Regretting fatherhood in Spain

Objective: In this first study of its kind in Spain, we analyse the scope of and reasons underlying paternal regret. Background: Research on parental regret, a subject only recently broached by analysts, tends to focus on motherhood. Regretting fatherhood has been only scantly researched. In this study we test the effects of intensive fathering, the use of different care resources, economic and employment conditions, and satisfaction with respondents’ partnership on their regret for having children . Method: The analysis is based on an online survey of parents of children under 7 years old (QUIDAN Survey) A total of 3100 parents were interviewed, with the sample evenly distributed by sex and youngest child’s age, and proportional by parents’ highest level of schooling and place of residence. The weighted subsample used in this article included 1374 fathers. The hypotheses are tested with logistic regression. Results: The social factors associated with a greater likelihood of regret include circumstances that challenge men’s role as primary breadwinner, a negative impact of fatherhood on job career, a high dependence on grandparents for balancing working and private lives and partnership dissatisfaction. A relationship between intensive fathering and regret could not be observed. Conclusion: The findings show the extent of paternal regret to be fairly limited, similar to the proportion reported for maternal regret. Paternal regret is mainly associated with family and working circumstances. factors on paternal regret. The results will be used in future comparisons to the results of research on maternal regret as a basis for discussion of the implications of gender roles and parental models in fathers’ reactions. From an institutional standpoint they conte xtualize regretting parenthood in a Southern European country.


Introduction
Parental regret has been studied primarily in connection with motherhood, for in conventional male breadwinner family models, responsibility for childcare is borne by mothers. In that context, maternal regret runs counter to standard societal expectations, as attested to in Orna Dornath's (2017) pioneering book, Regretting motherhood, which has been widely criticised, in Germany in particular (Heffernan & Stone, 2021a). Public reactions were shared on Twitter and Facebook, often tagged #regrettingmotherhood that kept the topic of maternal regret at the forefront of national debate in April and early May 2015 (Heffernan & Stone, 2021b). The Israeli author analysed regret in the context of the clash between norms that saddle women with the obligation to have and raise children and notions of individualism and freedom of choice prevailing today.
The moral and cognitive stance of motherhood is a normative act. Regret can be defined, following Donath (2015), as an emotion judgment when experiences or actions are not those expected in motherhood. Moreover, narratives on negative emotions, including regret, tend to sustain and endorse cultural systems of values and beliefs. As a result of the strong public interest the book in Spain, Orna Donath was invited to discuss her controversial work along with other women writers and researchers to the Centre of Contemporary Culture of Barcelona (CCCB) in 2017. She explained how she drew her study around three variables of a sociological triangle: parenthood/non-parenthood, society, and regret. This figure encloses what she called "a political usage of emotions", because women can only be fully recognized as such by becoming "a mother of others" (Pérez-Nieto, 2021:198).
Quantitative research has contributed little to the analysis of regret: studies from the life course approach do not usually model the dimensions of well-being considered relevant in qualitative gender studies, but rather employ a general measure of life satisfaction. Giesselmann et al.'s (2018) longitudinal survey on German mothers includes concepts of maternal well-being identified in gender studies, such as guilt. The authors assume that the maternal experience of guilt is accentuated by the tension between the model of intensive mothering and the norm of the working mother. Their results, while not specifically mentioning regret, show how the transition to motherhood in Germany is associated with a substantive decline in women's mental well-being. The lowest level of well-being is estimated around 6 and 7 years after the transition to motherhood. Given that maternal commodification increases with the age of the child, the authors assume that rising pressures to be employed outweighs the mothers' gradual relief from caring duties. The decline in mothers' well-being could be a consequence of societal mothering ideals.
Cesar et al. 's (2018:2) quantitative analysis of Facebook posts of Portuguese mothers sharing feelings shows how regret is an emotion that is considered a deviation from the norm. These mothers' self-doubts, feelings of guilt, and shame are induced whenever women, in some way, "fail to live up to ideals of womanhood and motherhood." Regretting fatherhood, however, has been scantly researched, most probably because social expectations around childcare and bonding with offspring are different for fathers than mothers. As O'Reilly (2020) puts forward, maternal regret means "I would rather be a father". A survey conducted in Germany (YouGov, 2016(YouGov, & 2022 estimated parental regret at around 20 %, one in the US (Newport & Wilke, 2013) among parents over 45 at 7 %, and one in Poland (Piotrowski, 2021) at 14 %. The author of the latter contended that political and institutional factors in Poland could explain the high percentage of regretful parents, in particular the country's highly restrictive legislation on abortion.
The model of fatherhood has evolved in recent decades from the conventional father-breadwinner and disciplinarian to a more plural concept that includes fathers' direct involvement in childcare (Flaquer et al., 2020;Miller, 2011). Miller (2011 related such bidirectionality to the expectations conventionally associated with remunerated work on the one hand and with greater participation in caring on the other. An analysis is therefore in order to determine whether or not that transition is associated with regret prompted by the tension between the breadwinner and carer roles or occupational aspirations and parental obligations, as observed among mothers. More generally, the societal factors that induce regretting fatherhood need to be identified.
The current study examines the effect of fathers' involvement in childcare and other occupational and socio-economic factors on paternal regret. The results will be used in future comparisons to the results of research on maternal regret as a basis for discussion of the implications of gender roles and parental models in fathers' reactions. From an institutional standpoint they contextualize regretting parenthood in a Southern European country.
The objectives here include an analysis of the extent of paternal regret in Spain and the identification of the family and occupational profiles and dynamics associated with that sentiment. More specifically, the analysis addresses how the time devoted by fathers to caring for their children affects the risk of regretting parenthood; how such regret is impacted by the difficulties posed by remunerated work demands; and the significance of respondents' satisfaction with their partnership in terms of paternal regret.

Intensive maternity as ideal and reproduction of role expectations
As noted earlier, maternity models are informed by societal expectations linking child raising to the 'proper' enactment of the maternal role (Hochschild, 1997, Katz-Wise et al., 2010. In intensive mothering, the prevalent model in Spain, women are expected to be 'perfect mothers', whilst men are not made to feel obliged to being 'perfect fathers' (Meeussen & Van Laar, 2018). French philosopher Elizabeth Badinter (2010) warned against over-internalization of perfection as a cultural norm in intensive mothering and male demands on their partners to bear children. Intensive motherhood springs from an essentialist belief in women's biological ability to afford their children the best possible care and consequently the need for them to be as close as possible to their babies. That view of motherhood brings a good deal of pressure to bear on mothers, for their interaction with their children is expected to follow along those lines (Sanchez-Mira & Muntanyola-Saura, 2020). Matley (2020) as well as the German #regrettingmotherhood and also the French #regretmaternel and the Spanish #madresarrepentidas show how social media might be a space in which women can break the taboo of regret. His analysis of "Mumsnet" online posts shows that the ''regretful" posters often depict motherhood as a role that is incongruent with notions of self-fulfilment: motherhood is not valued as a source of women's identity. The Finish authors Sihto & Mustosmäki (2021) replicated Donath's study with similar results, but claim that it is likely that the data does not represent 'all' women or mothers, because voices heard in the data are mainly those of white, relatively privileged middle-class women. Poverty and difficulties in providing for the family's basic needs do not emerge as a source of regret.
Such social pressure derives from ideals that call for a deep maternal commitment to care. In conventional or neo-conventional ideals (Flaquer et al., 2021) that commitment complements the father's role as breadwinner, in which childcare is not a priority. That gendered division of tasks explains why regretting parenthood, like other childcare-related conditions, such as post-partum depression and the empty nest syndrome, is associated with women. Societal expectations around parenthood are changing, however, and the idea that fathers must become actively involved in raising their children is widespread in affluent societies. The latest statistics from Pedersen (2015) on a "Mumsnet" census show that men make up 16% of all users of "Mumsnet". Pedersen argues that men use the site because they wish to be seen as parents rather than fathers and because they are attracted by the variety and quality of discussion. Although collective behaviour changes slowly, a growing proportion of fathers is taking an active part in childcare, with the building of closer bonds with their children as a side effect, especially after caring for their very young children alone (O'Brien & Wall, 2017). Higher expectations around the quality of a partnership and childcare and fathers' performance of tasks conventionally associated with motherhood may clash with career expectations and demands, inducing occupational and parental tension and burnout. Psychological studies show how the relationship between maximize the number of options and choice in decision making is often experienced in terms of regret, adaptation, and self-blame (Ogle, 2017). Moreover, in quantitative questionnaires and qualitative reports Varian (2021) reported very high levels of satisfaction and very low levels of regret regarding women's decision to be sterilized, and many unsterilized women noted the negative impact of not being sterilized on their lives.
Given that societal pressure for (intensive) fatherhood is lighter than for motherhood and the associated opportunity costs are likewise lower, fathers might be assumed to regret being parents to a lesser extent than mothers. That is our first hypothesis (H1). Studies on parental burnout seem to point in that direction, although they also warn that such frustration may be temporary and not necessarily lead to parental regret. Roskam & Mikolajczak (2020) and Aunola et al. (2020) reported a higher rate of burnout syndrome among women, which they associated with the goal of perfect performance of the maternal role, i.e., with internalized ideals about deep involvement in child-raising. Piotrowski (2021), however, found a clearer connection between burnout and regret in fathers than in mothers. That author concluded that regret grew out of exhaustion and a perceived lack of sufficient material and emotional resources to engage effectively in parenthood.
In Spain, Barbeta-Viñas & Cano (2020) identified what they defined as the 'fatigued father' profile in committed fathers who found their own conditions unacceptable for childcare. Similarly in Sweden, Pedersen et al. (2021) observed some men committed to childcare to suffer post-partum depression and regret. Hence it is important to verify whether regret is greater among fathers deeply involved in childcare. Reframing regret as exhaustion, stress, and burden-states more commonly allied with the experience of parenting and thus, more compatible with cultural norms of parenthood-arguably dilutes the more radical emotion of regret. Barbeta-Viñas & Cano (2020) also observed a 'desired paternity deficit' defined as the unease and frustration felt by young fathers unable to conform to ideals and expectations in connection with participation in childcare. Like the mothers analysed by Donath, fathers would therefore be expected to harbour ambiguous and contradictory feelings about, and even regret, in certain circumstances.
Our second hypothesis (H2) is that everyday reality and intensive childcare may raise the likelihood of regretting fatherhood. Underlying that hypothesis is the idea that engaging in fatherhood resembles intensive motherhood, for the two parental roles and associated feelings such as regret are similar, despite the differences in societal pressure.

Occupational and economic conditions and regret about having children
Despite the trend toward greater acceptance of involved fatherhood, the idea that fathers should be the family breadwinner continues to prevail in today's society. The gendered division of labor into care and employment translates into the "motherhood penalty" and the "fatherhood bonus" (Kmec, 2011;Munsch, 2016). That notwithstanding, paternity may also change men's career outlook. Johansson (2010), for instance, described career-oriented middle-class men who, after becoming fathers, moved on to less timedemanding jobs, more compatible with intensive fathering.
Of the parents analysed by Piotrowski (2021), educational level did not raise the likelihood of regret, attesting to the cross-sectional nature of that feeling and its presence in different social circumstances. Conversely, financial problems have been associated with lower satisfaction with parenthood (Hansen, 2011) and with a higher probability of parental regret (Piotrowski, 2021). The mothers surveyed by Dornath (2017) also stressed that job insecurity, other occupational difficulties, and unstable domestic finances or a weak welfare support system greatly hindered engagement with motherhood.
Although the economic impact of fatherhood on household expenses is particularly intense among the least advantaged classes, Hansen (2011) noted that it is also detrimental to middle-class fathers due to the opportunity costs of paternity as related to career, income and training. Nonetheless, in countries with developed welfare states where a considerable share of social spending is allocated to lightening family loads, the impact is less severe.
In that regard, social class is relevant to the ability to engage in inclusive fatherhood (Hansen, 2011;Johansson, 2010) and the least advantaged classes are the ones with the fewest resources to meet societal expectations around more involved fatherhood. Thus, our third hypothesis (H3) is that the rate of regretting fatherhood rises with greater financial and job insecurity.

Pivotal importance of harmonising employment and family life
The effect of difficulties in attaining life/work balance on workers' mental and emotional health and on family dynamics has long been researched (Leineweber et al, 2018;Newton et al., 2012). It should consequently come as no surprise that regretting parenthood has been associated with the difficulties involved in reconciling employment and family life. Glass et al. (2016) draw on theories about stress and mental health, as well as research on the effects of public policies on the experience of parenting, and claim that the larger policy context helps explain why adults residing with children report less happiness than those not living with children.
Our fourth hypothesis is based on Piotrowski's (2021) findings that in Poland the gap between fathers' resources and childcare demands raises the likelihood of regret. On those grounds, the partners' use of outside care is considered an indication of difficulties to harmonise employment and childcare. More specifically, fathers who depend on such resources for balance (such as active grandparental participation in childcare, hired domestic care or a reduction in working hours) are expected to have a higher likelihood of regretting fatherhood than others (H4).

Life satisfaction and parenthood
One of the factors associated with regretting parenthood is identification with romantic ideals (Dijkstra & Barelds, 2008), and the notion of romantic love prevails among Spanish families (Roche, 2020). That association is unsurprising, given that emotions, such as regret are not atomised psychological products that arise out of nowhere, but the fruit of social experience and interaction with loved ones. The ideological bounds of intensive mothering and paternal involvement govern emotions tailored to a given time and place (Hochschild, 1979).
Negative emotions, such as regret may persist beyond the transitional stage of child birth (Moore, & Abetz, 2019). Emotions are ambivalent, as mothers who narrate the brighter and darker sides of their postpartum experiences are well aware (Bernardi et al., 2015;Sevón, 2005). The difference between Donath's "Regretting Motherhood" and Lazarre's "The Mother Knot" is that, although both mothers love their children, the women interviewd in Donath's study think motherhood was a mistake, while Lazarre embraces the lifelong ambivalence of becoming a mother. Both are different approaches to what being a mother means: a relationship with a new person which cannot be undone (Pérez-Nieto, 2021). However, Preisner et al. (2020) show how all else being equal, German fathers rank above childless men in terms of life satisfaction. The emotional rewards of having children may close the gender gap of happiness.
Consequently, from the standpoint of the symbolic interaction perspective, emotions are not an individual but a collective product. Illouz (1997) defended the need to analyse emotions sociologically, limiting her study to heterosexual love in capitalist societies. She specifically contended that romantic love and capitalist culture have elective affinities that define what her respondents deemed romantic and what they valued in their love lives. She added that romantic love has been infused with an aura of naughtiness while at the same time being raised to a supreme value.
In that context, our fifth hypothesis (H5) is that a failed partnership is associated with regret over having children. The success of the affective-sexual life plan, linked to ideals such as romantic love, endurance of the couple relationhip, a steady marriage or continuity of the family commitment, appears to be a safety net to guard against possible regret around the project for parenthood.

Data
This study is based on an online survey of parents with children under 7 years old (QUIDAN Survey) conducted by NETQUEST, an organization specialized in online surveys, during May and June 2021. The authors of this paper designed the survey. The questionnaire included items primarily concerning the resources used to balance employment and family, partnership satisfaction and conflict, and opinions on family. A total of 3,100 parents were interviewed, with sample quotas evenly distributed by sex and youngest child's age (in order to get enough fathers of children of all ages) and proportionally to parents' highest level of education and place of residence. The raw data were weighted to counteract the demographic oversampling of younger children´s parents and educational level to ensure statistical representativeness (established according to the distribution of parents drawn from the Labour Force Survey). For the purpose of this article, we analyze a subsample of 1,377 fathers, that excludes those with missing data and those not living with a partner. The hypotheses were tested with logistic regression of variables related to family and occupational circumstances, work/life balance strategies, opinions on childcare and socio-economic characteristics.

Dependent variables
Data on regretting parenthood were collected with the question, How fully do you agree with the following statement?: "If I could start over, I wouldn't have children"'. The five replies provided in the questionnaire ranged from completely agree to completely disagree and 'I prefer not to answer'. That question was preceded by other on regret over respondents' level and type of schooling, satisfaction with their partnership and satisfaction or conflict with their partner. Only 4.8 % of the respondents preferred not to answer the question, with no statistical difference between fathers and mothers. For the purpose of logistic regression, the answers 'completely agree' and 'agree', representing parental regret, were assigned a value of 1, and the other answers were assigned a value of 0.

Independent variables
The following variables on family circumstances were included in the analysis. In connection with children, respondents were asked whether they had children from an earlier relationship (value=1); the oldest child's age (under 6, 6 to 9, and 10 or over); and whether any of the children had a condition or disability that interfered with their performance of everyday activities (value=1). To measure partnership dynamics, respondents were asked to assess their satisfaction with their partner on a scale of 1 (completely dissatisfied) to 10 (very satisfied).
The variables on parents' job circumstances and experience at the time of the survey were coded as dummy variables under the categories unemployed, working fewer than 35 hours/week, working 35 to 40 hours/week (used as the reference) and working over 40 hours /week. A dummy variable on the difference in the spouses'/partners' incomes was also included, where 'the man earns more than the woman' was assigned a value of 1 and any other combination, 0. A further question aimed to determine the respondents' subjective perception of whether fatherhood had been an obstacle in their career, likewise on a 10-point scale, with 1=not at all and 10=indisputably.
The items considered in connection with work-life balance included the father's involvement in childcare and the use of outside resources. The former was determined by the use of leave time to care for the children as dummy variables: used (fully paid) paternity leave (value=1; reference category not used=0); used part-time (unpaid) parental leave (1); and used full-time (unpaid) parental leave (1). The time spent by fathers caring for their children was also taken into consideration, along with the time devoted to housework, measured not linearly but by contrasting full commitment (defined as over 1.5-fold the mean devoted by all male respondents), scored as 1 and other situations, as 0. The items used to determine the use of outside resources were frequent grandparental participation in childcare (1=at least several times a week) and resorting to hired help for childcare or domestic chores, both used as dummy variables.
The questionnaire also included data on respondents' agreement with or ambiguity around (value=1) the statement: 'Having children has more drawbacks than advantages' and disagreement with 'Child-raising is a source of everyday satisfaction'. The demographic variable controlled for was respondents' and their partners' highest level of education, with three choices: primary, secondary and higher education. Because many respondents failed to answer the question on income, the family's financial situation was measured based on their replies to the question on the ability to make ends meet, where the five choices ranged from comfortably to barely. The percentages, means and standard deviations for the subsample for the independent variables defined in the analysis are shown in Table 1.

Work-life balance indicators
Paternity leave taken for at least one child 85.9 Full-time parental leave taken to care for at least one child 7.8 Part-time parental leave taken care for at least one child 5.1 1.5 more time devoted by father to housework than the mean 26.6 1.5 more time devoted by father to childcare than the mean 29.8 Intense grandparental involvement in childcare (at least several times a week) 13.

Results and discussion
While infrequent, regretting parenthood is felt by a small fraction of parents of young Spanish families, even though family planning has been routinely implemented for decades. The percentage of parents clearly stating they would not have children again, at 5.1 %, is lower than the 20 % in Germany (YouGob, 2016(YouGob, & 2020, the 14% in Poland (Piotrowski, 2021) and the per cent among parents 45 and over in the United States (Newport & Wilke, 2013). Regret was found somewhat more frequently among fathers than mothers when the entire spectrum of choices was taken into consideration (see Table 2). When only the options most clearly expressive of regret (completely agree and agree) were considered, however, the difference between the sexes was smaller. This is consistent with the findings form surveys conducted in Germany (YouGob 2016) (19 % among mothers and 20 % among fathers) and Poland (13.1 % and 14.4 % respectively) (Piotrowski, 2020). Consequently, in the bivariate case, the first hypothesis was not confirmed, for fathers did not regret parenthood less frequently than mothers, despite the lighter pressure on them to engage in intensive fatherhood and the lower career opportunity costs of father-than motherhood (de Quinto et al., 2020;Hansen, 2012;Kmec, 2011;Munsch, 2016). A possible inference of this result is that the intense societal pressure on women in Israel to have children as identified by Donath (2017) may not be the actual reason for regretting motherhood. Moreover, the arrival of a child also has consequences for traditional fathers because it changes the material needs of the family and conditions time use after work time. As might be expected, paternal regret was associated with negative opinions about the role of childraising in life satisfaction. The odds ratio for regret among fathers who deemed child-raising to have more drawbacks than advantages (or responded ambiguously) was 11.4-fold greater than the ratio among those who deemed the opposite (p=.000). Those who did not feel child-rearing to be a source of everyday wellbeing also had a much higher odds ratio for regret than those who linked the two (odds ratio= 4.85; p=.000).
Intensive fathering has been related to the likelihood of regretting fatherhood as premised in the second hypothesis, with signs of burnout among deeply involved fathers (Aunola et al., 2020;Roskam & Mikolajczak, 2020). Our analysis does not support this hypothesis, controlling for its impact among those with a university degree (therefore with higher opportunity costs for being highly involved) or whose fatherhood had a negative impact on their work career (results not shown in table 3). It could be that the development of positive feelings and intense bonding with the child(ren) dervided from intensive caring (O`Brien & Wall, 2017) could partly outweight the negative effects, such as lack of time and stress. Further analysis should focus on this relationship. Table 3 presents the odds ratios of regretting fatherhood. Controlling all other factors in the model, the likelihood of regret was also found to increase when one or several children had a disability that interfered with their performance of everyday tasks. Risk of regret was 4.3-fold higher in such cases than when the children had no disability (odds ratio=4.334; p<.05). Such regret may stem from the greater time demanded for childcare and/or from the thwarting of the expectations initially placed in the family project. Our analysis revealed the absence of any effects of other family factors on regretting fatherhood. Furthermore, although paternity has lower direct and career opportunity costs than maternity, and the literature has even reported a so-called fatherhood wage bonus (de Quinto et al., 2020;Kmec, 2011;Munsch, 2016), the use of paternity leave may be detrimental to fathers' careers (Meil et al., 2018b). As noted in the preceding section, the present survey asked respondents to subjectively assess whether paternity had entailed any adverse effect on their careers, with 57 % responding that it had. The multivariate analysis shows that the greater the perceived adverse impact of paternity on a respondent's career, the higher was the risk of regret (odds ratio =1.174; p<0.01), controlling other variables. It might be added that when paternity entailed the need to take part-time parental leave (i.e., fewer hours with proportionately lower pay), the odds ratio of regret rose nearly five-fold (odds ratio= 4.927; p<0.01). That was not found for full-time parental leave, however, probably because its duration is normally very brief, whereas in part-time parental leave the term tends to be much longer (Meil et al., 2018b).

Work-life balance characteristics
Paternity leave taken for at least one child 0.656 Full-time parental leave taken to care for at least one child 0.509 Part-time parental leave taken care for at least one child 4.927** 1.5 more time devoted by father to housework than the mean 1.060 1.5 more time devoted by father to childcare than the mean 1.662 Intense grandparental involvement in childcare (at least several times a week) 3.626*** Paid household help 1.056

Opinions on parenthood
Having children has more drawbacks than advantages 12.141*** Parenting is not a source of satisfaction 5.097*** Socioeconomic characteristics On the whole, the likelihood of regret rose when parenthood entailed obstacles in fathers' careers. Nonetheless, where parenthood had no adverse impact on their traditional breadwinner role, the risk of regret declined substantially. Fathers who earned more than their partners had an odds ratio for regret onethird the value of those who earned the same or less than theirs, when all other relevant variables were controlled for (odds ratio=.299; p<0.01). That finding confirmed our third hypothesis, but contrary to the results of earlier studies (Piotrowski, 2021), here the probability of regretting fatherhood was not associated with family financial problems (odds ratio=.877, p>.05) or job insecurity (odds ratio=1.498; p>.05). Paternal regret was not directly correlated with social class or financial difficulties, but derived from what some fathers perceived as a threat to their traditional breadwinner role. Regret was also seen to be unrelated to fathers' level of schooling, where no statistically significant differences were observed, a finding consistent with the results of Piotrowski's (2021) study in Poland. However, because social class is a social indicator that impacts many aspects of life, future research should evaluate further the role of social class on paternity emotions.
Fathers who were not working at the time of the survey were much less likely to feel regret than those working full time (odds ratio=0.172; p<0.05). Whilst fathers' unemployment may cast doubts on their breadwinner role, the possible causes of such unemployment, by contrast to those in play among women, is commonly attributed not to their family circumstances but to the economic environment in their place of residence (as Spain is characterized by chronic high unemployment rates, particularly among young people). Regretting fatherhood would not therefore be expected to be higher in this case. The lower frequency of regret among unemployed men might also be attributable to the fact that they had more time available and therefore perceived childcare to be less of a burden. Not only that, but in such cases, childcare may contribute to affording fathers a role, that of carer, highly appreciated socially that would partially offset their inability to provide for their family (Castrillo et al., 2021). Nonetheless, the risk of regret was higher when the partner was unemployed (odds ratio=3.943; p<0.01), perhaps also an expression of dissatisfaction with the family's adoption of a conventional reconciliation strategy. In other words, it may also denote dissatisfaction with the adverse consequences of maternity for mothers' careers (Kmec, 2011;Munsch, 2016), which are detrimental not only to women's well-being and aspirations, but also to the family economy and therefore fathers' well-being.
Our fourth hypothesis (H4) about the effect of resorting intensively to outside care to harmonise work and family life was also partially confirmed. Whereas resorting to paid assistance to perform household tasks and/or care for children did not affect the likelihood of regret, using intensive grandparental support increased the risk of regret (odds ratio= 3.626; p<.001). That may well be attributable to the appearance of a feeling of dependence and guilt for having to resort to grandparents rather than to the lack of sufficient financial resources to pay for outside care, for as noted above, regretting fatherhood was not conditioned by income level. Whilst grandparents play a significant role in reconciliation strategies for some Spanish families with small children (Meil et al., 2018a), their participation as carers is not necessarily desired, with most families preferring to involve them more sporadically than intensively (Valarino et al., 2018). Fathers' part-time work was not observed to be associated with a higher risk of regret because the reason fathers took such jobs was seldom to balance work and family life, whereas a correlation was found, as noted, when they had to take part-time parental leave for that purpose.
Lastly, as premised in our fifth hypothesis, the degree of partnership satisfaction was observed to condition feelings around parenthood, not only among mothers as might be deduced from the cases analysed by Donath (2017), but among fathers as well. The analysis showed that the higher the satisfaction, the lower was the risk of regretting fatherhood (odds ratio= .793; p<.001) and vice-versa. The origin of that association probably lies in the prevalence of romantic love among Spanish families (Roche, 2020), given the relationship observed between romantic love and regret (Dijkstra & Barelds, 2008). That feeling, like all feelings, may be ambiguous and transitory until the dissatisfaction with the partnership resolves.

Conclusions
In the wake of Donath's (2017) book, regretting motherhood has begun to be analysed and discussed in academia and in public opinion in countries where the norm is planned parenthood and intensive mothering. Studies conducted to quantify that sentiment, among others the one described here, have shown that the rate of incidence varies substantially from country to country. It is much less frequent in Spain than in the other countries for which we found comparable information and, as there, it has been observed not only among mothers but also in fathers, with no statistically significant differences between the sexes. Those findings question one of the primary contentions around the origin of regretting motherhood, according to which it lies in societal pressure on women to have children and the adverse consequences of those demands for individual career and life projects. The challenge to that premise arises because men are presumably under no such pressure, nor does fatherhood penalise their careers.
The present findings reveal that regretting fatherhood has roots in a number of social factors. One is associated with the fact that not only mothers but also some fathers perceive parenthood to have adverse consequences for their careers, and the more intense that perception the higher the likelihood of regret. Such feelings are also more probable when paternity poses life/work balance issues (heavy dependence on the care provided by grandparents, need to resort to part-time leave or partner unemployment) and when the man ceases to be the family's primary breadwinner. Regretting fatherhood is consequently closely related not only to the type of work/life balance strategies adopted, but also to the challenge fathers may perceive to their role as family breadwinner and their personal development through their careers.
The hypothetized relationship between intensive fathering and regret could not be supported, as the statistical significance level of the relationship is over the conventional threshold, even controlling for specific groups of fathers such as professionals with university degree and those stating that fatherhood had a negative impact in their work careers. This is probably due to the lack of enough cases in both circumstances. Further research should explore under which circumstances intensive fathering turns to a burden that translates into regretting becoming a father.
Respondents' dissatisfaction with their partnership was observed to be another source of regret, which was the more likely where satisfaction was lower. That association appears to be more or less transitory, however, at least among men who have reconstructed their partnership, for the presence of children from a former relationship appears not to be a risk factor for paternal regret.
The present study, a first approach to regretting fatherhood in Spain, is subject to a number of limitations that merit mention and must be resolved in subsequent research. First, the fact that regret over being a father is a minority condition prevented us from applying specific analytical models to the data for that group of fathers. Inasmuch as building a sufficiently large subsample of repentant fathers would call for a broader database, future studies might adopt a qualitative approach to regretting parenthood to acquire a fuller understanding of its societal implications. Second, as the data denote regret at the time of the survey only, we do not know for how long the respondents felt that sentiment or whether it varied over time. Future research based on panel data where changes in paternal regret can be monitored over time would be recommended to further our understanding of the variability of regret and its relationship to children's ages. Last, as this study analysed fathers only, the factors inducing maternal regret in Spain have yet to be addressed. The present findings may be compared to such future data to establish possible differences in the way societal expectations may affect women's and men's feelings around parenthood.