Abstract
Family values operate in both immediate (i.e., familial) and broader social contexts. This study used a mixed methods approach to examine both forms of family values in the Australian context. A convenience sample of 856 people completed a measure of family values about both their own values and their perceived values of one of their parents, and a measure of familism. Using a story completion approach, a majority of the sample also responded to three story stems focused on the perceived values held by fictional families. Quantitative findings identified relationships between participant and perceived parent values in terms of gender. Participants reported high levels of familism, predicted by religiosity, age, and being a parent. Qualitative findings suggested that some participants were mindful of discrimination faced by the fictional families, but many participants also provided deficit accounts.
Introduction
Values, typically understood as “guiding principles” held by individuals (Schwartz, 2012), shape the way we live our lives, they help us to make important decisions, and they determine the commitments we make. Our values are shaped by both our immediate and broader social contexts. Our immediate contexts are our families: both birth families and chosen families. Our broader social contexts are the political, ideological, and institutional forces that promote certain values over others. In this paper, we report on a study that sought to explore how people living in Australia view their own and their parent’s values, factors that are predictive of traditional family values, and how people account for the values of other families in the Australian community.
Key to understanding the values that people hold has been the work of Schwartz (e.g., 2012), who has outlined an evidence-based approach to understanding core values that are likely shared across many cultures. Schwartz has proposed 10 core value dimensions, namely, power (i.e., social status), achievement (i.e., personal success), hedonism (i.e., pleasure seeking), stimulation, universalism (i.e., concern for others and nature), benevolence, tradition (i.e., respect for customs), conformity (i.e., not violating social expectations), and security (i.e., stability). These 10 core values dimensions are encompassed by two binary groupings: openness to change versus conservation, and self-transcendence versus self-enhancement (Knafo-Noam et al., 2020).
In terms of families, researchers have used the core value dimensions developed by Schwartz (2012) to examine how values are transmitted across generations, and thus the degree to which values are similar between parents and children. Much of this research has focused on transmission between parents and their adolescent or young adult children, and has consistently found gender differences. Specifically, the strongest value similarities have been found between young women and their mothers, and the weakest between young men and their mothers (e.g., Barni et al., 2012; Hoellger et al., 2021).
Importantly, however, researchers have emphasised that values are not simply a process of transmission from one person (i.e., a parent) to another (i.e., a child). Rather, and as noted above, values are shaped not only in immediate contexts such as our families, but also in broader social contexts. Barni and colleagues (2012) have suggested that “value transmission similarity should be composed of a culturally stereotypic component, shared within the common social-cultural context of belonging, and a unique component, which could potentially be interpreted as that excess in value that the family either is or is not able to create and then is shared among its members” (p. 47). Research suggests that cultural differences—particularly when comparing individualistic and collectivist cultures—are likely to influence the degree to which values in immediate contexts are shaped in their transmission by broader cultural values (e.g., Prioste et al., 2017).
The transmission of family values in the context of collectivist cultures has been routinely examined through the concept of familism. Familism refers to a strong identification with the importance of family, and strong feelings of commitment to family members across generations (Losada et al., 2020). Research, primarily undertaken in the South American context, has consistently identified familism as playing a key role in the transmission of family values, and the role of values in individual wellbeing and harm prevention (e.g., Meija et al., 2021). However, research to date has assumed that familism is primarily a feature of collectivist cultures. As such, it has failed to explore whether or not familism is also equally evident in the context of cultures that are thought to be individualistic, as well as the extent to which familism in individualistic cultures (if it exists) affects the transmission of family values.
In the Australian context, which is the focus of the present paper, collectivism is certainly not absent, with a general agreement that Australia is a multicultural nation. However, despite this diversity, Australian family related policies are often premised on an assumption of the singular, individual, nuclear family (Uhlmann, 2013). Moreover, research on family values in Australia is scant and dated, and to date has not focused on values transmission or similarities. What we do know is that there are not only gender differences in values (with men holding more traditional values, such as the belief that men are the head of the household and women should work primarily in the home), but that there are generational gaps, such that younger people are likely to hold more progressive values (such as showing greater support for family diversity) (De Vaus, 1997). Whether such values are largely individualistic in nature, or encompass aspects of collectivist values (such as familism), has not to date been a focus in Australian research.
Given all of the above, the study reported in the present paper sought to examine family values in Australia using a mixed methods approach focused on both immediate and broader social contexts. As a multicultural nation, Australia offers a unique family values perspective that has international relevance. The research was guided by the work of Gillis (1996) on family values, who suggests that there are both families we live with, and families we live by. The former are our immediate family members, their values and practices. The latter are our imagined families: our ancestors and the values we believe they held, and the type of family we like to hope we belong to. Gillis suggests that families function as “temporal convoys,” propelling us collectively through time via a shared sense of values, and through a presumed relationship between the past and the present. The binaries of collectivist and individualist, immediate and social contexts, then, are encompassed by the role of our real and imaged families as proxies for imagined communities that themselves are proxies for an imagined national unity.
Rather than looking at individual family members and comparing their self-reported values, then, the study reported in this paper utilised one family member to report on both their own and one of their parent’s values. This individual approach thus shifts our focus away from an emphasis on direct values transmission and similarities, and towards a focus on how individual people imagine their families to be: do they perceive their parents like them or unlike them? More broadly, and using story completion methods, the study also looks at the perceived values held by other people: people potentially different in a number of ways to the participants. Story completion uses brief fictional stories and prompts to encourage participants to hypothesise about the imagined lives of people in the stories (Clarke et al., 2017). In so doing, it allows for an examination of broad culturally available narratives or stereotypes about an imagined community. When it comes to family values specifically, story completion allows for an examination of the types of attributions that people make about the values of families likely different to their own.
Research Questions
Drawing and extending on the research summarised briefly above, the study reported in this paper used a mixed methods design to explore perceptions of family values in a convenience sample of people living in Australia, focussing on both immediate and broader social contexts. The quantitative component of the study focused on immediate contexts, and sought to answer the following research questions: 1. Which of the 10 values identified by Schwartz (2012) demonstrate a relationship between participants and their perceptions of one of their parents? 2. Are perceived similarities in values differentiated by the gender of participants and the gender of their parent? 3. What are the predictors of familism in an Australian sample?
The qualitative component of the study, using a story completion task, sought to answer the following research question: 4. How do people living in Australia perceive the values of a diversity of families other than their own?
Focussing on adult participants and their views of their parents contributes to the relative lack of research on this population with, as Hoellger and colleagues (2021) note, much of the previous research focussing on younger people and their parents. Further, focussing on traditional understandings of family (i.e., familism) in the context of a notionally individualistic culture makes a significant contribution to how we understand the existence of familism across a diversity of contexts. Finally, the mixed methods approach allows the study to address both immediate and broader social contexts. As such, the mixed methods approach provides data that are complementary, but which speak to the topic of family values in differing ways and from differing vantage points.
Method
Participant Recruitment
Ethics approval was granted by the Flinders University Social and Behavioural Research Ethics Committe. Participants were recruited via paid advertisements placed through the Australian Family Diversity Twitter account, and the Australian Family Diversity Facebook page. Inclusion criteria were living in Australia and being aged 18 years or older. Recruitment opened in February 2021 and continued for 6 weeks, and was closed once the minimum sample size (of complete responses, see below) had been exceeded (based on the population size of Australia, a 95% confidence level, and a 4% margin of error). Participants were asked to give consent to participation, and were advised that they could withdraw at any time prior to submitting their completed responses. No financial incentive was offered to participants.
In the time the survey was open, 1043 people started the survey. Of these, 49 people completed none of the survey questions. 138 people completed the demographic questions but did not complete enough of the measures to be included in the final sample. There were no statistical differences in terms of demographics between these 138 partially completed responses and the final sample of 856 participants. Given the wide distribution of the call for participants via paid advertisements on two social media platforms, it is not possible to provide an estimated response rate.
Materials
Participants completed a survey designed by the authors, hosted on Qualtrics, an online survey hosting platform. The survey opened with an information screen, detailing the purpose of the study, outlining inclusion criteria, and then asked participants to consent to participation. Participants were then presented with a series of demographic questions: 1) gender, 2) sexuality, 3) age, 4) whether or not they were in an intimate relationship, 5) whether they were Aboriginal, Torres Strait Islander, both, or neither, 5) if they were born in Australia, 6) if English was their first language, 7) their degree of religiosity (where 1 = not at all, 2 = somewhat, 3 = quite a bit, 4 = a lot), 8) their political views (liberal, centrist, or conservative), 9) whether or not they were a parent, 10) what State or Territory they lived in, 11) their highest educational qualification, 12) their average household annual income, and 13) their current employment status.
Utilising a story completion approach, participants were then presented with three story stems, and identical prompts for each story stem. Derived from projective testing, story completion involves giving participants the first part of a fictional story (or “stem”), and asking them to write more about the story (Clarke et al., 2017). Participants are often also given prompts to help encourage their writing. By asking people to imagine how the rest of a story or scenario might proceed, story completion provides researchers with access to the beliefs, stereotypes, and ways of thinking that participants hold about a particular situation, group of people, or cultural phenomenon.
The story stems used in the present study are included in the results below when introducing the results from the analysis of responses to each stem. The prompts were the same for each stem: “Thinking about the brief fictional story of [parent/s], please consider the following prompts and share your thoughts about this family. How do you imagine the home life of [parent/s and children]? What types of values do you think [parent] conveys to their children? What other people might be related to or involved with their family?” Having responded to the three story stems, participants then completed two measures.
Schwartz Values Survey
As outlined in the introduction to this paper, the Schwartz Values Survey (SVS, Schwartz, 2012) asks participants to respond to a series of prompts, asking them to rate the degree to which they are like the fictional person in each prompt. The SVS is comprised of 21 items, encompassing the values, of security, conformity, tradition, benevolence, universalism, self-direction, stimulation, hedonism, achievement, and power. Each value is comprised of two items, other than universalism, which is comprised of three. Examples of the fictional-person based questions include “It’s very important to them to help the people around them. They want to care for other people” (benevolence) and “Thinking up new ideas and being creative is important to them. They like to do things in their own original way” (self-direction). The SVS typically uses he or her for the fictional prompts, but for the present study the gender neutral “they” was used. Participants responded to each prompt on a seven-point Likert scale, from “1 = not at all like me” to “7 = very much like me.” Given each value is separate to a degree they are not summed into an overall value score, and nor can an overall alpha value be calculated. Participants were first asked to complete the SVS about themselves, and then to choose either their mother or father, and to complete the SVS a second time, where they provided responses that reflected what they believed were the values of their chosen parent.
Revised Familism Scale
The Revised Familism Scale (RFS) asks participants to indicate the degree to which they agree with a series of statements that focus on traditional family values reflective of an emphasis on assisting family members, respecting family members, and being close to one’s family members. Example items include “I would help with my means if a relative told me that they were in financial difficulty,” “A person should respect their elders regardless of their differences in views,” and “A person should live near their parents and spend time with them on a regular basis.” Participants responded to each prompt on a five-point Likert scale, from “1 = strongly disagree” to “5 = strongly agree.” Previous research has found strong alpha values for the RFS, a = 0.85, and the same was true in the present study, a = 0.93.
Data Analysis
Upon closure of the survey, data were exported from SurveyMonkey into SPSS 25.0. The data were then cleaned in the following ways. Statistical tests were run to identify if there were any differences between completers and non-completers, and non-completers were then removed from the file. Item means were calculated for the RFS. Given that while the SVS is not a singular measure, it is conceptually comprised of four areas (Openness to change, Conservation, Self-transcendence, and Self-enhancement), a factor analysis was conducted using varimax rotation, identifying a four-factor solution, with the factor contributing the greatest amount of variance being comprised of the three values of conformity, tradition, and security (constituting the Conservation value area). This was true in terms of responses pertaining to both self (26.84% of variance) and key parental figure (25.72% of variance). Given the relatively low contribution of the other values to explaining variance, only the combined Conservation values are used in the results below, reported as Self-Conservation Value and Parent Conservation Value.
Dyadic correlations (Kenny et al., 2006) were then performed to test the relationship between Self and Parent Conservation Values, and the split data function in SPSS was used to differentiate these by participant gender and parent of key parental figure. Two-way MANOVAs were used to explore interactions and simple main effects for participant gender, gender of key parental figure, and both Self and Parent Conservation Value. T-tests, ANOVAs and correlations were used to identify any statistically significant relationships between demographic variables and the RFS, and statistically significant variables were then included in a regression.
Of the 856 participants who completed all measures, 625 provided responses to story stems one and two, and 615 provided responses to story stem three. Responses to each story stem were analysed via conventional content analysis. As outlined by Hsieh and Shannon (2005), conventional content analysis involves (a) repeated readings of the data corpus, (b) developing codes by highlighting key words that capture frequently occurring concepts, (c) reducing codes in order to minimise overlaps, (d) examining codes for patterned responses, in order to group codes into categories, and (e) examining categories to determine whether or not they accurately reflect the data corpus. As noted by Hsieh and Shannon, a limitation of conventional content analysis is that it does not utilise member checking or interrater reliability. Given the survey was anonymous, member checking was moot for this study. In terms of interrater reliability, Hsieh and Shannon note that all analyses are subjective, and thus should be viewed as offering one interpretation derived by the researcher. That said, given the brevity of responses provided by participants (most participants provided only a brief one-two sentence response to each story stem), it is reasonable to suggest that the content analysis undertaken captures the semantic meaning of each response.
In terms of the specific undertakings of the content analysis, the first author read through each open-ended response, making note of potential codes from each response. This initial coding was used to generate a list of all possible codes. All responses were then read again by the first author, allocating one code to each response. Where more than one potential code was possible for a response, the code allocated represented the most salient code within the response, based on the greatest amount of text written (i.e., if a response included two possible codes, the length of text for a given code was used to determine if it was likely the most salient code). 92% of responses contained only one possible code. For the other 8% of responses, the second author reviewed the single codes allocated to responses that contained multiple possible codes, and was in agreement in all instances with the single code allocated by the first author. The final categories for each story stem reported in the results below are different for each story stem given they are based on codes developed from the data, rather than a pre-existing set of codes used to uniformly examine responses to each story stem.
Results
Participants
Demographics of Sample (n = 856).
Correlations Between Matched Values for Self and Parent and Between Familism and All Values.
p = .01.
Quantitative Findings
Family Continuity
Dyadic Correlations Comparing Self and Parent Conservation Value × Participant Gender and Gender of Key Parental Figure.
Means for Self and Parent Conservation Values × Gender of Participant and Gender of Chosen Parent.
Familism
Participants reported on average relatively high levels of familism (M = 4.03, SD = 0.60). Statistically significant relationships were identified between Familism and three of the demographic variables. There was a weak positive correlation between age and familism, r = .256, p = .001, and religion and familism, r = .256, p = .001. Participants who were parents reported higher levels of familism (M = 4.09, SD = 0.57) than participants who were not parents (M = 3.85, SD = 0.65), t = 4.40, p = .001. Given these relationships a simultaneous multiple regression was conducted utilising Familism as the outcome variable, and age, religiosity, and being a parent (yes = 1, 2 = no) as predictor variables. The model predicted 39.7% of the variance (R 2 = .397, F (4, 856) = 16.67, p = .001). Of the predictor variables, religion contributed the largest proportion of unique variance (β = .240, p = .001), followed by whether or not participants were parents (β = −.222, p = .05), and age (β = .205, p = .01).
Qualitative Findings
Overview of Content Analysis Categories for Each Story Stem.
Story Stem 1: Sarah and Mary and their Children
For this first story stem, participants were asked to reflect on the prompt questions having read the following story stem: “Sarah and Mary are the parents of two children, Mark and Kate. Sarah works full-time as a site engineer and Mary works part time as a teacher. Sarah and Mary and their children live in an inner-city suburb. Sarah and Mary are married to one another.” Five categories were developed from participant responses: 1) The family would have a focus on equality, 2) Lesbian-parent families are just like heterosexual-parent families, 3) Lesbian-parent families offer unique benefits to children, 4) There is something wrong with lesbian-parent families, and 5) Lesbian mothers need to provide male role models to their children.
The first two categories may be construed as liberal accounts of lesbian-parent families, echoing the extensive research of Clarke (2002), which has found that heterosexual participants often frame lesbian-parent families as inclusive (and thus a positive place to raise children) and as being just like heterosexual-parent families (and thus not deserving of discrimination). Examples of an emphasis on equality appear in participant responses such as “The values they would have are love is love, tolerance and accepting of difference” and “I would imagine they are very liberal people with values closely aligned with the green party.” While these types of accounts are positive, they do not necessarily pair an account of inclusivity with an account of why lesbian-parent families might be more inclusive (i.e., experiences of marginalisation). Examples of the second liberal category include “I think Sarah and Mary would be just like any heterosexual couple” and “their values wouldn’t be any different to the conventional marriage”. It can be argued that these types of accounts emphasise a homonormative understanding of lesbian-parent families, homonormativity referring to the privileging of normative understandings of sexuality diverse people, including with a focus on monogamous coupledom and parenthood (Duggan, 2012).
Different to the two liberal accounts, accounts that focused on the unique benefits of lesbian-parent families did not make direct comparisons to heterosexual-parent families. Instead, they often paired recognition of discrimination with recognition of how this might result in specific forms of parenting. Examples include: “They would have values that include being resilient against possible judgements outside the family, which would make them a strong family” and “Having probably experienced a lack of acceptance, the children will probably be raised to accept diversity. What makes these types of accounts distinct from both forms of liberal accounts is that they do not simply praise lesbian-parent families for being accepting of diversity, for example, but rather situate such potential strengths in a broader socialcontext.
The final two categories represent negative views about lesbian-parent families. The first was explicitly negative, and included responses such as “The children would not be taught the right perspective of the world” and “The family is immoral, confusing for children, and their values would be man hating.” These types of accounts draw on both pathologizing and stereotyped accounts of lesbian-parent families. Less obviously negative were accounts that emphasised the need for male role models in lesbian-parent families. As Clarke and Kitzinger (2005) has argued, however, such accounts nonetheless emphasise a deficit account of lesbian-parent families, which are seen as “lacking” men, and which is presumed to have significant implications for children (and specifically for boys). Examples of this type of deficit account included “I hope that men are a valuable part of the extended family and that masculinity isn’t discouraged for the male child” and “Sarah and Mary actively seek male role models for their children.” While not as clearly negative as accounts located within the fourth category, accounts in this final category nonetheless positioned lesbian-parent families as inadequate.
Story Stem 2: Farid and Fatima and their Children
For this second story stem, participants were asked to reflect on the prompt questions having read the following story stem: “Farid and Fatima are the parents of three children, Mubarak, Alia, and Farida. Farid works full-time in correctional services. Before having children Fatima worked in childcare, but now works solely in the home caring for her family. All of the family were born in Australia, however Farid and Fatima’s parents were born overseas. Farid and Fatima and their children attend religious services regularly.” Five categories were developed from participant responses: 1) Family seen as conservative, 2) Family seen as enacting inequality through gender roles, 3) Food as central to the family, 4) The family’s experiences of discrimination, and 5) Strengths presumed to be inherent to the family.
The first two categories emphasised negative accounts of the family. Many participants presumed that the family were Muslim, and endorsed values for the family that reflected their interpretation of Islam. The first instance of this pertained to the assumption that the family would be conservative, with examples being “Farid is the head of the family, probably very conservative and traditional” and “Judging by religious beliefs probably a conservative household with traditional values.” In these types of examples, “conservative” and “traditional” are collapsed together, precluding the possibility that the family might hold beliefs aligned with a particular religion without enacting those beliefs in conservative ways. The second category focused on a specific form of conservatism, namely, normative gender roles. Examples include: “The religious teachings are a concern as Islam’s teachings can be misogynistic and not inclusive of gays,” and “Children are loved but male child is favoured more. If any of the kids are gay, they would probably keep it a secret so their parents don’t abandon them.” Here only conservative accounts of religious families are emphasised, ignoring other contemporary accounts of religious families.
In many ways paired with the two more obviously negative accounts, the third category emphasised food as central to the family, endorsing what could be framed as an exoticising account of what is read as cultural diversity, one in which the celebration of the cultural practices of certain groups is taken as a form of liberal inclusion (May, 1996). As such, while the accounts included within this category were positive, they nonetheless draw on stereotypes or culturally limiting ways of providing a positive account. Examples include: “The family life is full of delicious food,” “I imagine they would eat fabulous food” and “They probably have lots of delicious food to eat.” Given the prompt questions asked, the emphasis on food here is reductive.
Similar to the previous story stem, some participants provided responses that were mindful of the discrimination potentially faced by the family. Examples include “I imagine they might be Muslim, and therefore despite being Australian-born, have to deal with a lot of racism and anti-Muslim sentiment,” and “They may face racism in their wider community.” While these types of accounts were productive in that they were mindful of the broader social context, they did not explicitly link this context to the values of the family. The final category, however, emphasised the strengths of the family in the face of broader (potentially discriminatory) social contexts. Examples of this final category include: “Being religious could take off some of the stress of living in hostile places as they have the backing of their religious community” and “Religious education would impart the children with good values that would enable to be positive members of a society that may not accept them.”
Story Stem 3: Mark, Jesse, and Arnold
For the final story stem, participants were asked to reflect on the prompt questions having read the following story stem: “Mark is the father of Jesse. They live together with their dog Arnold. Mark works when Jesse is at school. Mark is a plumber. Mark, Jesse and Arnold live in a small country town.” Four categories were developed from participant responses: 1) The family values hard work, 2) The family is probably sad and lonely, 3) The family is missing a woman’s influence, and 4) Their homelife is probably re/laxed. Despite the mention of Arnold the dog in the story stem, very few participants oriented to this aspect of the story stem, though those who did typically mentioned that any family living with a dog must have positive values.
In terms of the first category, a combination of Mark being a plumber and Mark being a sole parent translated for many participants into the assumption that he and Jesse valued hard work. Examples of this include “I think that loyalty and hard work are values conveyed by Mark” and “Mark would teach Jesse the importance of hard work and a strong work ethic.” Rarely did participants unpack why hard work would be a core value of the family, but those who did emphasised that being a sole parent alongside undertaking paid work was emblematic of being a hard worker. Likely this is at least in part a gendered phenomenon: men as sole or primary parents are typically framed as hard workers, in comparison to women in similar roles who are typically framed as having the automatic capacity to care for children and work outside the home (Hunter & Riggs, 2019).
Despite framing the family as hard working, some participants nonetheless framed the family as being sad, and Mark in particular as being lonely. Examples of sadness primarily pertained to Jesse, and included: “Jesse’s Mum is not around so that seems like a possible point of sadness” and “Some sadness about not having a biological mother around.” Here the heteronormative assumption is that Jesse does have a mother, she is just not around, an assumption that presents a limited framing of who the family might be. In terms of Mark, participants emphasised his “lack” of a partner as producing loneliness, such as in the following examples: “Mark is often lonely because he doesn’t have a partner, so that may have some depression” and “Dad probably feels lonely back lacks confidence to date anyone.” Examples such as this endorse an account of singlism, referring to the normative assumption that all people should want to be in an intimate relationship (DePaulo & Morris, 2006). Not having a partner means Mark is positioned as potentially depressed, and as not confident enough (an excuse for not having a partner, rather than, for example, conjecturing that he doesn’t want a partner).
In many ways building on the second category, the third category emphasised a deficit account of the family, such that there being no adult woman in the house was viewed by some participants as detrimental to Jesse. Some participants presumed that Jesse was a boy, and made comments such as “I hope the child has some female influences in his life”, while other participants did not presume Jesse’s gender, but nonetheless made similar comments: “Depending on the age and gender of Jesse, things may become harder is there is no mother figure.” Some participants made direct comparisons to the story stem with Sarah and Mary, noting “the importance of having women as well as men in Jesse’s life,” and others presumed the family would have negative views of women, such as “Mark may be sexist and convey negative thoughts about women to Jesse.” In these examples, not only is the family positioned as being in deficit, but Jesse is also positioned as being at risk.
Finally, and again relying upon a particularly gendered account of the family, participants frequently indicated that family life for Mark and Jesse would be relaxed, if not lax. Examples include “Mark would be a bit lax in his discipline and meals would be anything they wanted,” “I imagine a bachelor pad and not very healthy meals,” and “I feel their home would be a bit rough around the edges.” While not specifically referencing gender in these examples, that they are about male-parent household potentially speaks to views about men’s capacity for household duties and childcare, views that may not be entirely out of step with the experiences of some families, but certainly oversimplifies the lives of all households with single male parents (Hunter & Riggs, 2019).
Discussion
The findings reported above provide answers to the four research questions. In terms of research question one, the findings suggest that conservation values, comprised of conformity, tradition, and continuity, were those most strongly related between participant ratings of themselves and a key parental figure. Given the high average familism values in the sample, it is understandable that values related to family conservation were especially privileged. Further, given the average age of the sample and predominance of women in the sample, this may further explain the emphasis on family conservation. Research suggests that older women may be especially committed to what is referred to as kin keeping: preserving close relationships across generations (Attar-Schwartz et al., 2009).
In terms of the second research question, the findings contradict previous research to a degree. Stronger correlations were found for men and women’s views on their fathers, though the interactions between parent and participant gender give nuance to this finding. This difference to previous research might be a product of the fact that participants in the present study chose one parent, whereas in previous research, the values of both parents have been a focus. It might well be that the parent chosen by participants was the parent they most identified with. Additionally, differences from previous research might again reflect the average age of the sample, and the predominance of women in the sample. Previous research on value transmission suggests that women are taught to conform to the views of their parents more than men are, and to their fathers in particular (e.g., Barni et al., 2012). Unique to the present study was the inclusion of nonbinary participants. Nonbinary participants did not report a strong relationship between their own values and those perceived of one of their parents is not surprising, given what we know of the negative experiences of many nonbinary people with their families (Fuller & Riggs, 2018).
The findings of high levels of familism in the sample, while not surprising given the sample, adds weight to the idea that collectivist views about family are most certainly not absent in notionally individualistic societies. That religiosity, being a parent, and being older were predictors of familism affirm the suggestion in previous research that values in the Australian context may display generational effects (De Vaus, 1997). This suggests the importance of follow-up research focussing on younger cohorts of people who are not parents, who may (or may not) hold differing views about familism. That English speaking background and country of birth were not significant predictors of familism is also of note, and warrants further investigation in the Australian context.
Finally, in terms of research question four, the analysis of the story stems provides significant insights as to the broader Australian context in which values about families circulate. Certainly, for all of the story stems, some participants provided positive accounts of the fictional characters, and for the first two stems, some participants were mindful of the discrimination they might face. But for all three story stems, some participants also produced highly normative, even if at times liberal, accounts of the fictional characters, and this included negative accounts that relied on stereotypes. Most interesting to note is that while family conservation and familism were rated highly by participants, this was not uniformly applied to other families who may be different to the participants’ own. While conserving family traditions was valuable for participants when it came to their own families, the traditions and values of other families were not uniformly seen as worthy of preservation, as evident in the less than positive responses provided by some participants. Future research with a more diverse sample would benefit from examining in closer detail whether participant responses to story stems about family values are differentiated by participant demographics.
As already indicated, the study reported in this paper is limited by the specific skew of the participant demographics. Whether or not relationships between values might be stronger or differently distributed in younger, more diverse samples of people living in Australia is a topic for further research. While the aim of the study was to investigate participant perceptions of their own and one of their parent’s values, further research would benefit from matched samples of adult children and their parents in the Australian context, so as to examine if there are differences between perceptions of parent values, and the values that parents report. Finally, while the study used story completion as a way to tap into social values about families, future research would benefit from including a measure of adherence to socially prescribed values (such as gender ideology), to examine whether the social context mediates the immediate (i.e., familial) context.
Returning to the work of Gillis (1996), the study reported in this paper gives weight to the idea that the families we live by are just as important as the families we live with. Given this was an adult sample reporting on their views about their parents, it is reasonable to suggest that the quantitative component taps into a sense of families as temporal convoys, with participants holding onto views about their parents that they perceive as aligning with their own. That this was most true with regard to values about family conservation is both logical, but also speaks to the ongoing importance of family connections, at least in some cohorts in the Australian context. That such values do not easily map across to full respect for other families suggests a gap, however, between what is important for one’s own family, and what one may see as important for other people’s families. In other words, conserving the family that one lives by does not automatically translate into respect for the families that other people live by. This suggests the importance of further research that unpacks what factors, values, and worldviews create this gap. An easy answer would be to emphasise ingroup and outgroup differences, but needed is research that aims to identify the specific values and beliefs that enforce or indeed idealise differences between families.
In conclusion, in the specific convenience sample included in the study reported in this paper, family conservation was important for participant’s own families, but this same importance was not necessarily afforded to other families. This has implications for social cohesion in Australia, specifically with regard to respect for a diversity of family values. Clearly the topic of family values in the Australia context requires ongoing attention, with mixed methods approaches offering a means to exploring both the immediate and broader social contexts through which values are enacted, transmitted, and potentially changed.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study reported in this paper was supported by an Australian Research Council Future Fellowship, FT130100087, awarded to the first author.
