The social life of non-parents: A systematic review of social networks and social support among adults without children

Declining fertility rates

Fertility rates have been declining, with fewer people becoming parents and more people choosing not to have children. Over the last 70 years, fertility rates have decreased by 50%, with an average of 5 children per family in 1952, compared to 2.4 in 2020 (World Economic Forum, 2022) ¹ . This decline is attributed to multiple factors, including economic instability (Sobotka et al., 2011), increasing costs of parenting (McDonald, 2002), increased educational attainment, family planning and access to contraception for women (Liu & Raftery, 2020) and shifting views on parenthood and motivations (Kearney et al., 2022; Mitchell & Gray, 2007). Reasons for foregoing parenthood include–but are not limited to–internal motivations such as disinterest in children (Park, 2005), opportunities for freedom and independence (Höglund & Hildingsson, 2023), maintaining close relationships (Gillespie, 2003), and broader concerns regarding overconsumption, overpopulation and climate change (Helm et al., 2021; Nakkerud, 2024) (see also Agrillo & Nelini, 2008).

Demographic trends also demonstrate an increase in the number of individuals not expecting or intending to have children (e.g., Berrington, 2017), and a trend of delaying parenthood, with women increasingly having children in their thirties rather than their twenties (ONS, 2024). Therefore, understanding the social dynamics of non-parenthood has become an increasingly relevant area of research. Yet, in empirical work, non-parents are often treated as a single, undifferentiated group, which risks obscuring the heterogeneity within this population (Dykstra & Hagestad, 2007). To improve our knowledge of this population and choices around parenthood, it is important to more fully differentiate among non-parents. The current review thus aims to characterize potential differences in social networks and support systems between parents and non-parents, including distinct non-parent groups.

Stigma of foregoing parenthood

Across many cultural contexts, including contemporary Western societies (McCutcheon, 2020), parenthood is still conceived of as natural and fixed, and pronatalist views prevail. Compared to parents, non-parents are often perceived more negatively: they are viewed as less warm, less fulfilled, and having less positive marital relationships (Ashburn-Nardo, 2017; Bays, 2017; Çopur & Koropeckyj-Cox, 2010; Ekelund & Ask, 2021). Similar negative perceptions and social exclusion have been found in other regions of the world, such as India (Babu, 2024), Indonesia (Rismarini & Adira, 2024) and Japan (Tanaka & Lowry, 2018). Non-parents are aware of these negative views and feel targeted for not having children (Doyle et al., 2013; Park, 2002; Turnbull et al., 2016).

Importantly, not all non-parents are perceived equally. Childfree individuals–those who choose not to have children–face moral outrage (Ashburn-Nardo, 2017) and are perceived as deviant (Blackstone & Stewart, 2012). Childfree individuals tend to be evaluated more negatively than those who are childless or delaying parenthood and can be viewed as an outgroup by those who (want to) have children (Iverson et al., 2020; Koropeckyj-Cox et al., 2007, 2018; Watling Neal & Neal, 2021). Thus, the desire or intention to have children may act as a buffer against becoming the target of others’ negative attitudes. Childbearing intentions can be shifted by perceived social pressure (Albertini & Brini, 2021), which may be stronger in some regions, such as Central and Eastern Europe, where parenthood is often regarded as even more of an ideal, a societal obligation, and an important component of personal fulfilment (Miettinen & Szalma, 2014; Sobotka & Testa, 2008).

Social network and support system theories

Social support can help manage the impact of stigma. Defined as “the provision of assistance or comfort to others, typically to help them cope with biological, psychological, and social stressors” (American Psychological Association, 2018), social support is intrinsically connected to the relationships people have with others, i.e. their social network. Theories of social networks and support systems highlight the central role of children in providing support, suggesting that their absence may shape both the structure and function of non-parents’ support systems in distinct ways. In this review, we apply the Hierarchical Compensatory Model and Task-Specific Support Theory to help interpret how individuals without children may rely on alternative sources of support.

The Hierarchical Compensatory Model suggests a well-ordered hierarchical preference for support, with spouses and children typically being the primary providers (Penning, 1990). When the primary or preferred support providers are unavailable (e.g. no spouse or children), relationships lower in the support hierarchy, such as extended family, friends, and neighbours, are called upon (Dykstra, 2015); components of support networks are interchangeable (Penning, 1990). For parents, adult children are suggested to serve as key support providers. In contrast, non-parents may rely more on relationships lower in the support hierarchy, such as friends. These relationships, however, may be less stable, as peer-based support is often more susceptible to deterioration than parent-child ties in unbalanced exchanges of support (Dykstra, 2015). Complementarily, Task-Specific Support Theory proposes that relationships tend to be specialized in their support provisions (Dykstra, 2015), shaped by network characteristics such as proximity, role, and shared experience (Penning, 1990). Family members, including children, often provide instrumental help (e.g. transportation, shopping) and foster a sense of belonging, while friends are better suited to emotional support (e.g. providing comfort) and tasks based on similarities of interests or experience (e.g. giving advice; Dykstra, 2015; Penning, 1990; Teerawichitchainan et al., 2024). Therefore, non-parents may develop more diverse networks and rely on other family members, such as siblings, to fulfill instrumental support needs, reflecting structural and functional differences in the social networks of parents and non-parents.

While parenthood is presumed to integrate individuals into social networks and provide a sense of purpose in life (Hansen et al., 2009), empirical evidence for its psychosocial advantages is limited (but see Hudde & Jacob, 2025, on parents across Europe reporting higher meaning in life). Many individuals choose to be childfree for the advantages it provides, including better relationships with others (Gillespie, 2003; Magennis, 2024). It has been suggested that research casts a one-sided negative view of non-parents as isolated and outright lacking in support (Klaus & Schnettler, 2016), overlooking compensatory ties, including the increasing number of online communities that provide advice and support (mutually supporting reference groups; Basten, 2009).

Review aims and objectives

In line with the aims of this review, the following research question was formulated:

(1) What is known about social networks and social support of non-parents, including not-yet parents, childless, and childfree individuals?

More specific questions informed the focus of the review, including:

(1a) How are social networks and social support for non-parents different from parents?

Accordingly, the key objectives of this review were:

(1) Identify social network characteristics (e.g. network size) for non-parents and the amount and forms of support they receive,

Method

This review followed the guidelines of the Preferred Reporting Items for Systematic Review and Meta-Analysis guidelines (PRISMA; Page et al., 2021). Further details regarding the method can be found in the pre-registered protocol with the Open Science Framework (https://doi.org/10.17605/OSF.IO/SFUE6).

Information sources and search strategy

The primary source of literature was a structured electronic search using the following databases: Medline, APA PsycINFO, Web of Science Collection, International Bibliography of the Social Sciences, Sociological Abstracts, and Scopus. Grey literature included the first 1000 articles on Google Scholar, due to viewing limitations (Haddaway et al., 2015), ProQuest Dissertations and Theses and searches of websites of relevant organisations, supplemented by hand searches of the reference lists from prior reviews and included studies. Search terms and strategy were developed in collaboration with a subject librarian with specialist knowledge (see Table S1-7) and in accordance with the Peer Review of Electronic Search Strategies (PRESS) checklist (McGowan et al., 2016). Table S8 shows how the Sample, Phenomenon of Interest, Design, Evaluation, Research type (SPIDER) tool was applied to the research question and eligibility criteria.

Inclusion and exclusion criteria

No limitation related to the years of publication was applied. The inclusion and exclusion criteria were identified a priori. Studies must have included a direct comparison of social network characteristics and/or social support between either parents versus non-parents or different groups of non-parents. Social network characteristics broadly included structural and quality aspects of individuals’ relationships, such as network size, composition, frequency of contact, closeness, reciprocity. Social support involved any measure assessing perceived or received support or assistance, including emotional, informational, and instrumental forms. Full inclusion and exclusion criteria can be found in the pre-registered protocol and document listing amendments. Any studies not meeting all the relevant inclusion criteria were excluded.

Study selection

Study selection involved three stages. Studies were first screened for relevance by title and abstract, followed by full-text review against the inclusion and exclusion criteria. Then, lead authors of all included studies were contacted to identify any additional relevant studies meeting the inclusion criteria, and the reference lists of each included study were searched. At all stages except regarding grey literature, the second author and third author screened 50% of articles (25% each). There was high agreement regarding inclusion/exclusion amongst the reviewers, with only 17 conflicts recorded out of 339 articles assessed at full text screening (94.99% agreement).

Quality appraisal and risk of bias

The quality of each study and risk of bias were assessed by the first author using the Joanna Briggs Institute (JBI) Appraisal Tools checklists (e.g. Checklist for Analytical Cross-Sectional Studies; JBI, 2020; available online). These tools evaluate methodological rigor across several domains, such as clearly defined inclusion criteria, appropriate measurement of outcomes, and adequate control of confounding factors. However, they do not give extra consideration to sample representativeness or to the robust sampling designs common in large-scale national surveys. Each question was answered “Yes,” “No,” “Not applicable,” or “Unclear,” in line with JBI guidance and the overall score reflected the proportion of criteria met. Scores were interpreted as high (≥70%), moderate (50–69%), or low (<50%) quality, guided by recent systematic reviews (Alfeo et al., 2024; Kachabian et al., 2024; Kundu et al., 2024) and an overview by George et al. (2014). Of note, low scores often reflected gaps in reporting or lack of detailed information on key variables (e.g. descriptions of parent/non-parent status in some studies) rather than weaknesses in study design. Studies were not excluded based on scores/overall appraisal from the JBI checklists–all eligible studies were included.

Data management and extraction

Rayyan was used to manage all levels of screening and to record the reasons for excluding articles at each level. Relevant quantitative data for the findings section was extracted by reviewing each study’s reported descriptives, such as the mean, standard deviation, and sample sizes. A summary of findings from included studies can be found in Table S9 and quantitative data for three aspects of the review in Tables S10, 11 and 12.

Analysis

Narrative synthesis is utilised in this review using the guidance by Popay et al. (2006). Findings from included studies are synthesised into themes, and together with the patterns found across the whole dataset (e.g. similarities and differences) form the narrative synthesis, allowing for interpretation beyond the primary findings of individual papers. Additionally, where deemed appropriate, descriptives provided for parent and non-parent groups from the included studies were used to calculate inverse-variance weighted average effect sizes for three key aspects: network size, received social support, and satisfaction with network/support.

Quality

The majority of articles had detailed information regarding inclusion criteria, study subjects and setting, and measurement of outcomes. Lack of detail in the reporting was most often a result of using secondary data. Some studies lacked detailed analysis (e.g. significance testing) or reporting of findings for data that was directly comparable (e.g. means for parents and non-parents) as these results were largely used for descriptive purposes rather than the main analysis. JBI appraisal scores ranged from 25% to 100%, with a mean score of 67.42%, indicating a medium quality (see Table S9).

Study characteristics

Demographic differences between parents and non-parents

There were some notable demographic differences between parents and non-parents. Non-parents were much more likely to live alone (Edgar, 1989; Hill et al., 2009; Kivett & Learner, 1980; Vicente & Guadalupe, 2024) or live without a partner/spouse (Albertini & Kohli, 2009; Graham, 2018) compared to parents. Non-parents were also more likely to live with siblings, friends, or in a care facility or nursing home (Boerner et al., 2016; Kivett & Learner, 1980). Fewer non-parents were in a relationship or married compared to parents (Balbo & Mills, 2011; Graham, 2018; Hsieh & Zhang, 2021; Langlois, 2003; McNamee & James, 2012; Vicente & Guadalupe, 2024) and had shorter current relationship durations (Chatterjee, 2016; DeMino, 2004). There were no clear differences in education, employment or wealth, and limited comparisons between different groups of non-parents.

Network characteristics

Received social support

Almost half (37; 46.25%) of all included studies included a general measure of perceived or received support, sources of support and proportion of informal or formal/paid help. Here we provide a brief overview.

Earlier studies demonstrated that non-parents often rely more on a broader range of non-familial relationships, including friends, neighbours, and siblings, for support, especially in emergencies (Hogan & Eggebeen, 1995; Johnson & Catalano, 1981). The likelihood of receiving support from outside the household for parents varies with age and life stages, such as childrearing and health deterioration, while it remains relatively stable for non-parents, with an increase in later life for non-parent women (Albertini & Mencarini, 2014).

In more recent research, parents were shown to perceive and receive greater social support than non-parents, with larger support networks. For instance, Klaus and Schnettler (2016) found that non-parents perceive a significantly smaller number of potential supporters than parents, and receive significantly less actual support. Harwood-Gross et al. (2023) also found that parents reported significantly greater perceived social support than non-parents.

More recent studies have placed greater emphasis on the distinction between informal and formal support. Older non-parents are more likely to receive formal support, such as professional care, compared to parents who receive more informal support, predominantly from adult children (Albertini & Kohli, 2009; Deindl & Brandt, 2017). Across 12 European countries, Deindl and Brandt (2017) found that older non-parents more often relied on either formal care alone or a combination of formal and informal care. In countries with greater social service availability, such as those in Northern Europe, particularly Scandinavia, the use of formal and combined support within households was more likely compared to Southern and Eastern European countries, where public support for those in need is less comprehensive. Consequently, older non-parents in lower-service countries may face a higher risk of formal care shortages.

In general, parents typically receive more social support through larger informal networks, while non-parents rely more on formal support and non-familial relationships, with more recent work incorporating cross-country comparisons. For 19 out of 37 studies, descriptive data on social support in parent and non-parents groups was available (Table S11). Again, we mostly found small to medium effect sizes, ranging from −0.009 to 0.84 and amounting to an inversely variance-weighted average Hedges g of 0.25, suggesting a rather small effect with non-parents perceiving less support than parents.

Support provision

Seven studies (8.75%) examined the support parents and non-parents provide to others. The earliest study by Wagner et al. (1999) found that parents aged 70 and over provided more help to their network than non-parents, though the difference was not statistically significant. More recent research, using large cross-national datasets like the Survey of Health, Ageing and Retirement in Europe (SHARE), provides clearer insights. Albertini and Kohli (2009) found among adults aged 50 and older, parents were significantly more likely to give financial support to others compared to non-parents (28.8% vs 15.5%), the majority of which (89%) was given to descendants, while non-parents were more inclined to support non-relatives (33%). A similar pattern was observed in later work by Albertini and Kohli (2017) who observed that generally, non-parents were less likely to give economic support compared to parents (17.2% vs 32.9%) although they gave similar amounts as parents who had minimal or no contact with their children (17.2%). There were less clear differences between parents and non-parents regarding the provision of social support, except that non-parent women spent significantly less time providing social support to others than parent women. Furthermore, Albertini and Mencarini (2014) observed that non-parents were slightly less likely to support others outside the household. The largest differences were observed among men, with this support declining earlier with age for non-parents. Conversely, Hurd (2009) noted that individuals without children, especially those with higher income, were more likely to provide financial assistance to parents, in-laws, friends, or relatives. While differences between non-parents and parents tended to be smaller for couples than for single individuals, in some income/wealth quartiles, the propensity among non-parents to provide financial support was twice as high as parents.

Regarding emotional and practical (e.g., household chores, transportation) support, Koller (2008) reported that parents provided significantly more practical help to family, not including childcare, than non-parents and significantly more emotional support. Comparatively, non-parents provided significantly more practical help to friends than parents and significantly more emotional support. Pesando (2019) found that non-parents in several countries were more likely to offer emotional and practical support to their elderly parents (upward support), with significant differences in Bulgaria, Georgia, Poland and France. In general, upward support was notably low in the Netherlands and Sweden, representing Northern European countries in the study characterised by publicly funded, universal welfare programmes. No substantial differences were observed in the overall likelihood of providing upward support between Eastern and Western European countries. Using logistic regression, the authors further conclude that across all 11 countries, non-parents were significantly more likely to provide support to their elderly parents compared to parents. Non-parent men were estimated to be 30–40% more likely to provide upward support compared to their parent counterparts.

Overall, parents tend to provide more support compared to non-parents. However, non-parents often give more support to non-relatives and their own parents or parents-in-law (upward transfer) while parents may provide the majority of familial support to their children (downward transfer). The nature and extent of support may also vary based on factors such as age, gender, and income.

Satisfaction with network and support

Seven studies (8.75%) provided insights into how the presence of children relates to satisfaction with social networks and support, finding mostly small differences. Becker et al. (2019) examined 16 European countries and observed slightly higher network satisfaction among parents aged 50 or older with three or more children compared to non-parents. The positive correlation between children and network satisfaction was attributed to children who have left home as opposed to resident children. When country-specific analysis was conducted, the overall findings remained largely unchanged. However, some variation was apparent, with some countries showing no significant effects (e.g. Poland, Portugal, Germany, Denmark) while others were consistent with the overall positive association (e.g., Austria, Estonia, Hungary). Similarly, Pushkar et al. (2014) reported marginally higher social support satisfaction among parents, regardless of the gender of their children compared to non-parents. However, two studies (Graham, 2018; McNamee & James, 2012) found that women with children reported slightly lower satisfaction with their support networks compared to women without children, though differences were not statistically significant. Rempel (1985) noted little difference in friendship satisfaction among older individuals (M age = 72.3 years), whereby parents were slightly more satisfied than non-parents. Thus, due to the non-significant difference and relatively high means the authors suggest that the elderly as a whole were satisfied with friendships.

Marital status and gender also play a role. Cwikel et al. (2006) found a significant difference in the proportion of women satisfied with social support, which was greatest among previously married women with children (71.6%), followed by married women with children (67.5%), previously married women with no children (61.5%), married women with no children (58.7%) and lastly never-married women without children (57.2%). Whilst more parents, regardless of marital status, were satisfied with support compared to non-parents, a larger proportion of previously married women were satisfied with support than their married counterparts. Cobb (2002) observed that parenting can impact satisfaction differently based on gender. Parent husbands in the early stages of parenthood (6 months after the birth of first child) reported lower satisfaction and fewer available supports than non-parent husbands while parent wives reported higher and more available supports than non-parent wives. This finding may suggest that fathers are less satisfied as they experience more difficulty sourcing support compared to mothers.

Overall, satisfaction can vary depending on marital status, gender, and the number of children; however the majority of these studies found minimal or non-significant differences. For 4 out of 7 studies, descriptive data on satisfaction with social networks or support in parent and non-parents groups was available (Table S12). Small to medium effect sizes were found, ranging from −0.15 to 0.24 and amounting to an inversely variance-weighted average Hedges g of 0.23, indicating that parents tend to be more satisfied than non-parents but the difference is small.

Differences between non-parents

Nine studies (11.25%) included two or more different groups of non-parents, with many of these studies being conducted in the last decade and the majority focusing on social support. Štambuk et al. (2019) observed significant differences for LGBTIQ not-yet-parents, childfree, and undecided individuals in perceived social support from family, friends, and significant others. Specifically, those who wanted children perceived significantly more support from friends, significant others, and family compared to those who did not want children, with no significant difference compared to those who were undecided. Using data from the Australian Longitudinal Study on Women’s Health, Jin et al. (2023) found a significant difference in social support, noting that among women aged 40–45, childfree women had lower social support than childless women and mothers. A few years earlier, Turnbull et al. (2016) found among women aged 25–44 also based in Australia, circumstantially childless women (those who wanted children but cannot due to reasons other than infertility e.g. absence of a partner, or a lack of resources or opportunity) reported significantly lower social support compared to involuntarily childless women, women who were undecided, childfree women, and women planning to have children in the future. However, Turnbull et al. (2016) recruited a younger sample than Jin et al. (2023), thus, social support may decrease for childfree individuals as they age and are perceived as having less opportunity to change their minds. Adding to the recent research, Ribeiro et al. (2024) found that 72.3% of involuntary definitive childless women (had tried to conceive for at least 2 years but did not achieve pregnancy and were not a fertility treatment patient or an adoption candidate) reported that they received low social support compared to 48.7% of women with infertility who were actively trying to conceive and 49.2% of fertile women planning to have children (not-yet-parents), though the difference was not significant.

Regarding social network characteristics, an early study from 1974 found a significant difference among groups of women in sources of primary friendships. Voluntarily childless (childfree) women primarily formed their closest friendships at work, with 60% identifying work as their main source of primary friends (Bram, 1974). The same study showed women who delay childbearing (not-yet-parents) to have an equal split between friendships formed through work and their neighbourhood, with some also forming friendships through their spouses. Parents, however, were more likely to form their closest friendships in their neighbourhoods, with 54.2% of women and 31.8% of men doing so (Bram, 1974). Jeffries and Konnert (2002) found no significant differences in positive relations with others between mothers, childfree, and childless women.

Overall, some differences between different groups of non-parents were found. Early research suggests not-yet-parents are more likely to form close relationships through neighbourhoods, similar to parents, while childfree women may tend to form close friendships at work. More recent research indicates not-yet-parents perceive and receive more support than childless and childfree individuals, although research is conflicting concerning support differences between childless or childfree individuals.

Differences between parents and non-parents

Overall, there appear to be differences in the size, type, and composition of social networks, as well as sources of support, according to parental status. Our findings suggest parents tend to have larger, more robust family-centred networks, more frequent contact with relatives, and more informal support, predominantly from adult children. This aligns with the Hierarchical Compensatory Model, which positions children as primary sources of support when available and the Task-Specific Support Theory, which highlights family as key providers of instrumental support. While the Task-Specific Support Theory proposes that emotional support is more typically provided by friends, we found that many parents also receive emotional support from family members. This may reflect the overlap between support types, where emotional and instrumental support are often intertwined, potentially facilitated by closer geographic proximity and more frequent contact with relatives.

Conversely, non-parents are often reported as having smaller and more restricted networks but also more diverse social ties outside the immediate family, typically with more friends and extended kin, with the exception of markedly closer relationships with siblings. According to the Hierarchical Compensatory Model, non-parents tend to rely more on secondary and tertiary support providers, receiving more formal sources of support and assistance from non-relatives, when primary providers like children are absent. Similarly, the Task-Specific Support Theory suggests that non-parents supplement their smaller networks with task-specific support providers (e.g., care professionals) or by nurturing broader social ties outside of immediate familial ties, including “families of creation”. However, while friends often provide emotional support, they may be less equipped or expected to offer consistent instrumental help. Accordingly, we found that non-parents tend to receive less instrumental support overall. Additionally, reliance on these less stable secondary and formal supports may be linked to more changeable networks over time, vulnerable to life transitions such as moving house, shifts in social circles, or health-related changes (e.g. Doherty, 2021; Finchum, 2003; Köhler et al., 2021).

It is important to note that we cannot infer a causal relationship between parental status and social networks and support. For instance, Lois and Arranz Becker (2014) found that couples who became parents had networks that were more highly child-compatible compared to couples who did not start a family. However, in further longitudinal path model analyses, the authors determined that child-compatibility was related to fertility intentions: the more child-compatible a couple perceived their network to be, the more intentions they had to start a family. Therefore, parental status may drive some changes in social networks, but importantly, social network characteristics may also influence the decision to have children (see also Keim et al., 2009).

Some studies suggest that social networks and support vary with both age and parental status. Parents generally have larger networks, with network size peaking in early adulthood (and older age) and potentially decreasing or stabilising during midlife for both parents and non-parents (Lang et al., 2013; Lois & Arranz Becker, 2014), aligning with established research on age-related social network changes (Wrzus et al., 2013). Across the lifespan, non-parents may maintain more stable networks (Albertini & Mencarini, 2014), potentially engaging in relationships that are based more on closeness, companionship or connections, rather than reciprocal exchanges of support (Burholt & Wenger, 1998; Vicente & Guadalupe, 2024). However, more research is needed to better understand how age interacts with parental status to influence social support patterns.

Differences between groups of non-parents

This review aimed to explore differences in social networks and support between different groups of non-parents, such as not-yet parents, childless, and childfree individuals. However, only 9 out of 80 studies included in this review allowed comparison between two or more groups of non-parents. Evidence on social network structure is limited. One early study (Bram, 1974) found that childfree women tended to rely on work environments for social connections. However, given the increased proportion of mothers remaining in or returning to the workforce (ONS, 2022), this raises questions about the continued relevance of career-focused stereotypes of childfree individuals (Magennis, 2024) and how work and personal lives intertwine for those without children more generally.

Additionally, despite the mixed findings regarding whether childfree or childless individuals receive less support, it is clear that individuals without children, whether because they choose to or not, perceive less support compared to those intending to become parents in the future. This potentially reflects the societal pressures and negative attitudes towards those who deviate from traditional family structures. Within the childless group, circumstantially childless women received less support than involuntarily childless women, highlighting variation within this group (Turnbull et al., 2016). Our findings also suggest that not-yet parents may have a social life that resembles that of parents, forming friendships through neighbourhoods, similarly to parents (Bram, 1974) and receiving more support than other non-parent groups.

Notably, compared to studies that treat non-parents as a single group, those that distinguish types of non-parents focused on younger adults. This may be the result of both–a shrinking number of non-parents between (late) young adulthood and midlife (i.e. most people, especially former not-yet-parents, following the “normative” life course by having had children before turning 50) and researchers’ assumptions that reproductive distinctions among non-parents (such as whether not having children was intended or not) matter less as people age. While we can only offer speculations here, we encourage researchers to include such finer-grained distinctions in their studies on individuals beyond the typical reproductive years and older adults, as it could help clarify how social networks and support change across the lifespan and whether these experiences differ across non-parent groups. Ultimately, the limited number of studies comparing non-parent groups restricts our theoretical understanding of how these groups compensate for the absence of children.

Limitations and directions for future research

The lack of qualitative studies and inclusion of lower quality studies are sources of potential bias in the review. Of note, lower quality ratings often resulted from lack of detail in reporting in the articles rather than weaknesses in study design; JBI quality appraisals are provided to detail these issues (see Table S9). While methodological quality may affect the robustness of the conclusions drawn, only 15% of studies were rated as low quality. Readers are encouraged to consider the appraisal scores alongside the broader context of each study when interpreting the findings of this review.

Definitions and measures of social networks and support varied considerably. For instance, few studies included relationships beyond family and close friends. While social support was often measured via established scales, subtypes of social support were measured heterogeneously. This diversity limits formal meta-analysis and effect sizes could only be calculated for outcomes deemed sufficiently comparable.

Included studies were primarily conducted in Europe and the US, limiting our findings beyond Western and industrialised contexts (Henrich et al., 2010). Reproductive status may have more profound effects in regions with tighter norms around family formation and more traditional or pronatalist values (Dyer, 2007, see also Nauck & Klaus, 2007). Some studies suggested differences in social networks and support may be influenced by country-specific factors like culture, economic development, and welfare systems. Moreover, the findings from studies included in this review span 50 years, further limiting the ability for direct comparison due to changing historical and cultural contexts. Age-related variations may also be overlooked, as most studies in this review focused on individuals aged 50 and older and few examined changes over the lifespan. Longitudinal studies are needed to assess whether observed differences persist at different life stages.

Further research is needed to understand how demographic differences intersect with social networks and support (e.g., Kersten et al., 2024). For example, non-parents are more likely to be single and live alone; observed network and support differences between parents and non-parents may partly reflect characteristics like these rather than parental status itself. Health conditions, socioeconomic status, gender, and a range of other factors may impact both social networks and parental/reproductive status. Therefore, differences in social networks and support should be interpreted in the context of broader demographic and life circumstances.

This review highlights a significant gap in the literature regarding distinctions among non-parents. 87.5% (70/80) of studies included a non-specified non-parent sample (i.e. no details of whether participants were childless, childfree, etc.) and only 11.25% (9/80) included two or more different groups of non-parents. Clear and consistent operationalisation of non-parent status is vital in establishing a cumulative science on how social networks and support may differ for different non-parent groups as well as the correlates and consequences of reproductive status and childbearing decisions.