, hablando con las maestras, nunca la pusieron en el programa de gifted. Ya cuando se va a graduar de 5to grado, me llaman una reunión y... ¡a! no aceptaron las pruebas de Puerto Rico que se supone que sí. Eso fue algo que yo, ahí yo note si un poco de, yo diría que esa es la primera vez que yo me atrevo a decir con certeza que yo note un poco de racismo.

The difficulty that parents experienced was also mirrored by their children, who were struggling to adjust to the U.S. mainland culture and found it difficult to get along with their classmates. A participant described her child as having the following experience:

It was frustrating. I cried so much, young man, I cried so much. Most teachers and assistants were American and I don’t know how to speak English. I couldn’t defend my children, I couldn’t advocate for them because they [administrators] did not speak Spanish, and I don’t speak English. I did not know how to explain the situation to them, so for that reason I decided to take them out of their school and put them in home school because the situation was affecting them and it was affecting me.

Frustrante. Yo lloraba tanto, caballero, yo lloraba tanto Entonces, la mayoría de los maestros y los ayudantes eran americanos y yo no sé inglés. Yo no podía defender a mis hijos yo no podía abogar por mis hijos. Porque ellos no sabían en español, yo no sé inglés, yo no sabía cómo defender la situación, yo no sabía cómo explicarles a ellos, por tal razón yo dije los cambio de escuela, los pongo home schooling porque esta situación, no puede ser, porque se afectan ellos y me afecto yo.

Profile 4: Low Language Stress

Profile 4 encompassed 10% of the sample (n = 33). Via survey measures, participants in this profile reported moderate levels of negative context of reception and discrimination, but low levels of language stress. Members of this profile were also much more likely to report “very good” spoken English (52% compared to 15% in the full sample). Additionally, members of this profile were more likely to be male (55% compared to 29% in the full sample), less likely to report very low income (less than $15,000 per year; 33% compared to 44% in the full sample), and less likely to have relocated in 2017 (39% compared to 60% in the full sample).

Qualitative information indicated that participants in this cluster had well-established lives in Puerto Rico and were doing well before the hurricane hit. However, the storm and its aftermath significantly disrupted their lives and prompted them to move to the mainland. Many reported having family or friends already living on the U.S. mainland, noting that these support persons were key in their adjustment process. These support persons helped them settle into their new homes and provided help in finding employment and getting established. Participants in this profile discussed having a basic knowledge of English before coming to the U.S., but underscored that family and friends helped them to improve their language skills and become more comfortable speaking. Participants in this profile stated that language was no longer an issue for them, noting that they now feel comfortable speaking English. For example, a participant mentioned the following:

I knew some [English], but there were words I didn’t understand, I mean, I knew the basics… there were words that I didn’t understand but little by little I was learning, I especially had to learn at work… in my university studies [on the U.S. mainland] everything was in English, but my teachers helped me.

Yo sabía hablar un poco pero había palabras que no entendía, o sea, yo se lo básico… hay palabras que no entendía pero poco a poco fui aprendiendo y más por el trabajo tenía que aprender… en la universidad todo era en ingles pero los maestros me ayudaban.

Although participants in this profile stated that they did not experience overt discrimination, one participant mentioned an experience of racial profiling. In this instance, the individual (female) encountered a police officer who did not believe she owned her car or had a legitimate driver’s license because she was Latina. She noted:

There was one situation that we did feel some prejudice, that we realized after it happened and we were so angry… My husband was driving and a sheriff stopped us… he asked for his license and my husband gave him his Florida license… And so he was looking at us… and he tells us where did you get these licenses?... And like that a series of intimidating questions

En una situación si nos sentimos con prejuicio, que nos realizamos después de que paso y nos dio mucho coraje… Mi esposo venia guiando y lo detuvo un Sherif… le pidió la licencia y todo y pues mi esposo le dio la licencia de Florida… Entonces se nos queda mirando… y nos dice y ustedes de donde sacaron esas licencias?... Y así una serie de preguntas intimidantes.

This participant noted that her husband was darker skinned, which she felt led the police officer to ask prejudicial and unnecessary questions for a simple traffic stop.

Mental Health vis-à-vis Cultural Stress Types

In addition to the latent profile modeling and qualitative coding, we examined the relationships between latent subgroup membership and mental health outcomes while simultaneously examining covariates, indicators, and mental health outcomes by means of the the Bolck-Croon-Gangenaars (BCH) bias-adjusted three step approach (see Bakk & Kuha, 2021 and Nylund-Gibson et al., 2019 for greater detail).

Using effect coding, we found that the levels of depression and anxiety symptoms among the Moderate Cultural Stress (depression: B = −0.105, SE = 0.058, p = .070; anxiety: B = −0.091, SE = 0.072, p = .205) and Low Language Stress (depression: B = −0.007, SE = 0.101, p = .941; anxiety: B = 0.015, SE = 0.120, p = .898) classes did not differ from the grand mean. In contrast, levels of depression and anxiety symptoms were significantly lower among the Overall Low class (depression: B = −0.304, SE = 0.060, p < .001; anxiety: B = −0.404, SE = 0.069, p < .001) and elevated among the High Cultural Stress class (depression: B = 0.417, SE = 0.072, p < .001; anxiety: B = 0.480, SE = 0.089, p < .001).

Discussion

We conducted the present study to ascertain the degree of heterogeneity in cultural stress experiences among Puerto Rican Hurricane Maria survivors who migrated to the U.S. mainland (primarily Central Florida) following the storm. Consistent with a cultural stress theory perspective, we assessed discrimination, negative context of reception, and language stress, reflecting a range of adverse cultural experiences that recent migrants were likely to undergo following their arrival on the U.S. mainland. Although Puerto Rico is a U.S. territory, many Puerto Ricans—especially those residing outside the San Juan metro area—do not speak English fluently. As a result, language barriers, including difficulties speaking and understanding English and discomfort interacting with monolingual English speakers, are likely to cooccur along with overt discrimination and a sense of being unwelcome and unwanted. U.S. citizenship is unlikely to provide substantive protection against these types of challenges, as many of our respondents indicated in their qualitative excerpts.

The qualitative excerpts we included serve to confirm and expand upon the profile solution extracted in the quantitative analyses. For example, individuals in the High Cultural Stress profile reported conflictual interactions with monolingual English speakers and perceived that they were not wanted on the U.S. mainland. Individuals in the Moderate Cultural Stress profile experienced some initial negative context of reception, but this poor reception appeared to ease once migrants had settled into ethnically dense areas with other Latin American people. Respondents in the Overall Low profile were proficient in English and familiar with mainland U.S. culture, and they experienced few or no cultural difficulties upon arriving on the mainland. These individuals were pleased with the amount of cultural diversity they encountered in their new neighborhoods and generally enjoyed their work. Individuals in the Low Language Stress profile knew some English prior to moving to the U.S. mainland, so their adjustment process was less stressful compared to individuals in the Moderate and High Cultural Stress profiles. However, members of the Low Language Stress profile still experienced other cultural stressors, such as having police officers doubt that someone who looked like them could afford a nice car or have a legitimate driver’s license.

Also consistent with cultural stress theory, individuals in the Overall Low and High Cultural Stress profiles reported significantly distinct symptoms of depression and anxiety, with the Overall Low class reporting lower symptom scores and the High Cultural Stress class reporting elevated symptom scores. A key component of cultural stress theory is that the amalgamation of discrimination, negative context of reception, and language stress is associated with symptoms of psychological distress (Salas-Wright & Schwartz, 2019). These findings, along with qualitative excerpts suggesting that participants in the High Cultural Stress profiles were feeling anxious and depressed, appear to support this fundamental tenet of cultural stress theory.

Within the qualitative component of our study, participants’ first-person perspectives were generally consistent with the profile descriptions—that is, people classified in the Moderate and High Cultural Stress profiles generally provided statements suggesting more isolation and exclusion from English speakers on the U.S. mainland. On the other hand, individuals in the Overall Low profile and the Low Language Stress profile provided more positive statements about their lives on the mainland, suggesting that they felt more welcomed and included. Participants’ statements drew upon all three cultural stress indicators that we measured quantitatively—discrimination (e.g., being told to go back to Puerto Rico), negative context of reception (e.g., feeling unwelcome and being shut out of opportunities), and language stress (e.g., difficulty making oneself understood when communicating with non-Spanish speakers).

The present study is part of a series of efforts to extend cultural stress theory to U.S. citizen crisis migration (see also Montero-Zamora et al., 2023; Salas-Wright et al., 2023; Schwartz, Maldonado-Molina, et al., in press). Although internal migration—individuals migrating within a nation—is qualitatively different from international migration (Bell et al., 2015), Puerto Rican migration to the U.S. mainland lies somewhere between these two types of migration. Because Puerto Rico maintains a distinct language and culture from that of the U.S. mainland, Puerto Ricans migrating to the mainland will likely experience a more challenging adjustment process and more cultural stress than people migrating from one state to another (see Acosta-Belén & Santiago, 2018, for a review). However, Puerto Rican migrants to the U.S. mainland do not have to go through formal immigration channels and therefore enjoy the full benefits of citizenship as soon as they arrive. Nonetheless, as our findings suggest, Puerto Ricans moving to the mainland are not immune to cultural stress. Previous research with Puerto Rican migrants living in the U.S. has found similar results (Capielo Rosario & Dillon, 2020; Capielo Rosario et al., 2019; Capielo Rosario et al., 2022). In a qualitative study, Capielo Rosario et al. (2023) interviewed 24 recently-arrived Puerto Rican migrants and found that after migrating to the U.S. mainland, they experienced ethnic discrimination in their communities and at work. The study also found that participants often described experiencing discrimination from more established Puerto Ricans, which recently-arrived Puerto Ricans found unexpected and hurtful (Capielo Rosario et al., 2023). A quantitative study investigating acculturation profiles and depressive symptomology among 367 Puerto Ricans found that participants in the partial marginalization group—those who did not identify with the Puerto Rican culture, nor with the U.S. culture—were more likely to have elevated symptoms of depression, corroborating the results found in the present study (Capielo Rosario & Dillon, 2020). Our study further extends this literature by identifying specific patterns among Puerto Ricans who migrated in the aftermath of Hurricane Maria.

Puerto Rican Hurricane Maria migrants are likely qualitatively different from other Puerto Rican migrants, most of whom have migrated for economic, educational, or family-reunification reasons (see Capielo Rosario et al., 2023; Gonzalez et al., 2020, for more information). The present results might therefore reflect some combination of pre-migration hurricane trauma and post-migration cultural stress (see Montero-Zamora et al., 2023, for an example of research examining such combinations and their links with mental health). Our data, however, do not speak to how the migration-related cultural stress experiences of Maria migrants may impact individuals’ relationships with friends, family, and their communities in Puerto Rico, but future research should examine this topic.

Implications

The present findings highlight important implications and considerations when working with Puerto Rican Hurricane Maria survivors. As discussed earlier, experiences of cultural stress, compounded with post-traumatic stress related to the hurricane, exert negative effects on mental health among immigrant populations, including Puerto Rican Hurricane Maria survivors. Consequently, clinicians should have knowledge and understand the ramifications of the traumatic experiences that are specific to Hurricane Maria survivors. This entails providing Hurricane Maria survivors with trauma-informed care, in addition to cognitive behavioral therapy or other evidence-based intervention approaches that can ameliorate symptoms of depression and anxiety (Pineros-Leano et al., 2017). Furthermore, when working with Hurricane Maria survivors, clinicians should also inquire about experiences related to cultural stress among this population, such as experiences of discrimination, feeling unwelcomed, and difficulty communicating and being understood. Clinicians should be aware of experiences related to discrimination and negative context of reception even among those who experience little to no language stress. Further, clinicians should be sure to differentiate reports of cultural stress that are relatively moderate from those that are severe and tailor their intervention accordingly.

It is also critical for mental health professionals, as well as community leaders, to move beyond the one-to-one clinical relationship and address meso- and macro-level factors that contribute to cultural stress. It would be beneficial for clinicians to develop innovative and wholistic treatment plans, including services that could help address experiences of cultural stress, when working with migrant populations. For instance, including language supports outside of therapy can be an important part of the therapeutic process. Additionally, our findings suggest that careful attention and consideration is needed vis-à-vis meso- and macro-level factors that contribute to experiences of cultural stress. Considering the rhetoric around immigration in the U.S. and attitudes towards Latin American and Latino populations could also be an important part of the therapeutic process (see Montero-Zamora et al., 2023). Specifically, activities might be developed to help clients avoid internalizing harmful rhetoric or depictions of migrants. Lastly, paying careful attention to individual experiences of discrimination and serving as public advocates for Latin American and Latino clients can also serve as important aspects of the therapeutic relationship and process.

Limitations

The present results should be interpreted in light of several limitations. First, our sample predominantly included females, which may particularly impact our findings around depressive symptoms since previous research has shown that existing measures are more likely to capture symptoms of depression that are commonly identified by females. Another limitation is that all data were gathered via self-reports, with no attempts to collect information from other informants such as partners, friends, coworkers, or children. Although we collected both quantitative and qualitative data—which may help to correct for shared method variance issues in which correlations among reports of various constructs from the same person may be inflated (Podsakoff et al., 2003)—it may nonetheless be helpful to collect informant reports where possible. Also, Profile 4: Low Language Stress, which constituted 10% of the sample, included qualitative information from only two participants, which may be indicative of the low prevalence of this profile and should be considered when interpreting the findings. Furthermore, acculturative stress was not included as a measure in the present study, but future cultural stress research may benefit from careful consideration as to how its main constructs may interface with acculturative stress. Additionally, the measures that were only available in English were translated and back translated, which is only one step in the cultural adaptation of a measure (Byrne, 2015, 2016). Finally, although our survey data include a variety of adults ages 18 and older (parents and non-parents), our qualitative data was limited to parents and caregivers of adolescents, which may have impacted our qualitative findings.

Conclusion

In conclusion, despite these limitations, the present study offers one of the first mixed method examinations of cultural stress theory within the context of U.S. citizen crisis migration. The present results strengthen cultural stress theory by suggesting that participants’ own words and statements exemplify the component stressors described by the theory (Salas-Wright & Schwartz, 2019; Schwartz, Zamboanga, et al., in press). Additionally, our results carry important implications for counseling U.S. citizen migrants. We hope that the present results will inspire additional work in this direction.

Public significance statement.

This article expands cultural stress theory by documenting the contributions of hurricane trauma and cultural stress vis-à-vis current mental health symptoms using a mixed method approach. Results of this study highlight the need for clinicians who work with Hurricane Maria survivors to inquire about migration-related stressors that can exacerbate mental health symptoms.