Parent–child relationships following spousal/parental death: An application of relational turbulence theory

Abstract

The death of a family member is a difficult experience. Although implications of loss are felt on intra- and interpersonal levels, little is known about how it affects the relational functioning of surviving family members, and in particular the parent–child relationship. Using data collected from 144 bereaved parent–child dyads, this study examined how the divergent experiences of spousally bereaved parents and parentally bereaved children impact the parent–child relationship following spousal/parental death. Drawing from relational turbulence theory (RTT), experiences with relational uncertainty and interference from a partner were explored. Findings indicate that parent–child pairs experience different types of relational uncertainty and interference from a partner and that within-dyad disagreement on these experiences can increase the amount of uncertainty and interference from a partner that an individual faces. This study extends RTT and demonstrates its generalizability to the parent–child relationship.

Keywords

Family death interference from partners parent–child relationship relational turbulence theory relational uncertainty

Grief is a complex experience that threatens people’s emotional and mental well-being (Shonkoff, Jarman, & Kohlenberg, 1987), impedes their social functioning (Wilcox et al., 2003), and compromises their physical health (Shahar, Schultz, Shahar, & Wing, 2001). These impacts are particularly apparent in familial death, as surviving family members face emotional and logistical upheaval and struggle to communicate and connect (Bosticco & Thompson, 2005; Shonkoff et al., 1987). This presents challenges to the relationships among bereaved family members, who struggle to understand the grief of their relatives (Bosticco & Thompson, 2005; Sedney, Baker, & Gross, 1994). Despite the relational implications of family death, challenges associated with studying bereavement have limited the capacity of past research, which has primarily centered the individual (Bosticco & Thompson, 2005; Keeley, 2017; Traylor, Hayslip, Kaminski, & York, 2003). Though valuable, these studies do not fully capture the interdependence of the family system and are thus unable to comprehend the extent to which individual experiences with loss impact the relationships between surviving family members.

This study examines the grieving process from a relational perspective, with the goal of understanding how the experience of one family member impacts the experience of another. Research suggests the familial context can offset the severity of grieving challenges (Hope & Hodge, 2006; Wolchik, Ma, Tein, Sandler, & Ayers, 2008), and dependability and cohesion are two indicators of positive adjustment for bereaved families (Greeff & Human, 2004; Traylor et al., 2003; Wolchik et al., 2008). In addition, open and honest communication is a valuable tool in the coping process, helping to generate a mutual understanding of the death (Bosticco & Thompson, 2005; Sedney et al., 1994). Indeed, it is through relationships that bereaved individuals make sense of loss, receive support for their grief, and adjust to a future without the deceased. This study explores how divergent experiences of spousally bereaved parents and parentally bereaved children impact the parent–child relationship following death.

One theory that may illuminate the parent–child relationship during bereavement is relational turbulence theory (RTT; Solomon, Knobloch, Theiss, & McLaren, 2016). RTT offers insight into the mechanisms that generate turbulence during relational transitions. By positioning family loss as a transitional experience capable of creating turbulence, this work extends RTT in two important ways. First, most applications of the framework have centered the romantic relationship, including dating, married, and empty-nesting couples (Nagy & Theiss, 2013; Solomon & Theiss, 2008; Steuber & Solomon, 2011; Theiss & Knobloch, 2009). Building on Scheinfeld and Worley’s (2018) work, this study tested the generalizability of the RTT to the parent–child relationship. Second, the framework has offered insight into varying relational transitions, including postdeployment military reintegration (Knobloch & Theiss, 2012) and infertility (Steuber & Solomon, 2008, 2011). This study strengthens the claims of the theory by evaluating its applicability within a new context: spousal/parental death.

The goal of this study was to examine relationships between perspectives of spousally bereaved parents and parentally bereaved children, with a focus on uncovering how the relational context creates relational turbulence. Emphasis was placed on uncovering relationally situated knowledge of familial death and extending the applicability of RTT. What follows is an overview of RTT that clarifies how the framework’s constructs are embodied following spousal/parental death. Next, a dyadic perspective of spousal/parental death is considered. Finally, using data generated from bereaved parent–child dyads and a convergent mixed methods design, guiding research questions are tested and findings are discussed.

A relational turbulence perspective

RTT (Solomon et al., 2016) highlights relational transitions as significant moments that may foster an environment wherein relational turbulence can occur. The framework argues that, during transitions, relationships are more susceptible to turbulence, because individuals are more prone to vigilance and are more likely to experience emotions intensely. The theory positions relational uncertainty and qualities of interdependence as key mechanisms that inform turbulence. The relationship between these key components and turbulence is cultivated by a set of associations between relational uncertainty, cognitive appraisals, and communicative engagement and interdependence, emotional reactions, and communication valence. RTT proposes that these processes inform perceptions of relational turbulence, leading to outcomes such as support, disclosure, and collaborative planning. The relational focus of the framework provides a means of capturing connections between spousally bereaved parents and parentally bereaved children, because it offers a way of identifying how individual perceptions impact relational outcomes.

Relational uncertainty

Relational uncertainty represents “the degree of confidence people have in their perceptions of involvement within interpersonal relationships” (Solomon & Knobloch, 2004, p. 797). It is positioned as a higher order construct, under which three types of uncertainty exist: self-uncertainty (questions people have about their own involvement in a relationship), partner uncertainty (the doubts people experience about their partner’s involvement in a relationship), and relationship uncertainty (“questions associated with the relationship itself”; Solomon & Knobloch, 2004, p. 797). RTT argues that relational uncertainty can generate turbulence by obscuring the conceptual framework of the relationship, creating challenges to sense-making, communication, and relational functioning (Solomon et al., 2016).

Although applications of RTT have not explored spousal/parental death, research supports the idea that bereaved parents and children face uncertainty. Parents and children are often unsure or unaware of their family members’ feelings and may feel inadequately prepared to communicate with them (Sedney et al., 1994; Toller & McBride, 2013). Parents and children may experience uncertainty about the functioning of the family unit and their ability to manage the emotional, logistical, and relational upheaval caused by the death (Greeff & Human, 2004; Shonkoff et al., 1987). The death may also prompt bereaved parents and children to question their relationships with surviving family members (Shonkoff et al., 1987). While families may anticipate experiences with uncertainty following a death, negative perceptions of uncertainty can increase feelings of grief and loss-related posttraumatic stress disorder (Boelen, 2010). RQ1 was designed to understand the specific aspects of relational uncertainty experienced by parent–child pairs who experienced the death of a spouse/parent:

RQ1: (a) What issues of relational uncertainty do bereaved parents and children face following spousal/parental death? (b) Do reported types of relational uncertainty relate to reported amounts of relational uncertainty?

Qualities of interdependence

The second cornerstone of RTT, qualities of interdependence, argues that relational transitions disrupt previously established patterns of relational interdependence. A primary argument of the theory is that individuals will inevitably interrupt the lives of their partners, but that their interruptions can function either as an interference or as a facilitation. Solomon, Knobloch, Theiss, and McLaren (2016) position emotional intensity as the primary manifestation of interruption, claiming that “heightened affective arousal sparked by interruptions from a partner prompt(s) stronger emotional reactions” (p. 515). Facilitative behavior typically provokes positive emotional reactions, while interference usually elicits negative emotional reactions. Despite the counteractive potential of positive emotional reactions, RTT suggests that negative emotional responses stemming from interference from a partner may be particularly unsettling during transitional moments. Therefore, interference from a partner was the focal point of this work.

Existing research suggests that loss has the capacity to interrupt the life and routines of a family unit. Bereaved parents often struggle to keep up with day-to-day routines and maintenance, finding chores, and daily activities to be exhausting and stressful (Hahn, Cichy, Small, & Almeida, 2014). They may experience financial interruptions following the death, including loss of health insurance, loss of home, and loss of primary income (McBride & Simms, 2001; Werner-Lin, Biank, & Rubenstein, 2009). Interruptions may also take the shape of logistical concerns related to immediate activities (e.g., funeral planning) as well as the ongoing functioning of the family unit (e.g., how will the kids get to school? Shonkoff et al., 1987). Interference is a crucial component to consider within the experience of family death, because families are asked to recreate new patterns of interdependence within a new familial context. This can lead to relational interferences; for example, a child may require additional support or may experience grief in a way that disrupts existing relational norms. This can also create logistical interferences; for example, a surviving parent may have to get a new job, which may interrupt childcare routines. RQ2 is posed to explore the specific types of interference from a partner (i.e., from their child or surviving parent) experienced by parent–child pairs who have lost a spouse/parent:

RQ2: (a) What types of interference from a partner do bereaved parents and children face following spousal/parental death? (b) Do reported types of interference from a partner relate to reported amounts of interference from a partner?

Dyadic processes

RTT accounts for the interactive nature of relationships, but it often prioritizes the individual as the unit of analysis; for example, RTT often highlights the role individual perceptions play in shaping relational outcomes such as collaborative planning and support. However, the past dyadic work using the relational turbulence perspective has uncovered partner effects, indicating that the perceptions of an individual impact their relational partner and the relationship (Knobloch, Knobloch-Fedders, Yorgason, Ebata, & McGlaughlin, 2017; McLaren & Solomon, 2014; Theiss & Nagy, 2010). For instance, in their work on sexual intimacy in married couples, Theiss and Nagy (2010) found that an individual’s sexual satisfaction was negatively associated with their spouse’s reports of relational uncertainty and interference from a partner. Other work has found that an individual’s evaluation of relational conflict and communication impacts their partner’s relational uncertainty and perceptions of the conflict (Knobloch & Theiss, 2010; Theiss & Knobloch, 2009). Although limited to the context of romantic relationships, this work implies that relational partners are linked during transitions.

The mutual dependence of relational partners extends into work on family death, where research highlights the need for a systems perspective, arguing that bereaved family members are interconnected (Traylor et al., 2003). For instance, past work has found that when one’s relational partner engages in productive coping techniques, a person is better able to adjust to the loss and engage in positive meaning-making (Albuquerque et al., 2017). Research has also found that perceived similarity in grief is a positive predictor of relational satisfaction, suggesting that reciprocity is important to the grieving process (Buyukcan-Tetik, Finkenauer, Schut, Stroebe, & Stroebe, 2017). These studies span varying familial death contexts but prioritize the spousal relationship (Albuquerque et al., 2017; Buyukcan-Tetik et al., 2017). RQ3 was posed to examine the ways in which the perspectives of spousally bereaved parents compare to the perspectives of parentally bereaved children:

RQ3: (a) How, if at all, do parents’ and children’s reports of relational uncertainty and interference from a partner differ following spousal/parental death? (b) Do parent/child dyads agree on reported types of relational uncertainty and interference from a partner? (c) How, if at all, does dyadic agreement on reported types of relational uncertainty and interference from a partner impact an individual’s reported amount of the variables?

Method

A convergent mixed methods design was used to answer guiding research questions. A convergent mixed methods design is defined by its ability to collect two (or more) types of data simultaneously and to subsequently treat that data with equal value during the analysis and interpretation phases of the work (see Creswell & Plano Clark, 2011 for review). This design allows the analysis to generate conceptually rich and contextually situated knowledge. In this study, qualitative and quantitative data were treated separately and with equal value initially but were merged when answering questions about the relationship between qualitative reports and quantitative indices.

Sample

Data were collected from 144 parent–child dyads who had experienced spousal/parental death. To be eligible for participation, both parents and children had to be at least 18 years of age. Traditional marriage, domestic partnership, common-law marriage, and non-married romantic partnerships were considered family, so long as both members of the dyad defined their relationship as a parent–child relationship.

Participants were recruited through snowball sampling that began with the researcher’s social network and included several grief support groups. A second round of recruitment used targeted Facebook advertising.¹ Using funds from a faculty research grant, in the second round of recruitment participants were offered a US$30 Amazon gift card for completing the survey. Institutional Review Board approval was granted for each round of data collection.

Children were mostly female (n = 117, 81.3%) and White (n = 127, 88.2%) and ranged in age from 18 to 63 (M = 31.56 and SD = 10.71), with their age at the time of their parent’s death ranging from 1.5 to 55 (M = 23.94 and SD = 12.18). Parents were mostly female (n = 129, 89.6%) and White (n = 131, 91%) and ranged in age from 38 to 84 (M = 58.93 and SD = 9.47), with their age at the time of their spouse’s death ranging from 19 to 84 (M = 51.04 and SD = 12.51). Most dyads were mother–daughter pairs (n = 105, 89.7%), and the average time since death was 88.58 months (SD = 80.67, range: 1.5–384 months).

Procedures

Data were collected as part of a larger project that included a 97-item questionnaire hosted via Qualtrics. A total of 66 items were used in this study, including two short-answer questions addressing relational uncertainty and interference from a partner (Knobloch & Theiss, 2012). Qualitative accounts of relational uncertainty were collected using an open-ended survey item that read: “It is normal for parents and children to have questions about their relationship. Please list and briefly describe issues of uncertainty you experienced within your relationship with your child/parent following the death of your spouse or partner/parent.” Reports of qualitative interference from a partner were collected using an open-ended survey item that read:

It’s normal for relational partners, such as parents and children, to interfere in each other’s everyday routines and activities once in a while. Thinking about your relationship with your parent/child (the one who is also taking this survey), please list and briefly describe ways in which they made it harder for you to complete your everyday activities following the death of your parent/spouse or partner.

To be included in the analysis, and thus compensated for their participation, both members of the parent–child dyad were required to respond meaningfully to the qualitative questions. This helped to ensure that participants did not skip questions. Responses to the relational uncertainty question ranged in length from 3 words to 530 words, with an average of 54.58 words per response (SD = 55.19). Responses to the interference from a partner’s question ranged in length from 7 words to 252 words, with an average of 38.14 words per response (SD = 33.36).

Measures

Relational uncertainty was measured using Knobloch and Solomon’s (1999) 12-item scale. Consistent with the past work that has applied RTT within parent–child relationships (Scheinfeld & Worley, 2018), the measure was modified to fit the relational and topical contexts. Each item began with “Since the death, how certain are you about…?” and was followed by a statement. The 12 statements reflected self-uncertainty (“how certain are you about your parent/child relationship”), partner uncertainty (“how certain are you about your child/parent’s view of your parent/child relationship?”), and relationship uncertainty (“how certain are you about the definition of your parent/child relationship?”). To parallel responses across the survey, responses were measured on a 7-point scale ranging from completely certain (7) to completely uncertain (1).

To identify the validity of the measure within the parent–child relational context, confirmatory factor analyses (CFA) were run. All items significantly loaded on their respective latent variables with loadings ranging from .70 to .93 (p < .001), and the model fits reasonably well for both parents (Root Mean Square Error of Approximation (RMSEA) = .10, Confirmatory Fit Index (CFI) = .96, Tucker-Lewis Index (TLI) = .95, Bentler-Bonett Nonnormed Fit Index (NFI) = .94, χ² = 127.99, p < .001) and children (RMSEA = .12, CFI = .96, TLI = .93, NFI = .92, χ² = 153.30, p < .001; Kline, 2005). Each subscale yielded good reliability (parent: self α = .92, partner α = .95, relationship α = .93 and child: self α = .90, partner α = .94, relationship α = .91). Responses were reverse-coded so that higher scores indicated greater uncertainty. Parents and children reported relatively low self-uncertainty (parent: M = 1.69, SD = 0.99 and child: M = 1.92, SD = 1.09), partner uncertainty (parent: M = 2.08, SD = 1.22 and child: M = 2.17, SD = 1.34), and relationship uncertainty (parent: M = 2.05, SD = 1.17 and child: M = 2.16, SD = 1.19).

Interference from a partner was measured using Solomon and Knobloch’s (2001) 5-item measure, wherein participants were asked to indicate their agreement with statements about their partner’s behavior. The scale was modified to fit the relational context and included statements such as “Since the death my child/parent interferes with the plans I make.” To maintain consistency across other items on the survey, responses were measured using a 7-point response scale ranging from high (1) to low (7) agreement, with higher scores indicating greater interference. CFAs found that all items significantly loaded onto the latent variable with loadings ranging from .82 to .95 (p < .001) and that the model was a good fit for both parents (RMSEA = .16, CFI = .97, TLI = .94, NFI = .96, χ² = 23.51, p < .001) and children (RMSEA = .12, CFI = .98, TLI = .95, NFI = .98, χ² = 14.65, p < .05). The 5-item measure yielded good reliability (parent: α = .94 and child: α = .93). Parents and children reported relatively low levels of interference from a partner (parent: M = 2.29, SD = 1.51 and child: M = 2.52, SD = 1.51).

To contextualize perceptions of parent–child communication, participants responded to the conversation orientation subscale of the revised family communication patterns scale (RFCP-C; Ritchie & Fitzpatrick, 1990). Family communication patterns (FCPs) are internal models of thinking that individuals draw from when processing the behaviors and interactions of their family unit (Koerner & Fitzpatrick, 2002). The framework highlights the role of open communication and homogeneity in creating a familial communicative climate. In their work, Carmon and colleagues (2010) identified that FCPs impact grief reactions following a family death and that this connection was particularly apparent for perceptions of conversation orientation, which predicted less detachment and more personal growth for the bereaved (Carmon, Western, Miller, Pearson, & Fowler, 2010). Building on their findings, this study included perceptions of conversation orientation (RFCP-C) as a covariate to account for differences in parent–child communication that may impact the individual and the relationship following a family loss. This 15-item measure asks individuals to indicate their agreement with statements such as “I usually tell my parents what I am thinking about things” on a 7-point scale. In this study, the scale yielded good reliability (parent: α = .94 and child: α = .93), and parents and children reported high levels of openness in their family communication (parent: M = 5.36, SD = 1.07 and child: M = 5.07, SD = 1.20).

Results

Data were analyzed at three levels: qualitative, quantitative, and mixed. First, inductive thematic analysis was conducted by the author and a team of trained assistants (Bulmer, 1979). Open-coding was used to generate an initial set of categories, which were verified and subsequently used to code the entire data set (Strauss & Corbin, 1990). While specific manifestations of uncertainty and interference from a partner were evaluated in the analysis, identifying the root cause of the reported experiences was a key to developing the coding scheme; thus, specific experiences may have varied within the individual themes, but the source of the uncertainty or interference from a partner remained the same. To ensure reliability of the resulting themes, two other assistants conducted the second part of the qualitative analysis. Cohen’s κ was calculated and indicated that both the relational uncertainty and interference from a partner’s thematic schemes were reliable (.80 and .82, respectively). Throughout this process, the unit of analysis was the individual, meaning each participant’s qualitative responses received codes independent of their dyadic partner’s.

Following the inductive thematic analysis, quantitative analysis was done using dyadic data analysis techniques. Finally, the numeric codes from the inductive thematic analysis were entered into SPSS, forming nominal variables for types of relational uncertainty and interference from a partner. This quantification process required that each qualitative response be assigned only one numeric code. In instances when disagreement between coders occurred, the response was discussed until a code was agreed upon. In cases where qualitative responses fit more than one thematic category, codes were decided on based on which theme was most apparent. These decisions were made at a joint meeting which involved the assistants who aided in the development of the initial set of categories and could, therefore, provide insight into the intention of each theme.

Preliminary analysis

Paired sample t-tests were run to identify any significant differences between participants recruited in the first and second round of recruitment (n = 29 and n = 115, respectively); none were found, so the data sets were merged. Additional t-tests were run to identify any significant differences between parent and child reports of key variables. Results showed that children’s reports of self-uncertainty (M = 1.92 and SD = 1.09) were slightly higher than parents’, M = 1.69, SD = 0.99; t(143) = 2.38, p = .02. Bivariate correlations evaluated if demographic variables such as time since the death and age (at time of death and survey) as well as responses to the RFCP-C scale were significantly related to any of the key variables (see Table 1). Although age and time since death varied across the sample, analysis revealed that only the RFCP-C was significantly and strongly correlated with the quantitative relational uncertainty and interference from a partner’s indices; therefore, the RFCP-C was a covariate in the substantive analysis.

Table 1.

Bivariate correlations.

	A	B	C	D	E	F	G	H	I	J	K	L	M	N
A
B	.82**
C	.89**	.81**
D	−.18*	.13	.30**
E	.37**	.39**	.39**	.07
F	.44**	.49**	.46**	.09	.87**
G	.40**	.45**	.44**	.13	.88**	.91**
H	.05	.14	.11	.21*	.20*	.29**	.39**
I	−.51**	−.51**	−.60**	−.31**	−.36**	−.38**	−.41**	−.14
J	−.29**	−.29**	−.26**	−.12	−.30**	−.38**	−.39**	−.16*	.38**
K	−.10	−.09	−.13	−.08	−.13	−.16	−.19	−.15	−.02	−.01
L	−.05	−.02	−.06	−.08	−.08	−.08	−.12	−.17	−.02	−.10	.81**
M	−.05	−.02	−.06	−.03	−.10	−.11	−.13	−.19	−.12	−.12	.66**	.54**
N	−.01	.08	−.001	−.05	−.04	−.02	−.04	−.18	−.09	−.19*	.47**	.72**	.74**
O	−.06	−.12	−.07	−.01	−.04	−.09	−.10	.03	−.07	.10	.13	−.43**	.09	−.45**

Note. (A) = self-child, (B) = partner-child, (C) = relationship-child, (D) = interference-child, (E) = self-parent, (F) = partner-parent, (G) = relationship-parent, (H) = interference-parent, (I) = RFCP-child, (J) = RFCP-parent, (K) = age (now)-child, (L) = age (death)-child, (M) = age (now)-parent, (N) = age (death)-parent, and (O) = time since loss. RFCP = revised family communication patterns.

*p < .05; **p < .01.

Relational uncertainty: Qualitative themes

Analysis revealed five themes in relational uncertainty, along with a category of those who reported no uncertainty (see Table 2).

Table 2.

Qualitative theme distribution—Relational uncertainty.

	Total N (%)	Parent n (%)	Child n (%)
Relational context“I felt so sorry for my daughter and son that were still at home with me. I spoiled them and let them have their way a lot.” (Parent 66).“I would like to keep my stepmother in my life, but there is a part of my that is concerned that we will grow apart now that my dad is no longer alive and she isn’t “obligated to be my stepmother” anymore (Child 106).	59 (20.5)	26 (18.1)	33 (22.9)
The future (well-being)“Sandy was already a very emotional child before her father died. I was really worried about her, and the impact his death would have on her” (Parent 38).“My only uncertainty with my mother was if she was how long she was going to be alive for, my worry was if when I woke up she wouldn’t be alive, or at any minute I could lose her” (Child 11).	47 (16.3)	21 (14.6)	26 (18.1)
The future (functional)“Scared very uncertain about the future. Being able to keep our home, etc. he died with no life insurance. Very unprepared.” (Parent 91).“My mother didn’t work when my dad was alive so we weren’t sure where the money was going to come from” (Child 52).	30 (10.4)	18 (12.5)	12 (8.3)
Talking about the death“I’m not sure how much to talk about the death with mom. I don’t want to make her cry” (Child 9).“I had issues about when or how to discuss my daughter’s father’s death with her” (Parent 42).	39 (13.5)	16 (11.1)	23 (16.0)
Support“I struggled with the balance of dealing with my own deep feelings of loss while also helping my kids whom I desperately wanted to spare the pain, handle their feelings of loss” (Parent 106).“Uncertain around how to make her feel wholly loved and supported” (Child 8).	32 (11.1)	16 (11.1)	16 (11.1)
Grief“I didn’t understand how she could go through all of his stuff so quickly and get rid of it. In hindsight, I can see that she needed to do it for her own process. But it hurt me. I felt like she was trying to erase his memory…I just couldn’t be there while she was taking all of his clothes and shoes and giving them away or donating them.” (Child 131).“Not sure she has really grieved the loss of her dad” (Parent 5).	31 (10.8)	23 (16.0)	8 (5.6)
No uncertainty	50 (17.4)	24 (16.7)	26 (18.1)

Relational context

Participants expressed uncertainty about their relational context, which had changed because of the death. These included concerns about adjusting to the “new normal,” (Parent 30), “After my husband’s death I was worried about how I would relate to my daughter. We had never had a bad relationship, but it always seemed that she was ‘daddys little girl’” (Parent 18). Others shared that the death had prompted changes to the family context, “It changed when he died. She is less sure of herself and less forceful as a parent. She doesn’t follow through on punishments and basically lets us do whatever we want” (Child 20).

The future

A second theme centered on uncertainty about the future, including concerns related to anticipated experiences and challenges. Participants who expressed this type of uncertainty shared anxiety about potential shifts that had occurred since the death. These concerns targeted two primary elements: the safety and health of their relational partner, and every day functioning of their family, resulting in two subthemes.

Future well-being centered uncertainty about the well-being of one’s relational partner. These concerns focused on how the death would impact the health and happiness of one’s relational partner, “How will this effect their desire to stay close? How does this change them as a person? Will they be stronger? Will they close themselves off?” (Parent 48). Within this theme, there were many children who expressed concerns about their surviving parent’s mortality, sharing uncertainty about whether their surviving parent would leave them too. For example, “I was worried my mom would die next and that I would be an orphan” (Child 63).

Future functional included uncertainty about logistical changes that had arisen because of the death. These concerns focused on the day-to-day family functioning, “I worried my mom couldn’t handle all the responsibilities of raising me, taking care of the house…paying bills all by herself” (Child 136), and

I felt like I was not up to the task of being both father and mother, that David was the person that held the family together and made the tough decisions. He was so wonderful at talking to them about sports, school and various disappointments they experienced…he was also the funny one, that made trips enjoyable, etc. I felt like I had to make things better, but had no idea how to. (Parent 121)

Talking about the death

Participants reported uncertainty related to talking about the death, “I wasn’t sure how much to tell them, or how much they would really understand. And then as they got older, I questioned how much detail to share with them, and how often to bring up their dad.” (Parent 57). Participants also shared uncertainty related to their communicative behaviors, often feeling like they were not communicating enough or in the appropriate way “I’m not sure how much to talk about the death with mom. I don’t want to make her cry” (Child 9).

Support

Participants expressed uncertainty about support, both the amount of support they received and their own abilities to provide support. For some, uncertainty stemmed from their own perceived (in)ability to provide support, “I was self-centered in my grief and I’m not sure if he felt that I supported, or was there to comfort him.” (Parent 73). However, others expressed concerns about their relational partner’s ability to provide support, “My mom was very material…while she provided our basic needs, we did not have the emotional support we needed” (Child 65).

Grief

A fifth theme focused on uncertainty about grief and concern related to the grieving process and the grief reactions of individuals and their relational partners. Participants were uncertain about if their parent or child had grieved properly, “Feeling like he wanted to move on too quickly. From wearing his wedding band, to clearing out her belongings (which I understand) to feeling like he wanted to find someone new” (Child 7). Participants were also uncertain about if their relational partner understood their grief “I feel the only issue we had was she didn’t understand the intensity of my grief as a spouse” (Parent 77).

Relational uncertainty: Mixed methods results

To answer RQ1b, multivariate analysis of covariance (MANCOVA) tests were run with qualitative types of relational uncertainty as the independent variable (IV), quantitative amount of relational uncertainty as the dependent variable (DV), and RFCP-C as the covariate. For both children and parents, Box’s M was significant (p < .001), and Pillai’s trace was used. Children’s type of relational uncertainty did not predict their quantitative reports of uncertainty, Pillai’s trace = .17, F(18, 408) = 1.40, p = .13, multivariate η² = .06. However, for parents, the MANCOVA was significant, Pillai’s trace = .21, F(18, 408) = 1.71, p = .04, multivariate η² = .07, meaning that the type of uncertainty they reported predicted how much relational uncertainty they experienced. This effect took place across all three subdivisions of relational uncertainty: self (F(6, 136) = 2.19, p = .04, partial η² = .09), partner (F(6, 136) = 2.73, p = .03, partial η² = .10), and relationship (F(6, 136) = 2.66, p = .04, partial η² = .09). Pairwise comparisons using the Bonferroni adjustment identified that those who experienced uncertainty related to support, as revealed by the qualitative data, reported significantly more partner and relationship uncertainty than those who shared no qualitative uncertainty (p = .04 and .03, respectively).

Interference from a partner: Qualitative themes

A total of four content themes were revealed, as well a fifth category for those who experienced no interference (see Table 3).

Table 3.

Qualitative theme distribution—Interference from a partner.

	Total N (%)	Parent n (%)	Child n (%)
Shifting roles and responsibilities“We almost immediately instituted a “plan” that I would be with her at least one weekend every month. This required me to talk to my current employer about flight patterns and the need for a day off of work each month. It required finding things to do to keep her entertained.” (Child 58).“Life is very difficult for many years having to be both the mother and the father” (Parent 41).	57 (19.8)	26 (18.1)	31 (21.5)
Loss of independence (internal)“I believe that I worry about her feelings and how she is adjusting to the death of her dad” (Parent 123).“I don’t think she ever did intentionally but, I sometimes I was worried about her feeling lonely so I would change plans to hang out with her” (Child 30).	40 (13.9)	21 (14.6)	19 (13.2)
Loss of independence (external)“They have wanted to be a part of EVERYTHING. If she wasn’t it was a problem” (Child 27).“She always concerned about me, overly protective” (Parent 53).	32 (11.1)	13 (9.0)	19 (13.2)
Expressive capacity“My Mom rarely answers the phone, preferring to talk by text but still doesn’t answer those for days sometimes” (Child 51).“She didn’t except that she wouldn’t really talk to me about him at all. She avoided me and everyone else for a while” (Parent 12).	34 (11.8)	15 (10.4)	19 (13.2)
Expectations“She became a depressed, single mother. She turned to alcohol and her friends rather than us so my sister and I spent a lot of time alone or with a babysitter. It made it hard for us to connect with her and build any kind of healthy relationship” (Child 19).“Acting out, doing things she would normally not do to get in trouble at home. Missing school more frequently, having emotional outbursts” (Parent 42).	31 (10.8)	15 (10.4)	16 (11.1)
Facilitation“My dad doesn’t make it harder in any way to complete my every day activities. It actually is easier now, he is able to help watch my kids more as he doesn’t have to care for my mom” (Child 7).“My daughter had not made it harder for me to complete daily activities since my husband’s death. If anything, she makes it easier by driving me to appointments and such” (Parent 115).	39 (13.5)	24 (16.7)	15 (10.4)
No interference	55 (19.1)	30 (20.8)	25 (17.4)

Shifting roles and responsibilities

A first theme centered on disruptions to the roles and responsibilities of participants that were caused by their relational partner following the death. This category focused on practical disruptions to day-to-day life,

Both of my kids were young, 6 and 1 1/2 years old, when my husband died. So my day-to-day for at least a few months was only taking care of them. Then I decided to go back to school and so I had to find babysitters, and was only able to do homework after 10 PM at night. I had to take it slowly so I could be a single-parent. Even projects around the house like painting, was a challenge and often happened late at night. I even hired a babysitter for a few hours each week just to give me time to take a shower and clean the house. (Parent 57)

Other experiences within this theme captured shifting or new responsibilities, “It was hard to stay a child. I had to cook when my mom was gone, I had to be responsible for myself because she wasn’t always around” (Child 52).

Loss of independence

A second theme captured a loss of independence that had occurred since the death. The parent–child relationship was central to this theme, as a key feature of this type of interference was the struggle between being an autonomous individual and being a compassionate and supportive parent or child. Individuals who experienced a loss of independence struggled with their relational expectations following the death, and these challenges manifested both internally and externally, resulting in two contrasting subthemes.

Internal loss of independence stemmed from a fixation with the emotional and physical well-being of one’s relational partner. For instance, “sometimes it added distraction and stress knowing my mother had a sadness I could not help with or fix” (Child 74), and “I lost sleep over worrying about how she was coping with losing a parent at a young age” (Parent 87). For these individuals, feelings of empathy had become excessive, causing them to be preoccupied with and overwhelmed by the welfare of their relational partner. Though the relational partner was not directly responsible for the interference, the relational context, and thus the individual, is central to the excessive emotional reaction that sparked the interference.

External loss of independence centered on experiences where the individual felt a loss of autonomy stemming directly from the dependency of their relational partner. These individuals saw the behavior of their relational partner as a disruption to their independence. For example, “She wants to be with me all the time. She calls me 3 or 4 times a day and always wants to know where I am” (Child 9), and “She’s always here. Relies on me to take her EVERYWHERE” (Parent 43).

Expressive capacity

A third theme embraced interference to one’s ability to express themselves and to communicate effectively with their parent or child. These participants shared that their relational partner created an environment wherein they did not feel comfortable sharing their feelings, “My Mother is a very stoic person, she will not tolerate my tears around her—‘death is a part of life’, therefore I have had to deal with my grief when my mother is not around” (Child 18). Other participants in this category shared that their relational partner had become distant from them, making communication difficult,

I may have thought they were doing ok and yet discovered they were not such as times that I would catch them breaking down when I thought things were going ok. In those times, I wish they would have been able to come to me and talk. (Parent 58)

Expectations

A fourth theme captured disruptions to one’s expectations for behavior and relationships following the death. This theme centered moments where one’s relational partner interfered with their hopes for themselves or for the relationship, “I did feel like we did not get to spend as much quality time together. This is really what upset me” (Child 106). Relational partners also differed in their expectations for aspects of the grieving process in disruptive ways, “Lack of motivation. Shutting himself off from everyone. Arguing over chores, leaving his room, getting a job, spending time with people who aren’t on his computer” (Parent 69).

Facilitation

Though not solicited in the survey, many participants responded to the question about interference by sharing explicit ways their relational partner had made their life better following the death. These responses went beyond sharing that their partner had not interfered with their life and included specific mention of ways their parent or child helped facilitate a better life. For instance, “I think that it’s been easier to complete everyday activities. My mom has me to help her around the house now, and help her pay bills and we do our grocery shopping together” (Child 91), and “My life is easier with her. I have someone to talk to. Do errands for me. We cook meals together. She makes my life easier” (Parent 39).

Interference from a partner: Mixed methods results

To answer RQ2b, analysis of covariance (ANCOVA) tests were run with qualitative types of interference as the IV, quantitative amount of interference as the DV, and RFCP-C as the covariate. For children, the ANCOVA was significant, F(6, 136) = 2.30, p = .04, partial η² = .09. Pairwise comparisons using the Bonferroni adjustment revealed that children who experienced external loss of independence, based on the qualitative data, reported significantly more interference than those who shared no interference (p = .03). For parents, the analysis was also significant, F(6, 137) = 8.59, p = .001, partial η² = .27. Pairwise comparisons revealed several significant between-group differences. Those in the external loss of independence category and the shifting roles and responsibilities category reported significantly more interference than those in the expressive capacity, no interference, and facilitation categories (p = .04, .000, and .001, respectively). Those in the internal loss of independence category reported significantly more interference than those in the no interference and facilitation categories (p = .01).

Dyadic processes

The last set of analyses dealt with RQ3 and reflected the dyadic approach taken in this study. Chi-squared analyses revealed that parents and children did not significantly differ in their reports of qualitative relational uncertainty, χ² (6, N = 288) = 11.16, p = .08; see Table 2, or their qualitative reports of interference from a partner, χ² (6, N = 288) = 4.70, p = .58; RQ3a; see Table 3. While RQ3a focused on between-group differences, RQ3b looked at within-dyad agreement on reported types of relational uncertainty and interference from a partner. The advisable method for testing dyadic agreement for categorical variables with two or more levels is Cohen’s κ, and with κ scores <.2 representing poor agreement (Kenny, Kashy, & Cook, 2006). Analysis indicated that parent–child dyads did not agree on reported types of relational uncertainty (κ = .10, p = .008) or interference from a partner (κ = .07, p = .02).

The final part of RQ3 focused on the impact of within-dyad disagreement on an individual’s reported amount of relational uncertainty and interference from a partner. To answer this question, two additional variables were created to identify the level of agreement between parents and children on the qualitative reports of uncertainty and interference. This question was answered using the nominal variables created by quantifying the qualitative codes. As with the mixed methods analysis, the unit of analysis was the individual. Level of agreement was categorized four ways: (a) pure agreement, where both individuals reported the same type of uncertainty or interference; (b) thematic disagreement, where both individuals reported some types of uncertainty or interference; (c) child-oriented disagreement, where a parent reported uncertainty or interference, but the child did not; and (d) parent-oriented disagreement, where a child reported uncertainty or interference, but the parent did not. By separating agreement into four categories, this analysis was able to account for dyads where one individual reported relational uncertainty or interference and the other did not, which is different than when relational partners both agree that uncertainty or interference has occurred but disagree on how it takes shape. After the variables were computed, analyses of variance (ANOVAs) were used to identify if type of agreement led to differences in parent and child reports of relational uncertainty and interference from a partner. To increase the conceptual clarity of the pure agreement category, parent–child dyads, where both individuals reported no uncertainty or interference from a partner, were excluded from this analysis.

While no significant impacts were found for children, the first ANOVA revealed that parents’ reports of self, partner, and relationship uncertainty were significantly different depending on level of agreement, F(3, 135) = 4.32, p = . 009; F(3, 135) = 3.75, p = . 02; and F(3, 135) = 4.42, p = . 009, respectively. Tukey’s post hoc analysis revealed that parents who reported no uncertainty but whose child reported uncertainty (i.e., parent-oriented disagreement) reported significantly less partner and relationship uncertainty than those in the other three categories, and significantly less self-uncertainty than those in the pure agreement and thematic disagreement groups (p < .05). For interference from a partner, the results were nonsignificant for children, but significant for parents, F(3, 124) = 10.86, p = . 001. Tukey’s post hoc analysis revealed that those in the parent-oriented disagreement group reported significantly less interference from a partner than those in the other three categories (p = .04–.003). Parents who reported interference when their child did not (i.e., child-oriented disagreement) reported significantly more interference than those in the other three categories (p = .03–.007)

Discussion

This study used RTT to explore the relationship between the experiences of spousally bereaved parents and parentally bereaved children, emphasizing how perceptions of turbulence were affected by the relational context. Results revealed specific ways relational uncertainty and interference from a partner were felt in parent–child relationships following spousal/parental death. RQ1 found that parents and children experience uncertainty stemming from shifts in communicative and relational norms, as well as anxiety about the future well-being of their relationship and relational partner. Parents experienced heightened levels of uncertainty when their insecurity stemmed from questions about their ability to support their grieving child. Consistent with past research (Greeff & Human, 2004; Shonkoff et al., 1987), this emphasizes that being a good parent and supporting one’s child remain a fundamental concern, even in the face of loss.

RQ2 revealed that parents and children experienced interruptions related to shifts in relational expectations and responsibilities, as well as their ability to communicate. An external loss of independence, marked by hypervigilance on the part of one’s surviving parent or child, was linked to higher levels of interference from a partner. While existing research suggests that a fear of abandonment is linked to higher levels of grief in parentally bereaved children (Wolchik et al., 2008), results from this study demonstrate the inverse of abandonment issues, highlighting the relational consequences of hypervigilance. RQ2 also demonstrates a key component of RTT that is often understudied in research on grief and death: facilitation. Though asked to share about interference from a partner, participants mentioned ways their relational partner supported them and made the transition to life post-spousal/parental death easier.

RQ3 explored the relational implications of uncertainty and interference from a partner from within the parent–child relationship. Results revealed that though relational partners do not always agree, spousally bereaved parents and parentally bereaved children do not experience different types of relational uncertainty and interference from a partner. This finding is important because it shifts the conversation from how the experiences of bereaved parents are unique from the experiences of bereaved children and recognizes that experiences with death may be similar across the family unit.

Theoretical significance

The experiences shared by participants in this study suggest that some types of relational uncertainty and interference from a partner may be similar across relational contexts and transitions. This was apparent when looking at themes identified in Knobloch and Theiss’ (2012) work on postdeployment reintegration for military couples. In both Knobloch and Theiss’ work and this study, participants emphasized logistical/household stressors, communication, and the well-being of their partner as causes of relational uncertainty. Both populations also shared about disruptions to everyday routines and chores, parenting, and independence. Although the experiences were contextually specific, both populations were required to create new routines and think differently about their family functioning as a result of a major change in the presence of a spouse. Further, much of what participants in both studies shared was centered around the act of coping with grief (Neimeyer, Prigerson, & Davies, 2002). While the specific cause of grief is a crucial part of the experience, these findings suggest that some challenges may be embedded in all relational transitions, regardless of context.

Findings also extend RTT by demonstrating its testability using dyadic data analysis techniques and further solidify its generalizability to the parent–child relationship. Results highlight the interdependence of parent–child pairs and emphasize the connection between individual appraisals and dyadic patterns. This study strengthens our conceptual understanding of the RTT by providing insight into the lived experiences of relational uncertainty and interference from a partner for bereaved families, but also broadens the contextual application of the framework by supporting its claims dyadically and within the parent–child relationship.

Social significance

Consistent with past work, participants in this study emphasized challenges related to logistical changes that had occurred since the loss, including redistribution of household chores, changes to primary income, and shifts in disciplinary styles (Greeff & Human, 2004; Shonkoff et al., 1987). However, participants also reported challenges that captured two extremes of the relational consequences of bereavement, hypervigilance, and detachment. These challenges help to explain several behavioral and relational changes bereaved individual’s experience. For instance, bereaved parents and children typically experience high levels of anxiety (Brent, Melhem, Donohoe, & Walker, 2009; Wilcox et al., 2003), which may be the result of a preoccupation with the well-being and adjustment of one’s relational partner. Further, the loss of a family member can create distance between surviving members of the family unit (Traylor et al., 2003; Walter & McCoyd, 2016). This study suggests that detachment may be the result of differing or disrupted expectations. Therefore, while practical support may be a key factor in creating an environment within which the family can grieve more effectively, it is also important to monitor shifts in the parent–child relationship, such as hypervigilance, which can create turbulence for both parents and children.

Additionally, results highlight the importance of considering family death from a relational perspective. Consistent with past research on bereaved marital couples, this study found that reciprocity is an important component of the relational coping process, particularly for parents (Buyukcan-Tetik et al., 2017). Specifically, disagreement about the existence or type of relational uncertainty and interference from a partner was related to an increase in the degree to which parents reported experiencing the two variables. This emphasizes the interconnected nature of bereavement within the family unit and suggests that mutually constructed understandings of the death, and its subsequent challenges, are vital to the coping efforts of a family (Bosticco & Thompson, 2005; Sedney et al., 1994).

Future directions and limitations

Several limitations to this study create pathways for future research. Namely, while age and time since death were not significantly related to outcome variables in this study, they are important variables to consider in any study on bereavement. Specifically, bereaved children who lost parents at a young age may have limited recollection of life when their deceased parent was alive. Further, the type of uncertainty and interference from a partner-bereaved person’s experience may shift over the course of their grieving process. For example, uncertainty related to the future and interferences stemming from shifting roles and responsibilities may be more discernible for relationships where the loss is recent and relational and logistical changes are occurring, while a loss of independence may be more visible years later when an aging parent or adult child needs support. Future RTT research should embrace a life span perspective and explore uncertainty and interference from a partner across life stages.

Finally, results indicated participants experienced relatively low levels of uncertainty and interference from a partner, which is typical of close relationships. Several participants even went as far as sharing ways their relational partner had made life easier since the death even when asked questions about interference. Research on family death often centers the adverse implications of loss, but these findings suggest that considering the optimistic and constructive effects of death may be equally important. Results also encourage a different perspective from which to explore RTT. Theoretically, negative emotional responses driven by interferences from a partner are thought to outweigh positive emotional responses that stem from facilitation (Solomon et al., 2016). However, participants in this study demonstrate the power of positivity by emphasizing the facilitative behaviors of their relational partners, even in the face of great loss. In fact, results suggest that the facilitative behaviors happened as a result of the death, suggesting that family death can encourage relational partners to find ways to help one another. Therefore, while not within the scope of this study, future research should build on Theiss, Carpenter, and Leustek’s (2016) work and examine the role of certainty and partner facilitation in relational transitions.

Conclusion

This study increased knowledge of family death and advanced RTT. It provided in-depth explanations of the relational uncertainty and interference from a partner faced by bereaved families. It extended knowledge of familial loss, underscoring the importance of managing logistical challenges and emphasizing that the parent–child relationship remains interdependent, even in the face of loss. Findings answer calls for more dyadic applications of the RTT, and further extend these requests by embracing a mixed methods approach. By centering the voices of the individuals, this study also provides a heuristic understanding of the experiences of bereaved families, making their problems more accessible to researchers and grief service providers.

Footnotes

Author’s note

This article is based on the author’s dissertation study.

Acknowledgements

The author would like to thank Dr. Mary Claire Morr Serewicz, her dissertation advisor, and Dr. Denise Solomon for their guidance and suggestions; and Corey Callahan, Kelsea Schoenbauer, and Caleb Green for their help in coding data.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was partially funded by a faculty research grant from the Department of Communication Arts and Sciences at The Pennsylvania State University.

ORCID iD

Veronica A. Droser

Open research statement

As part of International Association for Relationship Research’s encouragement of open research practices, the author has provided the following information: This research was not preregistered. The data used in the research are not available. The materials used in the research are available. The materials can be obtained by emailing: Vad18@psu.edu.

Note

References

Albuquerque

Buyukcan-Tetik

Stroebe

Schut

Narciso

Pereira

Finkenauer

(2017). Meaning and coping orientation of bereaved parents: Individual and dyadic processes. Plos One, 12, 1932–6203. doi:10.1371/journal.pone.0178861

Boelen

(2010). Intolerance of uncertainty and emotional distress following the death of a loved one. Anxiety, Stress, & Coping, 23, 471–478. doi:10.1080/10615800903494135

Bosticco

Thompson

. (2005). The role of communication and story telling in the family grieving system. Journal of Family Communication, 5, 255–278. doi:10.1207/s15327698jfc0504_2

Brent

Melhem

Donohoe

M. B.

Walker

(2009). The incidence and course of depression in bereaved youth 21 months after the loss of a parent to suicide, accident, or sudden natural death. The American Journal of Psychiatry, 166, 786–794. doi:10.1176/appi.ajp.2009.08081244

Bulmer

(1979). Concepts in the analysis of qualitative data. The Sociological Review, 27, 651–677. doi:10.1111/j.1467-954x.1979.tb00354.x

Buyukcan-Tetik

Finkenauer

Schut

Stroebe

(2017). The impact of bereaved parents’ perceived grief similarity on relationship satisfaction. Journal of Family Psychology, 31, 409–419. doi:10.1037/fam0000252

Carmon

A. F.

Western

K. J.

Miller

A. N.

Pearson

J. C.

Fowler

M. R.

(2010). Grieving those we’ve lost: An examination of family communication patterns and grief reactions. Communication Research Reports, 27, 253–262. doi:10.1080/08824096.2010.496329

Creswell

J. W.

Plano Clark

V. L.

(2011). Designing and conducting mixed methods research (2nd ed.). Thousand Oaks, CA: SAGE.

Greeff

A. P.

Human

(2004). Resilience in families in which a parent has died. The American Journal of Family Therapy, 32, 27–42. doi:10.1080/01926180490255765

10.

Hahn

E. A.

Cichy

K. E.

Small

B. J.

Almeida

D. M.

(2014). Daily emotional and physical reactivity to stressors among widowed and married older adults. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 69, 19–28. doi:10.1093/geronb/gbt035

11.

Hope

R. M.

Hodge

D. M.

(2006). Factors affecting children’s adjustment to the death of a parent: The social work professional’s viewpoint. Child Adolescent Social Work, 23, 107–126. doi:10.1007/s10560-006-0045-x

12.

Keeley

M. P.

(2017). Family communication at the end of life. Behavioral Science, 7, 45. doi:10.3390/bs7030045

13.

Kenny

D. A.

Kashy

D. A.

Cook

W. L.

(2006). Dyadic data analysis. New York, NY: The Guilford Press.

14.

Kline

R. B.

(2005). Principles and practice of structural equation modeling. London, UK: Guilford Press.

15.

Knobloch

L. K.

Knobloch-Fedders

L. M.

Yorgason

J. B.

Ebata

A. T.

McGlaughlin

P. C.

(2017). Military children’s difficulty with reintegration after deployment: A relational turbulence model perspective. Journal of Family Psychology, 31, 542–552. doi:10.1037/fam0000299

16.

Knobloch

L. K.

Solomon

D. H.

(1999). Measuring the sources and content of relational uncertainty. Communication Studies, 50, 261–278. doi:10.1080/10510979909388499

17.

Knobloch

L. K.

Theiss

J. A.

(2010). An actor—partner interdependence model of relational turbulence: Cognitions and emotions. Journal of Social and Personal Relationships, 27, 595–619. doi:10.1177/0265407510368967

18.

Knobloch

L. K.

Theiss

J. A.

(2012). Experiences of U.S. military couples during the post-deployment transition: Applying the relational turbulence model. Journal of Social and Personal Relationships, 29, 423–450. doi:10.1177/0265407511431186

19.

Koerner

A. F.

Fitzpatrick

M. A.

(2002). Toward a theory of family communication. Communication Theory, 12, 70–91. doi:10.1093/ct/12.1.70

20.

McBride

Simms

(2001). Death in the family: Adapting a family systems framework to the grief process. American Journal of Family Therapy, 29, 59–73. doi:10.1080/019261801750182414

21.

McLaren

R. M.

Solomon

D. H.

(2014). Victim and perpetrator accounts of hurtful messages: An actor–partner interdependence model. Human Communication Research, 40, 291–308. doi:10.1111/hcre.12031

22.

Nagy

M. E.

Theiss

J. A.

(2013). Applying the relational turbulence model to the empty-nest transition: Sources of relationship change, relational uncertainty, and interference from partners. Journal of Family Communication, 13, 280–300. doi:10.1080/15267431.2013.823430

23.

Neimeyer

R. A.

Prigerson

H. G.

Davies

(2002). Mourning and meaning. American Behavioral Scientist, 46, 235–251. doi:10.1177/000276402236676

24.

Ritchie

L. D.

Fitzpatrick

M. A.

(1990). Family communication patterns: Measuring intrapersonal perceptions of interpersonal relationships. Communication Research, 17, 523–544. doi:10.1177/009365090017004007

25.

Scheinfeld

Worley

(2018). Understanding the parent-child relationship during the transition into college and emerging adulthood using the relational turbulence theory. Communication Quarterly, 66, 444–462. doi:10.1080/01463373.2018.1443954

26.

Sedney

M. A.

Baker

J. E.

Gross

. (1994). “The story” of a death: Therapeutic considerations with bereaved families. Journal of Marital and Family Therapy, 20, 287–296. doi:10.1111/j.1752-0606.1994.tb00116.x

27.

Shahar

D. R.

Schultz

Shahar

Wing

R. R.

(2001). The effect of widowhood on weight change, dietary intake, and eating behavior in the elderly population. Journal of Aging and Health, 13, 186–199. doi:10.1177/089826430101300202

28.

Shonkoff

J. P.

Jarman

F. C.

Kohlenberg

T. M.

(1987). Family transitions, crises, and adaptations. Current Problems in Pediatrics, 17, 507–553. doi:10.1016/0045-9380(87)90017-x

29.

Solomon

D. H.

Knobloch

L. K.

(2001). Relationship uncertainty, partner interference, and intimacy within dating relationships. Journal of Social and Personal Relationships, 18, 804–820. doi:10.1177/0265407501186004

30.

Solomon

D. H.

Knobloch

L. K.

(2004). A model of relational turbulence: The role of intimacy, relational uncertainty, and interference from partners in appraisals of irritations. Journal of Social and Personal Relationships, 21, 795–816. doi:10.1177/0265407504047838

31.

Solomon

D. H.

Knobloch

L. K.

Theiss

J. A.

McLaren

R. M.

(2016). Relational turbulence theory: Explaining variation in subjective experiences and communication within romantic relationships. Human Communication Research, 42, 507–532. doi:10.1111/hcre.12091

32.

Solomon

D. H.

Theiss

J. A.

(2008). A longitudinal test of the relational turbulence model of romantic relationship development. Personal Relationships, 15, 339–357. doi:10.1111/j.1475-6811.2008.00202.x

33.

Steuber

K. R.

Solomon

D. H.

(2008). Relational uncertainty, partner interference, and infertility: A qualitative study of discourse within online forums. Journal of Social and Personal Relationships, 25, 831–855. doi:10.1177/0265407508096698

34.

Steuber

K. R.

Solomon

D. H.

(2011). Factors that predict married partners’ disclosures about infertility to social network members. Journal of Applied Communication Research, 39, 250–270. doi:10.1080/00909882.2011.585401

35.

Strauss

Corbin

(1990). Basics of qualitative research (Vol. 15). Newbury Park, CA: SAGE.

36.

Theiss

J. A.

Carpenter

A. M.

Leustek

(2016). Partner facilitation and partner interference in individuals’ weight loss goals. Qualitative Health Research, 26, 1318–1330. doi:10.1177/1049732315583980

37.

Theiss

J. A.

Knobloch

L. K.

(2009). An actor-partner interdependence model of irritations in romantic relationships. Communication Research, 36, 510–537. doi:10.1177/0093650209333033

38.

Theiss

J. A.

Nagy

M. E.

(2010). Actor-partner effects in the associations between relationship characteristics and reactions to marital sexual intimacy. Journal of Social and Personal Relationships, 27, 1089–1109. doi:10.1177/0265407510381254

39.

Toller

P. W.

McBride

M. C.

(2013). Enacting privacy rules and protecting disclosure recipients: Parents’ communication with children following the death of a family member. Journal of Family Communication, 13, 32–45. doi:10.1080/15267431.2012.742091

40.

Traylor

E. S.

Hayslip

Kaminsk

P. L.

York

(2003). Relationships between grief and family system characteristics: A cross lagged longitudinal analysis. Death Studies, 27, 575–601. doi:10.1080/07481180302897

41.

Walter

C. A.

McCoyd

J. L. M.

(2016). Grief and loss across the lifespan: A biopsychosocial perspective (2nd ed.). New York: Springer Publishing Company.

42.

Werner-Lin

Biank

N. M.

Rubenstein

. (2009). There’s no place like home: Preparing children for geographical and relational attachment disruptions following parental death to cancer. Clinical Social Work, 38, 132–143. doi:10.1007/s10615-009-0233-1

43.

Wilcox

Evenson

K. R.

Aragaki

Wassertheil-Smoller

Mouton

C. P.

Loevinger

B. L.

(2003). The effects of widowhood on physical and mental health, health behaviors, and health outcomes: The women’s health initiative. Health Psychology, 22, 513–522. doi:10.1037/0278-6133.22.5.513

44.

Wolchik

S. A.

Tein

J. Y.

Sandler

I. N.

Ayers

T. S.

(2008). Parentally bereaved children’s grief: Self-system beliefs as mediators of the relations between grief and stressors and caregiver–child relationship quality. Death Studies, 32, 597–620. doi:10.1080/07481180802215551