Overcoming Obstacles by Enacting Resilience: How Queer Adolescents Respond to Being Estranged From Their Parents

The Complications of Adolescence

Adolescence (early adolescence 10–13; middle adolescence 14–17; late adolescence 18–19; WHO, 2021) is a tumultuous time in any person’s life. Eccles and Gootman (2002) argue that adolescents must negotiate, (a) shifting relationships with parents, (b) new social and sexual roles, (c) intimate partnerships, (d) social and personal identity formation, (e) planning for the future, and (f) acquiring the necessary skills to successfully transition into adulthood. Exacerbating the demands of these tasks, adolescents must also contend with physical and biological changes such as dramatic shifts to the shapes of their bodies, increased hormones, and alterations to their brain architecture (Zarrett & Eccles, 2006). Indeed, adolescent brains are in the process of frontal lobe development, which impacts self-control, judgment, emotion regulation, organization, planning, and ultimately cognitive capacities more broadly (Begley, 2000). Navigating these cognitive and emotional demands are difficult but often supported by positive parent-child relationships (Morris et al., 2017). As such, positive parent-child relationships become increasingly important to adolescents’ self-esteem and their overall ability to thrive (Roberts & Bengton, 1996).

Difficulties Associated With Estrangement and Ideological Differences

Despite the positive role many parents enact for their adolescent children, Zarrett and Eccles (2006) argue that families of origin can also serve as a serious risk for those who lack a supportive family environment. Research suggests that adolescents estranged from their parents might have had many adverse experiences (Scharp et al., 2015). In these situations, adolescents with negative family relationships can find themselves without social capital in the forms of financial, emotional, and achievement-related support (Zarrett & Eccles, 2006). Estranged children also report experiencing extreme grief, trauma, and stigma given that they live in a culture that believes (a) parents and children are forever connected through biology, (b) family webs create unending obligations and interactions, and (c) shared history is irreplaceable (Scharp & Thomas, 2016). Thus, it might come as little surprise that estranged children generally report chronic uncertainty (Scharp & McLaren, 2018) and a lack of support (Scharp, 2016).

Although the majority of research on parent-child estrangement pertains to adult children who initiated distance from their parents; ideological differences can also serve as a catalyst for estrangement (see Scharp & Dorrance Hall, 2019). One of the most difficult challenges an adolescent might face is coming out to their parents as queer (Adams, 2011). Conron (2020) estimates almost 2 million people between the ages of 13 and 17 are LGBTQ+ and Grafsky (2018) found that adolescents, on average, come out to their parents at the age of 16. Researchers suggests that even though some family members are supportive, LGBTQ+ individuals often experience hostility, rejection, estrangement, and even violence from their parents (Adams, 2011). EQAs are often unhoused (Ecker et al., 2019) and at increased risk of mental health disorders and suicidality (National Alliance on Mental Illness, 2017). To complicate an already complicated experience, a study about unhoused queer youth suggests 70% of adolescents are non-white (Siconolfi et al., 2019), suggesting that on top of the estrangement and the difficulties of coming out, EQAs might also have to contend with the racism, inequities, and disparities associated with being non-white in the US. This is salient for EQAs, considering researchers agree that the more risk factors adolescents experience or are exposed to (e.g., poverty, housing/food insecurity, abuse, etc.) the greater the likelihood of deleterious outcomes (Coleman & Hagell, 2007). In light of the intersecting complications of being an EQA, we pose:

Data Collection and Participant Demographics

After receiving Institutional Review Board approval, we recruited participants who met the following criteria: (a) were between the ages of 12 ¹ and 19 years old (i.e., adolescents), (b) were comfortable reading and speaking in English, and (c) were distanced from their parents due to their sexual and/or gender identity. To recruit participants, we announced the study at various LGBTQ+ youth centers and youth homeless shelters in the United States and on social media. We built a relationship with centers and shelters by explaining the goals of our research, describing our study, and asking them to post our participant recruitment flier. Overall, the recruitment process, due the pandemic, took a total of 2 years. After emailing to indicate their interest in the study, we then sent participants a demographic questionnaire asking them to report their age, gender, race, sexual orientation, and length of estrangement, before scheduling a narrative and semi-structured interview. This process yielded 40 unique participants whose interviews ranged from 19 to 82 minutes (M = 42.43, SD = 17.44), totaling 516 single spaced pages of data.

Our 40 participants were between the ages of 13 and 19 (M = 17.18, SD = 1.58), which meant the majority of our participants were in the upper range of middle adolescence (WHO, 2021), although we had participants in early and late adolescence as well. Fourteen (n = 14, 35%) were male, 14 (n = 14, 35%) were female, eight (n = 8, 20%) were trans male, and four (n = 4, 10%) were agender or gender fluid. Most participants were Black (n = 19, 47.9%), White (n = 7, 17.5%), multiracial (n = 6, 15%), Hispanic/Latinx (n = 5, 12.5%), Hawaiian/Pacific Islander (n = 2, 5%), and Asian (n = 1, 2.5%). The sexual orientation for participants was gay (n = 15, 37.5%), lesbian (n = 8, 20%), bisexual (n = 7, 17.5%), pansexual (n = 3, 7.5%), heterosexual (n = 2, 5%), ² asexual (n = 2, 5%), other (n = 2, 5%), and queer (n = 1, 2.5%). Participants identified as being estranged from their parents, on and off again, for an average of 2.85 years (SD = 1.99, R: 0.8–6) and continuously for an average of 4.20 years (SD = 2.99, R: 1–12). One participant explained that their estrangement was gradual, and they do not remember when it truly began.

Data Analysis and Data Verification

For this analysis, we applied thematic co-occurrence analysis (TCA; Scharp, 2021). TCA is a method that allows researchers to explore the qualitative relationship between and across developed themes. The first step of TCA is to conduct a thematic analysis. We used Braun and Clarke’s (2021) phases of reflexive thematic coding including (a) data familiarization and writing familiarization notes, (b) systematic data coding, (c) generating initial themes from coded and collated data, (d) developing and reviewing themes, and (e) refining, defining, and naming themes, before (f) writing the manuscript. ³ For RQ2, we added an additional layer of analysis to compare the themes to Buzzanell’s (2018) five CR processes. Once themes emerged, we applied them to Owen’s (1984) standards of recurrence (i.e., how often they emerged), repetition (i.e., whether they were described with same or similar words), and forcefulness (i.e., the extent to which they were detailed and emphasized).

Scharp (2021) explains that the second step of TCA requires that the researchers create a co-occurrence matrix (see Table 1). A co-occurrence matrix illustrates whether (i.e., how recurrent, and repetitious) and to what extent (i.e., how forceful) a theme manifests in each of the participants’ accounts. We went row by row marking an X every time a theme emerged in a participant’s talk. We added a + after each X if the utterance was forceful.

OPEN IN VIEWER

Table 1.

Co-Occurrence Matrix of Participants.

Participant	Triggers				Communicative resiliency processes
	Negative reactions	Educating others	Ideological differences	Access to resources	Comm networks	Crafting normalcy	Alternative logics	Productive action
1	U			C1	U+	B1	B2
2	U	X	X+		U	B1	B2
3	U	X			U		X	C2+
4	U+		X+		U		X	C2+
5	U+	X+	A1	A2/C1+	U+	B1	B2+
6			A1	A2/C1	U
7			X		U+	X
8	U	X	X		U+	B1	B2+
9	U		A1	A2/C1	U		X
10	U+			X	U+	B1	B2	X
11			X		U	B1+	B2+	C2
12			X+		U	B1	B2
13	U	X	X		U+	B1+	B2	C2+
14	U				U	X+
15	U		A1	A2/C1	U	B1+	B2+
16	U+		X		U	B1	B2+
17	U+				U
18	U		A1	A2/C1+	U
19	U		X+		U	B1	B2
20	U			X	U+			X
21				C2	U
22	U				U
23	U+			C2	U		X+
24	U				U	B1	B2+
25	U+	X	A1	A2/C1	U	B1+	B2+
26	U	X		C1+	U		X+
27	U+	X		C1+	U		X
28	U	X+	A1+	A2/C1	U	B1	B2
29	U				U
30	U		A1	A2/C1	U	B1	B2+
31	U		A1	A2/C1	U
32	U		X		U		X	C2
33	U	X		C1+	U		X
34			A1+	A2/C1+	U+
35	U		X		U		X+
36			A1+	A2/C1	U	X
37			X		U+	X
38	U		X		U
39	U		A1	A2	U		X+	X
40			X		U

Note. X denotes presence of the theme and X+ denotes the expression of the theme was forceful. Letters (e.g., A and B) and colors (e.g., light gray and dark gray) denote a relationship (i.e., co-occurrence) between themes. Letter C denotes an inverse relationship; U denotes the theme was ubiquitous. Numbers (e.g., 1, 2) denote how many themes belong to each relationship.

Finally, we examined the matrix for instances of co-occurrence within and across research questions (Scharp, 2021). Specifically, four patterns emerged which we labeled U, A1/A2, B1/B2, and an inverse relationship between C1/C2 where the letter represents a relationship between themes (i.e., a co-occurrence) and the number represents how many themes characterize that relationship. Once we identified the relationships, we examined the quality of those relationships applying Scharp’s three TCA continua: (a) sporadic/pervasive, (b) unilateral/bilateral, and (c) balanced/unbalanced. The sporadic/pervasive continuum reflects whether the co-occurring relationship (A1/A2) pervades the data corpus. The unilateral/bilateral continuum speaks to whether the presence of one theme implies the presence of another (if A1 then A2) or whether the presence of one theme implies the other and vice versa (if A1 then A2 and if A2 then A1). Finally, balance speaks to the extent to which the theme is recurrent, repetitious, and forceful compared to the others in that relationship. For clarity, whereas balance speaks to the prevalence/force of each theme, pervasiveness speaks to the prevalence/force of the co-occurrence. Of note, two new relationships emerged based on a close examination of the TCA matrix. In our study, U stands for ubiquity such as that all EQAs discussed negative reactions to their queer identity (when that identity had been disclosed) and all EQAs discussed maintaining and using networks. In this regard, all themes co-occurred with those that were ubiquitous. Our findings also advance TCA in another meaningful way. Specifically, relationship C1/C2 illustrates an inverse relationship such that when EQAs discussed the lack of material resources, they also did not identify productive action as a resilience process. To ascertain this inverse relationship, we carefully examined the matrix and when we noticed this seemingly inverse pattern, and returned to our data to determine if the interview accounts supported this observation.

To verify our analysis, we engaged in five interrelated procedures: (a) referential adequacy, (b) peer debriefing, (c) negative case analysis, (d) audit trail, and (e) exemplar identification (Kidder, 1981; Lincoln & Guba, 1985). First, we began by splitting the data in half to achieve referential adequacy. Because all three coders reached saturation in the first half of the data (see Corbin & Strauss, 2008; Table 1), we then analyzed the second half of the data. No new themes emerged when compared to the second half of the data. Throughout the process, the first three authors engaged in peer debriefing during two meetings where they discussed their findings, resolved differences, and reached a consensus. For example, originally, the authors identified both religious beliefs and (inter)cultural values as separate triggers but, after discussion, decided they were part of a larger theme, ideological differences. The authors also engaged in peer debriefing throughout the TCA analysis. They then determined the labels that are presented in the findings. Fourth, we refined our analysis until we could account for 100% of the data, meeting the standard for negative case analysis (Kidder, 1981). Finally, the authors kept detailed notes for all the steps of the TCA (i.e., an audit trail), which informed both the analysis and the selection of evocative exemplars (Lincoln & Guba, 1985).

RQ1: Resilience Triggers

Four triggers served as catalysts for resilient responses: (a) negative parental reactions, (b) educating others, (c) ideological differences, and (d) a lack of access to material resources.

RQ2: CTR Processes

Our second RQ asked how EQAs engaged in the CR processes in response to their triggers. We identified four of Buzzanell’s (2010, 2018) CR processes. Specifically, EQAs discussed: (a) maintaining and using communication networks, (b) crafting normalcy, (c) foregrounding productive action while backgrounding negative emotions, and (d) putting alternative logics to work. In concert with existing CTR research (see Scharp et al., 2021, 2022), a tension emerged to characterize the processes. In our study, EQAs varied in their embeddedness in the LGBTQ+ community. Embeddedness reflected having access to a variety of resources such as support groups and even specific terminology to discuss their experience. Based on this tension, EQAs reported being more and less able to engage in the CR processes.

RQ3: Co-occurrences Between Themes

Our third research question asked about which themes co-occurred within and between research questions (see Table 1). In addition to the two ubiquitous themes (e.g., negative parental reactions and maintaining and using networks) which co-occurred with everything, three nuanced relationships emerged from our analysis, (a) a unilateral unbalanced sporadic relationship between ideological differences and a lack of access to resources, (b) a unilateral balanced sporadic relationship between crafting normalcy and putting alternative logics to work, and (c) an inverse relationship between a lack of access to resources and foregrounding productive action while backgrounding negative feelings.

Theoretical and Methodological Implications

Theoretically, findings from this study advance the CTR in a variety of ways. Specifically, this is one of the first studies to illustrate the process of CR for an adolescent population (see Craig et al., 2020 for exception). Indeed, two notable differences emerged compared to existing research based on adult populations. First, speaking positive futures into being emerged as a new facet of putting alternative logics to work. This subtheme attends to the past-future orientation Buzzanell (2019) discusses, animating the anticipatory-reactive dialectic. Put differently, speaking positive futures into being problematizes previous conceptions of anticipatory resilience that situate it as something that inherently occurs before trigger events. Thus, unlike previous conceptualizations of alternative logics that focus on retrospective sense-making and humor (see Wilson et al., 2021), this sub-process, constructed in the moment, focused on hope and the ways situations might improve over time or with access to more resources. We think this might have been especially salient for adolescents who recognized the ways their lives could change once they were adults and had more financial freedom. This finding also corresponds to theorizing about anticipatory resilience. Specifically, Betts et al. (2022) argues that anticipatory resilience is a narrative logic whereby people attempt to make sense of their disruption and in doing so utter incoherent fragments that reveal future-oriented logics. Yet in our study, speaking positive futures into being was not a sensemaking endeavor that reflected future logics or a way to accumulate resources in anticipation of disruption, but rather a communicative enactment that constituted a positive future reality. Thus, the inclusion of speaking positive futures into being is a logically consistent theoretical extension that strengthens the usefulness of CTR as a mid-range heuristic.

The second major difference between adult and adolescent accounts emerged in the (non)patterned co-occurrences between the CR triggers and processes. To date, researchers have illustrated strong patterns across triggers and processes. For example, in a sample of first-generation students, one CR process (i.e., crafting normalcy) co-occurred with multiple triggers (Scharp et al., 2022) and in the context of migration to US, multiple CR processes co-occurred to one specific trigger (Scharp et al., 2021). Although relationships among the triggers and processes independently manifested in patterned ways in this study, they did not do so across triggers and processes, with the exclusion of maintaining and using networks which EQAs enacted ubiquitously. There are multiple potential interpretations of this finding, none which are mutually exclusive. First, it is possible that the maintaining networks serves as a blanket band-aid that works to assuage triggers across the board. Interpretation of this finding then would suggest that regardless of what else EQAs are doing, networks are essential for survival, which would be logical considering these adolescents were often left without basic necessities. In the future, scholars should attend more closely to the role of others. For example, network members might be essential in not only enacting CR with others but for others as well (i.e., other resilience). Second, it is possible that EQAs, perhaps because of their developmental stage, have not determined which processes address which triggers and instead are trying whatever they can to respond to whatever they experience in the moment. This could mean that EQAs are not investing their effort where it might count best, especially if they have higher order concerns such as trying to find housing (Maslow, 1943). Thus, it might be that even though most of the CR processes do not differ from adults, the strategic pairing of process to trigger could be lacking. This is especially possible given that many migrants experienced a similar lack of resources as EQAs (Scharp et al., 2021). Finally, it is also possible that the specific enactments of each process are not corresponding to particular triggers. For example, EQAs simply might not have come across an alternative logic, etc. that specifically responds to their triggers even if they are enacting these CR processes broadly. In the future, researchers should attend to the ways populations might enact each CR process and whether certain enactments are more effective.

In addition to differences between adults and adolescents, this study also advances research methodology. Specifically, thematic co-occurrence analysis originally characterized the relationships between and across research questions based on whether those relationships were sporadic/pervasive, unilateral/bilateral, and (un)balanced (Scharp, 2021). Findings from this study introduce two new possibilities researchers can assess based on the co-occurrence matrix: (a) ubiquity and (b) inverse relationships. Specifically, we identified two ubiquitous relationships such that negative reactions and maintaining/using social networks were common across all EQA accounts. Ubiquity, then, becomes a meaningful determination because it lets the researcher identify a pervasive commonality across participants. In our study, the ubiquitous relationships tell a story of parental rejection and the need for support in the absence of parental approval. Perhaps even more salient in this study was the inverse relationship that emerged between a lack of material resources and the absence of foregrounding productive action. Recognizing this inverse relationship attunes researchers to the ways that resilience triggers might also serve as barriers to CR. Taken together, these two methodological advancements can serve to help researchers paint a more complete picture of their participants’ experiences.

Finally, findings from the present study contribute to mounting evidence that power is a distinct tensional dimension of the CTR. Although EQAs identified lacking access to material resources as a trigger, perhaps more notable was the importance that specific vocabulary and embeddedness in queer communities played into informing the tensions between the CR processes. EQAs who had the vocabulary to discuss their sexuality/gender identity and connectedness to/embeddedness in the LGBTQ+ community were able to better enact the CR processes than those EQAs who had less access to those language and relational resources. Although all CTR research foregrounds the importance of communication in constituting resilience, this study cements the importance of attending to power to understand resilience. Thus, we argue that CR might vary not only by access to ideological (i.e., privilege) and material resources, but also communicative (i.e., vocabulary/language) and relational (i.e., community connection) resources. In this regard, communicative resources are afforded to people who have both the cognitive ability and knowledge to describe and make their situation understood by others. In some regards this corresponds to the supportive communication literature that suggests it is difficult to garner support if people are unaware that someone has a problem, the content of the problem, and what is needed to address it (see MacGeorge, 2011). Finally, relational resources refer to whether people know about and have access to a specific community instead of a general one. For example, the combination of individual resilience practices and online community practices facilitated a robust resilience response where participation in an online community filled in gaps that individual resilience enactments did not completely address (Scharp et al., 2019). Thus, a specific vocabulary and access to a support community are factors that researchers should consider when studying resilience enactment.

Furthermore, the importance of power might be evident in the inverse relationship between lack of resources and productive action. Specifically, EQAs who did not have as many resources might have been constrained in their ability to engage in productive action. This relationship calls attention to both the theoretical and practical consequences for people who lack power such as EQAs. In the future, researchers should model how intersecting disadvantages predict the severity of the triggers people experience and people’s ability to enact resilience processes in response. This contribution is particularly meaningful in moving away from trait-based definitions of resilience and hold significant promise for practical application.

Practical Applications

Unlike trait-based conceptualizations of resilience that suggest adolescents might possess some sort of “ordinary magic,” findings from this study suggest that increased education and connection to support groups could play a major difference in how well EQAs can craft a new normal for themselves (Masten, 2001, p. 277). Better education about sexual and gender identities available on the internet or even in primary and secondary school education not only could give EQAs better language to understand and describe their own experiences but also reduce the triggers they report experiencing when educating others. Specifically, educational materials that help EQAs discuss their sexuality/gender-identity in ways that resonate with other cultures might be particularly useful considering the ideological differences trigger. Indeed, it might not be surprising that the EQAs who lacked the language to understand their own situation also struggled to explain it to their parents. Furthermore, it could be particularly frustrating when the materials available do not account for cultural differences. Better materials might shift the burden EQAs experience to an opportunity to discuss their sexuality/gender-identity with their parents in ways that could reduce anything from parental disbelief to outright rejection.

Findings from our study also connect theory to practice regarding the non-co-occurrent relationships between CR triggers and processes. Holistically, EQAs were not engaging in resilience processes to address specific triggers in any patterned or apparent ways. Yet, existing research has linked both crafting normalcy and alternative logics to structural issues ostensibly out of a person’s control in adult populations (see Scharp et al., 2021, 2022). Thus, it might be useful for counselors to help EQAs see how the CR processes they already are engaging in (i.e., crafting normalcy and putting alternative logics to work) might be particularly useful in making sense of the ideological differences they perceive and lack of resources at their disposal. Even helping EQAs strategize about how their resilience enactments are addressing their triggers might facilitate efficacy or give EQAs some additional ideas about what is within their control.

Lastly, although our study examined the experiences of EQAs in the US, other countries have made significant strides in supporting estranged adolescents with material resources especially when it comes to gaining a college education. The non-profit organization StandAlone, for example, has garnered a pledge from multiple institutions across the United Kingdom soliciting free tuition and housing for college students estranged from their parents (see https://www.thestandalonepledge.org.uk). Support like this can reduce unhousedness and start EQAs on a path to help them gain financial independence. LGBTQ+ advocates and nonprofits seeking to support EQAs might consider launching a similar campaign to partner with institutions of higher learning to reduce the inequities and disparities EQAs currently experience.

Limitations and Future Directions

All studies have limitations. First, there was a 6-year age gap between our youngest and oldest participants. For EQAs, not only are there major physical and psychological differences, but also differences in their legal status, especially regarding employment. Second, because of the COVID-19 pandemic, we were required to conduct our interviews through the online platform Zoom. Therefore, only participants who had access to technological resources were able to participate. When it is safe to do so, researchers should consider engaging more directly with the LGBTQ+ community to gain more diverse inclusion.

Despite these limitations, findings from this study advance theory, method, and practice by attending to the experiences of EQAs. In the future, researchers should attend closely to the implications of negative parental reactions and how even engaging in resilience processes could be dangerous for some adolescents. For example, hostile reactions might not only lead to rejection but open up EQAs to other adults who might take advantage of them. Indeed, researchers suggest that in addition to becoming unhoused, EQAs are more likely to experience violence, experience sexual trauma, and be recruited into sex-trafficking than their cisgender heterosexual counterparts (see Forge et al., 2018). Thus, especially for adolescents, researchers might consider the ways that CR processes might create triggers in other contexts.