Based on Self-Determination Theory and Separation-Individuation Theory, the present meta-analysis examined whether the association between parental autonomy support and offspring internalizing and externalizing problems differs between countries and regions of the world, and according to national levels of individualism–collectivism and power distance. A systematic search of electronic databases yielded 404 studies from 53 countries, which were included in a hierarchical meta-analysis. On average, we found significant negative correlations of autonomy support with internalizing and externalizing problems, with minor between-country variance and no variation between regions. Although national levels of individualism-collectivism were not significant moderators, levels of power distance interacted with the content of autonomy support. Parental promotion of independence showed a weaker association with internalizing problems in countries with higher power distance scores; however, associations of promotion of volitional functioning with child problems did not vary by levels of power distance. Our results support the suggestion that the association between parental promotion of volitional functioning and child outcomes tends to be similar across cultures, whereas the association between parental promotion of independence and those outcomes shows some cultural variation.
Different theoretical approaches – such as Self-Determination Theory (SDT; e.g., Ryan & Deci, 2017, 2019), Separation–Individuation Theory (SET; Blos, 1979; Mahler et al., 1975), and the concept of age-typical developmental tasks (Havighurst, 1972; Van Petegem et al., 2019) – consider the development of autonomy to be an important aspect in children’s and adolescents’ lives, although these approaches differ in the conceptualization of autonomy. According to SDT, autonomy refers to the sense of volition and being a causal agent of one’s own life, that is, the ability to perform actions through one’s own will (Ryan & Deci, 2017). It means making choices that reflect one’s true values and interests rather than feeling obliged or seduced to act in a certain way (van Petegem et al., 2013). In contrast, SET focuses on the independence from others and the degree of distance versus proximity in the parent–child relationship (Blos, 1979; Mahler et al., 1975; Van Petegem et al., 2013).
The development of children’s autonomy depends, upon others, on how autonomy-supportive their parents are. Parents can demonstrate autonomy support by providing choices, encouraging freedom of expression and action, acknowledging their children’s perspectives, and encouraging independent thinking and decision-making (Deci & Ryan, 2012; Soenens et al., 2007). Soenens et al. (2007, 2018) distinguished between two forms of parental autonomy support (PAS) based on different views of autonomy from SDT and SET: promotion of volitional functioning (PVF) which involves encouraging the offspring to make choices that reflect their true values and interests (not necessarily independently from the parents), and promotion of independence (PI) which involves encouraging independent thinking, decision-making and problem-solving without parental or others’ support. However, PVF and PI tend to be highly correlated (r = .68 for mothers and 0.70 for fathers, according to Barberis et al., 2021; r = .58 according to Liga et al., 2020), indicating the cooccurrence of behaviors that promote independence and volitional functioning, or problems with measuring both constructs separately.
Because autonomy is a fundamental human need, it has been argued that PAS plays a crucial role in enhancing children’s well-being (Grolnick, 2012). Similarly, SDT proposes that PAS protects against problem behaviors because of fostering coping mechanisms and self-control (Matte-Gagné & Bernier, 2011; Vansteenkiste & Ryan, 2013). In fact, a meta-analysis of 36 studies by Vasquez et al. (2016) found that PAS is related to psychological health (r = .38) and other positive outcomes. A meta-analysis on three studies by Yap and Jorm (2015) found a moderate correlation between PAS and anxiety symptoms (r = − .29). Finally, a meta-analysis by Manuele et al. (2023) found negative associations of PAS with depression (r = − .20 to − 0.21) and anxiety (r = − .13 to − 0.16).
However, the consistency of associations between PAS and child outcomes across cultures has been questioned. While proponents of SDT suggest that autonomy is a universal human need, thus indicating positive outcomes across cultures (e.g., Marbell-Pierre et al., 2019; Ryan & Deci, 2019), other researchers suggest cultural variability, referring to two cultural dimensions proposed by Hofstede—individualism/collectivism and power distance (Hofstede et al., 2010). Individualistic cultures value the development of autonomy, whereas collectivistic cultures promote cohesion and loyalty to the group, encouraging children to fulfill their duties to their families (e.g., Chao, 1994; Grusec, 2008; Rothbaum & Trommsdorff, 2007). Thus, PAS may not be as relevant and may not have the same positive effects in collectivistic cultures. However, this argument mainly holds true for PI rather than for PVF as the latter does not necessarily imply independence from the parents (Soenens et al., 2007).
The second cultural dimension considered, power distance, is defined as the extent to which less powerful members of organizations and institutions (such as families) accept and expect unequal distribution of power (Hofstede et al., 2010). In countries with high power distance, parents tend to teach children obedience, while in countries with low power distance, they treat children more as equals. Zhai and Gao (2009) suggested that PAS may be particularly incompatible with societies that entail hierarchical parent–child relationships and parental dominance. In fact, Mone et al. (2016) found that adults in countries that score higher in power distance were less likely to perceive independence as an important socialization goal for children and more likely to perceive obedience as a socialization goal. Since national levels of individualism-collectivism and power distance correlate with economic affluence (e.g., Basabe et al., 2002), research should statistically control this confounding factor.
To date, very few empirical studies have compared the correlates of PAS across cultures. Some results were consistent with the suggestion that autonomy-granting is more strongly associated with child outcomes in high-individualist and low-power-distance cultures. For example, Marbell-Pierre et al. (2019) found that parental provision of independent choice was related to positive developmental outcomes in US adolescents but not Ghanaian adolescents – a country with a lower individualism and a higher power distance (Hofstede et al., 2010). Manzi et al. (2012) found that parental promotion of autonomous thought showed a stronger negative association with depressive symptoms in the U.S. than in China which scored lower on individualism and higher on power distance (Hofstede et al., 2010). In a short-term longitudinal study, parental autonomy-support predicted increased psychological well-being and academic functioning among American and Chinese students. However, associations were stronger in the U.S. (Wang et al., 2007). In that study, PAS was defined as allowing choice-making and willingness to consider the child’s opinion. However, Xiong et al. (2022) found similar associations of PAS with child outcomes in the U.S. and China using the same scale. Interestingly, three studies found even stronger associations between PAS and child outcomes in countries that score low in individualism and high in power distance. Chirkov and Ryan (2001) reported that parental PVF is associated with lower depressive symptoms in Russian adolescents though the association was nonsignificant in U.S. adolescents. This is despite the fact that individualism scores are much higher in the U.S. than in Russia, while the reverse is true for power distance (Hofstede et al., 2010). Additionally, Filus et al. (2019) found the strongest association between paternal autonomy support and child well-being in Polish families as compared to families from Norway and Switzerland, despite individualism being lower and power distance higher in Poland than in Norway and Switzerland (Hofstede et al., 2010). That study did not observe any cultural differences regarding the correlations of maternal autonomy support. Finally, Manzi et al. (2012) found stronger associations between parental promotion of autonomous decision-making and depressive symptoms in China than in Belgium, even though China has lower individualism and higher power distance scores (Hofstede et al., 2010).
These inconsistent results highlight the need for more cross-cultural comparisons. Unfortunately, the meta-analyses by Yap and Jorm (2015), Vasquez et al. (2016), and Manuele et al. (2023) did not provide these comparisons. Thus, the present study aimed to compare associations of PAS with child outcomes across the globe. We focused on associations of PAS with internalizing and externalizing problems in the offspring because these symptom groups have been often empirically related to parental behaviors. This provides a larger database than academic achievement, academic motivation, or other outcomes (Vasquez et al., 2016). Internalizing symptoms refer to negative emotions, such as anxiety and depression, as well as social withdrawal and somatic complaints. In contrast, externalizing symptoms include delinquent and aggressive behavior (Achenbach, 1991).
The Current Study
Based on previous meta-analyses (Manuele et al., 2023; Vasquez et al., 2016; Yap & Jorm, 2015), the first research question asked whether there is, on average, a significant negative correlation of PAS with internalizing and externalizing problems. Given our focus on cultural similarities and differences, the second research question asked whether the size of the correlation of PAS with internalizing and externalizing problems varies between countries and regions. Based on the categorization of regions according to common cultural roots (Huntington, 1996; Spencer & Thomas, 1973), the present meta-analysis compared ten regions: North America, Central and South America, Western Europe (countries that were not part of the former communist bloc), Eastern Europe and Russia, North Africa and the Arabian Peninsula, Sub-Saharan Africa, Australia and New Zealand, East Asia (e.g., China, Japan), South Asia (e.g., India, Pakistan), and Southeast Asia (e.g., Indonesia, Thailand). These 10 regions were compared in a previous meta-analysis on cultural differences in the strength of the association between harsh parenting and child outcomes (Pinquart, 2021).
Based on the suggestion that PAS may be more culturally appropriate in individualistic societies than in collectivistic societies (Chao, 1994; Grusec, 2008; Rothbaum & Trommsdorff, 2007), the third research question asked whether stronger associations of PAS with children’s internalizing and externalizing problems would be found in more individualistic countries. Based on Zhai and Gao’s (2009) assumption, the fourth research question asked whether associations of PAS with internalizing and externalizing symptoms would be stronger in countries where power distance is less valued.
The final research question concerns the distinction of PI and PVF (Soenens et al., 2007). PVF may be culturally accepted in societies that value independence as well as in those that value interdependence. However, PI is more likely to be culturally accepted in societies that value independence and have a low power distance (Benito-Gomez et al., 2020; Soenens et al., 2007). Thus, the fifth research question asked whether moderating effects of individualism-collectivism and power distance on the association of PAS with internalizing and externalizing symptoms are stronger in studies on PI than in studies on PVF.
Method
Sample
Studies were found through the electronic databases PsycInfo, ERIC, Google Scholar, and PSYNDEX [search terms: autonomy AND (mother OR father OR parents) AND (internalizing problems OR depression OR anxiety OR externalizing problems OR delinquen* or aggress*). Reference sections of the identified papers were checked for additional studies. The final search was completed on December 31, 2025. The following criteria for inclusion of studies were applied:
1.
Correlations of autonomy support from mothers, fathers, or from both parents with internalizing and/or externalizing problems were reported or can be computed.
2.
The study assessed children, adolescents or college students.
3.
The study was completed before January 2026.
Studies were excluded if they.
4.
provided only multivariate associations of PAS with internalizing/externalizing symptoms as these effect sizes cannot be combined with bivariate effect sizes,
5.
used only a sum score that combined PAS with other parental behaviors, such as assessing parenting styles that combine parental warmth, control, and autonomy granting,
6.
used only a sum score of behavioral problems rather than differentiating between internalizing and externalizing problems.
To include studies from a broad range of countries, the included studies were not limited to those written in English. To minimize the effects of publication bias on the results, unpublished studies from databases (e.g., dissertations) were included.
The search identified 1440 records. After screening and assessing eligibility, 404 studies were included in the meta-analysis (Fig. 1). Selected characteristics of the included studies are reported as electronic supplements (ESM1 and ESM2). The review protocol is available from the first author on request. The following variables were entered: Number of children, mean age, percentage of female participants, percentage of members of ethnic minorities, country, region of data collection, method for assessing PAS, assessment of PI versus PVF (1 = PI, 2 = PVF, 3 = mix or insufficient information for coding), method for assessing internalizing and externalizing symptoms, support for validity of the PAS measure (2 = support provided in the individual study or other included studies, 1 = no support), support for validity of the behavior problem measures (2 = yes, 1 = no), monoinformant bias (i.e., possible overestimation of the effect size when information on parenting and child outcomes comes from the same respondent; 2 = yes, 1 = no), time of assessment (1 = all variables assessed at the same time, 2 = internalizing/externalizing symptoms assessed at a later time of measurement), publication status (2 = published, 1 = unpublished), and the size of bivariate association of PAS with internalizing/externalizing problems. The coding of PAS measures as focusing (mainly) on PVF versus PI was based on McCurdy et al. (2020). However, this coding was only possible for approximately 60% of the included studies. The first author coded all studies. The second author coded 40 studies. A mean inter-rater reliability of 90% (range 86% − 100%) was achieved, and discrepancies were resolved through discussion.
PAS was most often assessed with the Perceptions of Parents Scale (Grolnick et al., 1997; 44 studies), observations of parent-child interactions (39 studies), versions of the Parenting Styles Index (Lamborn et al., 1991; 29 studies), the Perceived PAS Scale (Mageau et al., 2015; 27 studies), the Parent-Child Relationship Inventory (Gerard, 1994; 21 studies), the Parents as Social Context Questionnaire (Skinner et al., 2005; 21 studies), the Parental Psychological Control and Autonomy Support Scale (Wang et al., 2007; 21 studies), the Mother-Father-Peer Scale (Epstein, 1983, 12 studies), and related measures (190 studies). Internalizing problems were most often assessed with the Child Behavior Checklist (CBCL; Achenbach, 1991; 56 studies), the Center for Epidemiological Studies Depression Scale (Radloff, 1977; 43 studies), the Children’s Depression Inventory (Kovacs, 1992; 23 studies), the Youth Self-Report (YSR; Achenbach, 1991; 21 studies), and related instruments (177 studies). Externalizing problems were assessed using the CBCL (48 studies), the YSR (23 studies), and related measures (121 studies).
Country-level scores of individualism-collectivism and power distance were extracted from Hofstede Insights (2025). Higher scores represent higher levels of individualism and power distance, respectively. Hofstede had originally provided data on national levels of individualism-collectivism as well as other cultural dimensions based on a survey of approximately 117,000 employees across 50 countries. Updated indices have been published by Hofstede et al. (2010) and on the Hofstede Insights electronic platform (2025). While these indices may not fully represent parents and children in individual countries and do not provide information on within-country variability, they offer a general overview of national differences in individualism-collectivism and power distance. Data on country-levels of individualism-collectivism and power distance were unavailable for two countries that provided data on the association between PAS and child problems: Barbados and Cyprus. Information on the gross domestic product (GDP) per capita was extracted from Eurostat (2025).
Statistical Analysis
The present study employed a hierarchical meta-analysis, acknowledging that individual effect sizes are nested within studies (e.g., if a study reports separate correlations for female and male participants), which are nested within countries, which are nested within regions (e.g., North America or Western Europe) (Cheung, 2014). This method allows researchers to split the variance of effect sizes into within-study variance (level 1), between-study variance within countries (level 2), between-country variance (level 3), and variance between regions of the globe (level 4). Hierarchical meta-analysis disentangles these sources of variability and analyzes whether significant heterogeneity exists at each level. We computed random-effects models assuming that the heterogeneity of effect sizes is based on random population effects and sampling error.
To analyse moderator effects, collectivism-individualism and power distance were introduced as predictors at the country level and the assessment of PI versus PVF was introduced as predictor at the effect size level. This is because separate effect sizes for PI and PVF may be reported for the same sample. To address possible publication bias, we compared results from published and unpublished studies and computed a trim-and-fill analysis (Duval & Tweedie, 2000). The data were analyzed using the metafor package in R (Viechtbauer, 2010).
Results
The included studies provided data from 53 countries. Most of the studies having been conducted in the United States (N = 125), China (N = 46), Canada (N = 30), Spain (N = 24), the Netherlands (N = 22), and Belgium (N = 21) (see ESM3 for the full list of countries). These studies provided data on 266,627 young people with a mean age of 13.99 years (SD = 3.51). Approximately 53.4% of the participants were female, and 99 of the included studies were unpublished. About 78.5% of the studies had collected data from one source, with 73.7% of them exclusively using child reports. In addition, about 61% of the studies used validated measures of PAS, while 88.5% used validated measures of child outcomes.
When analyzing associations of PAS with internalizing problems, we found a small-to-moderate, statistically significant negative correlation of r = − .18 (Table 1), which indicates 3.2% explained variance. Associations were stronger when focusing on PVF, in particular (r = .22, 4.8% explained variance). The Q-test indicates significant heterogeneity, with the largest amount of variance between studies of individual countries (69.7%), followed by the variance within studies (sampling variance; 22.1%), and variance between countries (8.0%). There was no variance between the ten regions of the globe (0%). We used four log-likelihood-ratio tests (LRT) to examine the significance of the variance components (Assink & Wibbelink, 2016). In these tests, the original model, in which all variance components are freely estimated, is compared with a model in which one variance component each is fixed to zero. The within-study variance component (LRT = 242.81, p < .001), the between-study variance component (LRT = 152.31, p < .001), and the between-country variance component (LRT = 7.26, p < .01) were significant. In contrast, the between-regions variance component was nonsignificant (LRT = 0.00, p = 1.00). Separate analyses were computed for each country if at least three effect sizes were available, which was the case for 23 countries. We found significant associations of PAS with internalizing symptoms in 18 countries (see ESM4) and in nine regions of the globe (except Subsaharan Africa where test power was restricted; see ESM5).
Table 1
Associations of parental autonomy support with internalizing and externalizing problems in the offspring
Internalizing problems
Externalizing problems
r
95%-CI
df
t
Q
r
95%-CI
df
t
Q
All studies
− 0.18
− 0.20
− 0.16
764
-17.85***
5804.04***
− 0.13
− 0.16
− 0.10
346
-10.10***
2544.01***
PI
− 0.13
− 0.17
− 0.09
151
-6.90***
972.27***
− 0.07
− 0.10
− 0.03
91
-3.87***
773.00***
PVF
− 0.22
− 0.25
− 0.19
328
-14.20***
1920.76***
− 0.18
− 0.21
− 0.15
87
-11.81***
470.03***
r = weighted mean correlation, CI = confidence interval, df = degrees of freedom, t = test for significance of the weighted mean effect size. Q = test for heterogeneity of effect sizes. PI = promotion of individuality, PVF = promotion of volitional function. ***p < .001
The association of PAS with externalizing problems has been less often examined than the association with internalizing problems (346 vs. 764 available effect sizes). A significant negative correlation was identified between PAS and externalizing problems (r = − .13, Table 1), indicating 1.7% explained variance. The Q-test indicated significant heterogeneity. Again, the largest amount of variance was found between studies of the individual countries (63.4%), followed by within-study variance (27.4%), and between-country variance (9.2%), while regional variance was found to be negligible (0%). The LRT indicated that the within-study variance component (LRT = 101.22, p < .001) and the between-study variance component (LRT = 41.57, p < .001) were statistically significant while the between-country variance component (LRT = 2.29, p < .32) and the between-region variance component (LRT = 0.00, p < 1.00) were not. A minimum of three effect sizes were available for 19 countries, and significant associations of PAS with externalizing symptoms were found in eight countries (Belgium, Canada, China, Israel, Italy, Mexico, Spain, and the USA; ESM4). When comparing regions of the globe, significant negative associations of PAS with externalizing symptoms were found for five out of ten regions (North America, Western Europe, Central/South America, East Asia, and Southeast Asia; see ESM5).
The trim-and-fill algorithm added 66 possibly missing effect sizes in the analysis of associations of PAS with internalizing symptoms, and the weighted mean effect size slightly increased (rcorr = − 0.20, CI − 0.22 to − 0.18, Z = -19.18, p < .001). In the analysis of externalizing problems, three potentially missing effect sizes were incorporated; however, the weighted mean effect size remained constant.
Moderating Effects
Before analyzing moderating effects of individualism-collectivism and power distance, we checked whether there were related differences in the study variables that could confound the results. Studies from countries with higher national individualism scores had older samples (r = .12, p < .001), higher percentage of female participants (r = .09, p < .007), were less likely to use the same informant for assessing PAS and internalizing/externalizing problems (monoinformant bias; r = − .17, p < .001), to be published (r = − .19, p < .001), to collect cross-sectional data (r = − .13, p < .001), and to assess PI rather than PVF (r = − .25, p < .001). Studies from countries with higher power distance scores had used older samples (r = .15, p < .001), more the same informant (r = .15, p < .001) and the same time of measurement (r = .09, p < .004). These studies were also more likely published (r = .24, p < .001), and more often assessed PI rather than PVF (r = .29, p < .001). No potential confounding effects were identified for the validity of the PAS and the child symptoms measures. Finally, country levels of individualism-collectivism and power distance were correlated with GDP per capita (r = .72 and r = − .62, p’s < 0.001).
Next, we examined whether the size of association of PAS with internalizing and externalizing symptoms varied by these study variables. Given the large correlation between country-levels of individualism-collectivism and power distance (r = − .65, p < .001), we followed the suggestion of Hox (2010) to analyze the moderating effects of individualism-collectivism and power distance separately before including them in a common analysis. The national levels of individualism-collectivism and power distance did not exhibit any substantial univariate moderating effects (Table 2). In contrast, correlations of PAS with internalizing symptoms were stronger in older samples, in studies with a higher percentage of female participants, with monoinformant bias, cross-sectional rather than longitudinal design, the use of a validated PAS measure, in studies that assessed PVF rather than PI, and in countries with lower GDP per capita. Studies with higher percentages of girls, utilizing a cross-sectional design, and assessing PVF found stronger negative correlations between PAS and externalizing symptoms.
Table 2
Univariate moderating effects of study and sample characteristics on the association of parental autonomy support with internalizing and externalizing problems in the offspring
Moderator variable
Internalizing problems
Externalizing problems
estimate
CI
df
Z
estimate
CI
df
Z
Age
− 0.0079
− 0.0106
− 0.0053
740
-5.86***
− 0.0039
− 0.0080
0.0003
332
-1.81
% female children
− 0.0008
− 0.0013
− 0.0002
713
-2.85**
− 0.0009
− 0.0016
− 0.0002
314
-2.37*
Monoinformant bias (2 = yes, 1 = no)
− 0.1068
− 0.1295
− 0.0841
762
-9.24***
− 0.0296
− 0.0632
0.0040
344
-1.73
Time of assessment (1 = concurrent, 2 = longitudinal)
0.0349
0.0149
0.0549
763
3.43***
0.0438
0.0084
0.0791
345
2.44*
Validity support: PAS (2 = yes, 1 = no)
− 0.0527
− 0.0787
− 0.0268
763
-3.99***
0.0072
− 0.0281
0.0425
345
0.40
Validity support: psych. probl. (2 = yes, 1 = no)
0.0032
− 0.0370
0.0435
763
0.16
− 0.0172
− 0.0589
0.0245
345
− 0.81
Published (2 = yes, 1 = no)
− 0.0151
− 0.0467
0.0165
763
− 0.94
− 0.0044
− 0.0467
0.0380
345
− 0.20
PVF (2) vs. PI (1)
− 0.0807
− 0.1089
− 0.0524
479
-5.61***
− 0.1102
− 0.1567
− 0.0537
178
-4.68***
GDP per capita
0.0009
0.0000
0.0018
762
1.97*
− 0.0000
− 0.0012
0.0011
344
− 0.08
Individualism (country level)
0.0007
− 0.0000
0.0013
761
1.91
0.0004
− 0.0006
0.0013
344
0.77
Power distance (country level)
− 0.0008
− 0.0018
0.0002
761
-1.55
− 0.0001
− 0.0015
0.0013
344
− 0.14
CI = 95%-confidence interval, df = degree of freedom, Z = testing for significance of the moderating effect. *p < .01, **p < .01, ***p < .001
To assess the final research question, two meta-regression analyses were conducted, incorporating interaction terms between PVF versus PI and individualism-collectivism as well as power distance. As demonstrated in Table 3, a statistically significant interaction effect was observed between power distance and PI/PVF. National levels of power distance did not moderate the association of PVF with internalizing problems. In contrast, the association of PI with internalizing problems became weaker (less negative) when the national level of power distance increased (estimate 0.0032, Z = 2.30, p < .025). Our findings revealed no interaction effects of PI/PVF with individualism-collectivism. The multivariate analyses confirmed the moderating effects of study design, monoinformant bias and PI/PVS from univariate analyses.
Table 3
Multivariate analysis of moderating effects on the association of parental autonomy support with internalizing and externalizing problems in the offspring
Moderator variable
Internalizing problems
Externalizing problems
estimate
CI
df
Z
estimate
CI
df
Z
Age
− 0.0035
− 0.0074
0.0004
367
-1.78
0.0026
− 0.0050
0.0103
167
0.69
% female children
− 0.0005
− 0.0011
0.0001
367
-1.80
− 0.0004
− 0.0014
0.0006
167
− 0.76
Monoinformant bias (2 = yes, 1 = no)
− 0.1037
− 0.1348
− 0.0726
367
-6.56***
− 0.0513
− 0.1005
− 0.0020
167
-2.06*
Time of assessment (1 = concurrent, 2 = longitudinal)
0.0330
0.0077
0.0582
367
2.56*
0.0424
− 0.0073
0.0921
167
1.68
Validity support: PAS (2 = yes, 1 = no)
0.0149
− 0.0497
0.0199
367
0.84
− 0.0187
− 0.0751
0.0378
167
− 0.65
PVF (2) vs. PI (1)
− 0.0604
− 0.0899
− 0.0309
367
-4.03***
− 0.1018
− 0.1528
− 0.0508
167
-3.94***
GDP per capita
0.0003
− 0.0020
0.0026
367
0.27
Individualism (country level)
− 0.0000
− 0.0017
0.0017
367
− 0.02
0.0004
− 0.0020
0.0028
167
0.34
Power distance (country level)
0.0006
− 0.0013
0.0025
367
0.62
0.0003
− 0.0028
0.0035
167
0.21
Individualism x PVF
− 0.0003
− 0.0020
0.0013
367
− 0.40
− 0.0013
− 0.0045
0.0020
167
− 0.77
Power distance x PVF
− 0.0032
− 0.0059
− 0.0005
367
-2.30*
− 0.0039
− 0.0092
0.0014
167
-1.46
Intercept
− 0.0160
− 0.1969
0.1650
367
− 0.17
− 0.0373
− 0.3451
0.2705
167
− 0.24
CI = 95%-confidence interval, df = degree of freedom, Z = testing for significance of the moderating effect. Only covariates that were significant in univariate analysis were included in multivariate analysis. *p < .05, ***p < .001
Discussion
The present meta-analysis found, on average, small-to-moderate negative associations of PAS with internalizing and externalizing problems in the offspring with size of the association showing some variability between countries. Country levels of individualism-collectivism did not explain between-country variability in the size of association of PAS with child problems. However, an interaction effect was observed between country level of power distance and operationalization of PAS, with associations of PI rather than PVF with internalizing symptoms becoming weaker in less individualistic countries. Additional main effects also indicated stronger associations of PAS with the two kinds of child problems if PVF rather than PI was assessed. Finally, multivariate analyses revealed moderating effects of gender and mono-informant bias.
Despite the absence of a consensus on the definition of the practical significance of findings, the explained variance of 3.2% and 1.7% (association of PAS with internalizing and externalizing problems) indicates that PAS has a small to moderate practical relevance (Cohen, 1992). When employing the binomial effect size display (Rosenthal & Rubin, 1982), a correlation of 0.22 would indicate that, when splitting the children into those with above and below median levels of internalizing problems, 41% of the former children would be in the group with autonomy support above the median, compared to 59% of those with below-median autonomy support—a 18% difference.
In accordance with SDT (Marbell-Pierre et al., 2019; Ryan & Deci, 2019), the present meta-analysis shows only slight between-country variability in the associations between PAS and internalizing as well as externalizing problems. Interestingly, there was stronger heterogeneity of the results within countries than between countries, which was, in part, explained by child age, gender, and use of the same versus different data sources. A cross-national study by Lansford et al. (2021) found stronger variability of individualistic and collectivistic values within countries than between countries, and related differences in cultural values between samples within individual countries might be another source of the present within-country heterogeneity.
The low between-country heterogeneity of the results may indicate that autonomy is a universal human need, particularly regarding the sense of volition and being a causal agent of one’s own life (Ryan & Deci, 2017). In accordance with suggestions by Marbell-Pierre et al. (2017), a moderating effect of cultural values was exclusively found in studies that assessed PI as opposed to PVF. In countries where power hierarchies within the family are more culturally accepted, parental attempts to promote independence in their children may be perceived as a lack of regard for the offspring rather than as a source of positive development.
The present study did not identify moderating effects of individualism-collectivism. Despite the finding that adults from collectivistic societies may prioritize different socialization goals than their more individualistic counterparts (Mone et al., 2016) and exhibit less autonomy promotion toward their offspring (Lekes et al., 2010; Supple et al., 2009), these differences did not result in a discernible weakening of the associations between PAS and internalizing and externalizing symptoms. However, an examination of the 25% of studies from countries with the lowest individualism scores revealed that nearly 85% of the effect sizes referred to PVF rather than PI. Therefore, PI, which is less culturally appropriate in highly collectivist countries, may have been assessed too infrequently in these countries, thus precluding the identification of a moderating effect of individualism-collectivism. Additionally, when interpreting the absence of moderating effects of individualism-collectivism, it is crucial to acknowledge that the data on national levels of individualism-collectivism and on the association of parental autonomy with child outcomes originated from distinct samples. Thus, it remains open whether moderating effects of individualism-collectivism would have been found when assessing these values in the participating children and parents.
A comparative analysis of research on PVF versus PI reveals that stronger associations with internalizing and externalizing problems are observed in the former studies. Volitional function is suggested to promote the fulfillment of a basic need which leads to positive emotions and reduces negative feelings (Grolnick, 2012). Need fulfillment may, again, promote self-control and resistance against peer pressure, thus reducing the risk of externalizing problem behavior (Matte-Gagné & Bernier, 2011; Vansteenkiste & Ryan, 2013). In contrast, aspects of independence tend to show inconsistent associations with mental health, and one of them – emotionally distancing from parents – even predicts an increase in psychological symptoms over time, potentially attributable to the loss of parental support (Pinquart & Pfeiffer, 2020).
The moderating effects of other study characteristics are only briefly discussed because of not being a central focus in the present meta-analysis. Regarding the stronger associations of PAS with psychological problems in samples with higher percentages of girls, this gender difference can be explained by the higher levels of concern shown by girls regarding interpersonal relationships and their higher interpersonal sensitivity compared to boys, as reported by Chen et al. (2019). The observed stronger associations of PAS with internalizing and externalizing problems in studies with a single informant (i.e., child or parent) can be explained by shared measurement errors. Stronger associations in cross-sectional as compared to longitudinal studies indicate stronger short-term as compared to long-term statistical effects. Although older children may seek greater autonomy and might benefit more from PAS (Qin et al., 2009), a moderating effect of child age was only found in univariate analysis. This univariate moderating effect is likely attributable to the fact that studies with older children more often assessed PVF rather than PI and more often used only one informant (McCurdy et al., 2020). A moderating effect of the methodological quality of assessing PAS was only found in univariate analysis, probably relating to limited validity checks of (observational) measures used in families with young children and the weaker associations of PAS with internalizing problems reported in studies with younger children.
As published and unpublished studies did not differ in their results and as trim-and-fill analysis did not find evidence for an overestimation of effect sizes, we conclude that our results were not influenced by a file-drawer problem.
Limitations and Conclusions
Several limitations must be mentioned. First, although studies from 53 countries were included, no studies from more than 140 countries could be located. Second, we analyzed bivariate correlations of PAS with internalizing and externalizing problems that do not allow for drawing causal conclusions. While a limited number of studies have documented cross-lagged associations, the overall number remains inadequate for conducting cross-cultural comparisons. Third, the mean size of associations of PAS with child outcomes may have been inflated by shared methods variance as most studies used only one source of information (in most cases the child). Nonetheless, child reports on parenting tend be less influenced by social desirability than parent reports (Bögels & van Melick, 2004), and child reports on emotional problems tend to be more valid than parental reports (Howells Wrobel & Lachar, 1998). Thus, the frequent use of child reports might not be interpreted as source of bias.
Fourth, as previously stated, data concerning individualism-collectivism and power distance were derived from samples other than those used for the PAS and child problems data. There is some evidence that national levels of individualism-collectivism and power distance may change over time (Beugelsdijk & Welzel, 2018; Ogihara, 2018) which would reduce the predictive power of the scores from Hofstede et al. (2010) when the interval between assessment of cultural values and PAS increases. Given the variability of cultural values within countries (Lansford et al., 2021), it would have been advantageous to incorporate data from parents and/or children regarding their agreement with these values in the analysis.
Fifth, substantial correlations between PI and PVF indicate the difficulty of measuring both variables as distinct, nonoverlapping constructs (Barberis et al., 2021; Liga et al., 2020). Even though the comparative analysis of PI versus PVF incorporated measures with a pronounced emphasis on PI or PVF (McCurdy et al., 2020), it was not possible to achieve complete separation between these variables (e.g., through the analysis of item-level correlations). However, the overlap of the measured constructs would not provide a satisfactory explanation for the observed differential results on PI and PVF. Finally, the moderator analysis was constrained by the uneven availability of studies on PI versus PVF when analyzing associations with internalizing problems.
Despite these limitations, the following conclusions can be drawn. First, we conclude that PAS has statistically small-to-moderate, but practically meaningful statistical effects, while PVF demonstrates a more pronounced correlation with behavior problems in the offspring when compared to PI. Second, the present study contributes to the resolution of the controversy surrounding the universality versus cultural specificity of associations of PAS (Grusec, 2008; Ryan & Deci, 2019; Soenens et al., 2007) by demonstrating that there are more similarities and that moderating effects of power distance refer to PI rather than PVF. However, the available data precluded the testing of causal effects. A stronger conclusion about universal statistical effects would necessitate the inclusion of a significantly larger number of countries in the evaluation and recording cultural values of the participants and in representative samples. To examine the role of culture in the relationship between PAS and behavioral problems more closely, a large-scale, cross-national longitudinal study is recommended in which the cultural values of participating parents and children are assessed, overlaps between PI and PVF are minimized, and various informants are used on parenting and child outcomes.
Third, given the stronger correlations of internalizing and externalizing problems with PVF than with PI, the meta-analysis indicates that parents would be more recommended to promote volitional functioning in the offspring. PI would be less accepted in cultures that place a higher value on power distance within the family unit. Two parenting interventions – the Parent Check-In (Grolnick et al., 2021) and the How-to Parenting program (Joussemet et al., 2014; Mageau et al., 2022) have been found to increase PAS, enhance psychological well-being and reduce behavior problems in the offspring. Given the only small-to-moderate associations of PAS with emotional and behavioral problems, we suggest that parenting interventions should combine the promotion of PAS with the promotion of other desirable parental behaviors.
Acknowledgements
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Achenbach, T. M. (1991). Integrative guide for the 1991 CBCL/4 ± 18, YSR, and TRF profiles. University of Vermont.
Assink, M., & Wibbelink, C. J. M. (2016). Fitting three-level meta-analytic models in R: A step-by-step tutorial. The Quantitative Methods for Psychology, 12(3), 154–174. https://doi.org/10.20982/tqmp.12.3.p154CrossRef
Barberis, N., Gugliandolo, M. C., Costa, S., & Liga, F. (2021). How parental autonomy support prevents from adolescents’ depression and low self-esteem: A mediational model with trait emotional intelligence. Mediterranean Journal of Clinical Psychology, 9(1), 1–18. https://doi.org/10.6092/2282-1619/mjcp-2898CrossRef
Basabe, N., Paez, D., Valencia, J., Gonzalez, J. L., Rimé, B., & Diener, E. (2002). Cultural dimensions, socioeconomic development, climate, and emotional hedonic level. Cognition & Emotion, 16(1), 103–125. https://doi.org/10.1080/02699930143000158CrossRef
Benito-Gomez, M., Williams, K. N., McCurdy, A., & Fletcher, A. C. (2020). Autonomy-supportive parenting in adolescence: Cultural variability in the contemporary United States. Journal of Family Theory & Review, 12(1), 7–26. https://doi.org/10.1111/jftr.12362CrossRef
Blos, P. (1979). The adolescent passage. International Universities.
Bögels, S. M., & van Melick, M. (2004). The relationship between child-report, parent self-report, and partner report of perceived parental rearing behaviors and anxiety in children and parents. Personality and Individual Differences, 37(8), 1583–1596. https://doi.org/10.1016/j.paid.2004.02.014CrossRef
Chao, R. (1994). Beyond parental control and authoritarian parenting style: Understanding Chinese parenting through the cultural notion of training. Child Development, 65, 1111–1119. https://doi.org/10.2307/1131308CrossRefPubMed
Chen, Y., Li, R., & Liu, X. (2019). The relationships among parental psychological control/autonomy support, self-trouble, and internalizing problems across adolescent genders. Scandinavian Journal of Psychology, 60, 539–547. https://doi.org/10.1111/sjop.12573CrossRefPubMed
Cheung, W. L. (2014). Modeling dependent effect sizes with three-level meta-analyses: A structural equation modeling approach. Psychological Methods, 19(2), 211–229. https://doi.org/10.1037/a0032968CrossRefPubMed
Chirkov, V. I., & Ryan, R. M. (2001). Parent and teacher autonomy-support in Russian and U.S. adolescents: Common effects on well-being and academic motivation. Journal of Cross-Cultural Psychology, 32, 618–635. https://doi.org/10.1177/0022022101032005006CrossRef
Deci, E. L., & Ryan, R. M. (2012). Motivation, personality, and development within embedded social contexts: an overview of self-determination theory. In R. M. Ryan (Ed.), The Oxford handbook of human motivation (pp. 85–107). Oxford University Press.
Filus, A., Schwarz, B., Mylonas, K., Sam, D. L., & Boski, P. (2019). Parenting and late adolescents’ well-being in Greece, Norway, Poland and Switzerland: Associations with individuation from parents. Journal of Child and Family Studies, 28, 560–576. https://doi.org/10.1007/s10826-018-1283-1CrossRef
Gerard, A. B. (1994). Parent-Child Relationship Inventory (PCRI): Manual. Western Psychological Services.
Grolnick, W. S. (2012). The relations among parental power assertion, control, and structure. Human Development, 55, 57–64. https://doi.org/10.1159/000338533CrossRef
Grolnick, W. S., Deci, E. L., & Ryan, R. M. (1997). Internalization within the family: The self-determination theory perspective. In J. E. Grusec, & L. Kuczynski (Eds.), Parenting and children’s internalization of values: A handbook of contemporary theory (pp. 135–161). Wiley.
Grolnick, W. S., Levitt, M. R., Caruso, A. J., & Lerner, R. E. (2021). Effectiveness of a brief preventive parenting intervention based in self-determination theory. Journal of Child and Family Studies, 30, 905–920. https://doi.org/10.1007/s10826-021-01908-4CrossRef
Grusec, J. E. (2008). What is the nature of effective parenting: It depends. In M. Kerr, H. Stattin, & R. Engels (Eds.), What can parents do? New insights in the role of parents in adolescent problem behavior (pp. 239–257). Wiley.
Havighurst, R. J. (1972). Developmental tasks and education. Longman.
Hofstede, G., Hofstede, G. J., & Minkov, M. (2010). Cultures and organizations: Software of the mind (3rd ed.). McGraw-Hill.
Howells Wrobel, N., & Lachar, D. (1998). Validity of self-and parent‐report scales in screening students for behavioral and emotional problems in elementary school. Psychology in the Schools, 35(1), 17–27. https://doi.org/10.1002/(SICI)1520-6807::AID-PITS2>3.0.CO;2-R.CrossRef
Hox, J. J. (2010). Multilevel analysis: Techniques and applications. Routledge.
Huntington, S. P. (1996). The clash of civilizations and the remaking of world order. Simon & Schuster.
Joussemet, M., Mageau, G. A., & Koestner, R. (2014). Promoting optimal parenting and children’s mental health: A preliminary evaluation of the how-to parenting program. Journal of Child and Family Studies, 23(6), 949–964. https://doi.org/10.1007/s10826-013-9751-0CrossRef
Kovacs, M. (1992). Children’s Depression Inventory. Multi-Health Systems, Inc.
Lamborn, S. D., Mounts, N. S., Steinberg, L., & Dornbusch, S. M. (1991). Patterns of competence and adjustment among adolescents from authoritative, authoritarian, indulgent, and neglectful families. Child Development, 62(5), 1049–1065. https://doi.org/10.2307/1131151CrossRefPubMed
Lansford, J. E., Zietz, S., Al-Hassan, S. M., Bacchini, D., Bornstein, M. H., Chang, L., & Alampay, L. P. (2021). Culture and social change in mothers’ and fathers’ individualism, collectivism and parenting attitudes. Social Sciences, 10(12), 459. https://doi.org/10.3390/socsci10120459CrossRefPubMedPubMedCentral
Lekes, N., Gingras, I., Philippe, F. L., Koestner, R., & Fang, J. (2010). Parental autonomy-support, intrinsic life goals, and well-being among adolescents in China and North America. Journal of Youth and Adolescence, 39(8), 858–869. https://doi.org/10.1007/s10964-009-9451-7CrossRefPubMed
Liga, F., Lo Coco, A., Musso, P., Inguglia, C., Costa, C., Cricchio, L., M.G., & Ingoglia, S. (2020). Parental psychological control, autonomy support and Italian emerging adult’s psychosocial well-being: A cluster analytic approach. European Journal of Developmental Psychology, 17(1), 37–55. https://doi.org/10.1080/17405629.2018.1532887CrossRef
Mageau, G. A., Ranger, F., Joussemet, M., Koestner, R., Moreau, E., & Forest, J. (2015). Validation of the Perceived Parental Autonomy Support Scale (P-PASS). Canadian Journal of Behavioural Science, 47(3), 251–262. https://doi.org/10.1037/a0039325CrossRef
Mageau, G. A., Joussemet, M., Paquin, C., & Grenier, F. (2022). How-to-Parenting-Program: Change in parenting and child mental health over one year. Journal of Child and Family Studies, 31, 3498–3513. https://doi.org/10.1007/s10826-022-02442-7CrossRef
Mahler, M. S., Pine, F., & Bergman, A. (1975). The psychological birth of the human infant: Symbiosis and individuation. Basic Books.
Manuele, S. J., Yap, M. B. H., Lin, S. C., Pozzi, E., & Whittle, S. (2023). Associations between paternal versus maternal parenting behaviors and child and adolescent internalizing problems: A systematic review and meta-analysis. Clinical Psychology Review, 105, 102339. https://doi.org/10.1016/j.cpr.2023.102339CrossRefPubMed
Marbell-Pierre, K. N., Grolnick, W. S., Stewart, A. L., & Raftery-Helmer, J. N. (2019). Parental autonomy support in two cultures: The moderating effects of adolescents’ self-construals. Child Development, 90, 825–845. https://doi.org/10.1111/cdev.12947CrossRefPubMed
Matte-Gagné, C., & Bernier, A. (2011). Prospective relations between maternal autonomy support and child executive functioning: Investigating the mediating role of child language ability. Journal of Experimental Child Psychology, 110, 611–625. https://doi.org/10.1016/j.jecp.2011.06.006CrossRefPubMed
McCurdy, A. L., Williams, K. N., Lee, G. Y., Benito-Gomez, M., & Fletcher, A. C. (2020). Measurement of parental autonomy support: A review of theoretical concerns and developmental considerations. Journal of Family Theory and Review, 12, 382–397. https://doi.org/10.1016/10.1111/jftr.12389CrossRef
Mone, I. S., Benga, O., & Opre, A. (2016). Cross-cultural differences in socialization goals as a function of power distance, individualism-collectivism, and education. Romanian Journal of Experimental Applied Psychology, 7(s 1), 330–334.
Ogihara, Y. (2018). Economic shifts and cultural changes in individualism: A cross-temporal perspective. In A. Uskul, & S. Oishi (Eds.), Socioeconomic environment and human psychology: Social, ecological, and cultural perspectives (pp. 247–270). Oxford University Press.
Qin, L. L., Pomerantz, E. M., & Wang, Q. (2009). Are gains in decision-making autonomy during early adolescence beneficial for emotional functioning? The case of the United States and China. Child Development, 80, 1705–1721. https://doi.org/10.1111/j.1467-8624.2009.01363.xCrossRefPubMed
Radloff, L. S. (1977). The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 385–401. https://doi.org/10.1177/014662167700100306CrossRef
Rosenthal, R., & Rubin, D. B. (1982). A simple general purpose display of magnitude and experimental effect. Journal of Educational Psychology, 74, 166–169. https://doi.org/10.1037/0022-0663.74.2.166CrossRef
Rothbaum, F., & Trommsdorff, G. (2007). Do roots and wings complement or oppose one another? The socialization of relatedness and autonomy in cultural context. In J. E. Grusec, & P. D. Hastings (Eds.), Handbook of socialization: Theory and research (pp. 461–489). Guilford.
Ryan, R. M., & Deci, E. L. (2017). Self-determination theory: Basic psychological needs in motivation, development, and wellness. The Guilford Press. https://doi.org/10.1521/978.14625/28806
Ryan, R. M., & Deci, E. L. (2019). Brick by brick: The origins, development, and future of self-determination theory. In A. J. Elliot (Ed.), Advances in motivation science (Vol. 6, pp. 111–156). Elsevier.
Soenens, B., Vansteenkiste, M., Lens, W., Luyckx, K., Goossens, L., Beyers, W., & Ryan, R. M. (2007). Conceptualizing parental autonomy support: Adolescent perceptions of promotion of independence versus promotion of volitional functioning. Developmental Psychology, 43(3), 633–646. https://doi.org/10.1037/0012-1649.43.3.633CrossRefPubMed
Soenens, B., Vansteenkiste, M., Van Petegem, S., Beyers, W., & Ryan, R. (2018). How to solve the conundrum of adolescent autonomy? In B. Soenens, M. Vansteenkiste, & S. Petegem (Eds.), Autonomy in adolescent development: Towards conceptual clarity (pp. 1–32). Psychology Press. https://doi.org/10.4324/9781315636511-1
Spencer, J. E., & Thomas, W. L. (1973). Introducing cultural geography. Wiley.
Supple, A. J., Ghazarian, S. G., Peterson, G. W., & Bush, K. R. (2009). Assessing the cross-cultural validity of a parental autonomy granting measure: Comparing adolescents in the U.S., China, Mexico and India. Journal of Cross-Cultural Psychology, 5, 816–833. https://doi.org/10.1177/0022022109339390CrossRef
Van Petegem, S., Vansteenkiste, M., & Beyers, W. (2013). The jingle–jangle fallacy in adolescent autonomy in the family: In search of an underlying structure. Journal of Youth and Adolescence, 42(7), 994–1014. https://doi.org/10.1007/s10964-012-9847-7CrossRefPubMed
Van Petegem, S., Zimmer-Gembeck, M., Baudat, S., Soenens, B., Vansteenkiste, M., & Zimmermann, G. (2019). Adolescents’ responses to parental regulation: The role of communication style and self-determination. Journal of Applied Developmental Psychology, 65, 101073. https://doi.org/10.1016/j.appdev.2019.101073CrossRef
Vansteenkiste, M., & Ryan, R. M. (2013). On psychological growth and vulnerability: Basic psychological need satisfaction and need frustration as a unifying principle. Journal of Psychotherapy Integration, 23(3), 263–280. https://doi.org/10.1037/a0032359CrossRef
Vasquez, A. C., Patall, E. A., Fong, C. J., Corrigan, A. S., & Pine, L. (2016). Parent autonomy support, academic achievement, and psychosocial functioning: A meta-analysis of research. Educational Psychology Review, 28, 605–644. https://doi.org/10.1007/s10648-015-9329-zCrossRef
Wang, Q., Pomerantz, E. M., & Chen, H. (2007). The role of parents’ control in early adolescents’ psychological functioning: A longitudinal investigation in the United States and China. Child Development, 78, 1592–1610. https://doi.org/10.1111/j.1467-8624.2007.01085.xCrossRefPubMed
Xiong, Y., Qin, L., Wang, Q., Wang, M., & Pomerantz, E. M. (2022). Reexamining the cultural specificity of controlling and autonomy-supportive parenting in the United States and China with a within-individual analytic approach. Developmental Psychology, 58(5), 935–949. https://doi.org/10.1037/dev0001329CrossRefPubMed
Yap, M. B. H., & Jorm, A. F. (2015). Parental factors associated with childhood anxiety, depression, and internalizing problems: A systematic review and meta-analysis. Journal of Affective Disorders, 175, 424–440. https://doi.org/10.1016/j.jad.2015.01.05CrossRefPubMed
Zhai, F., & Gao, Q. (2009). Child maltreatment among Asian Americans: Characteristics and explanatory framework. Child Maltreatment, 14(2), 207–224. https://doi.org/10.1177/1077559508326286CrossRef
Pinquart, M. (2021). Cultural similarities and differences in the association of harsh parenting with internalizing and externalizing symptoms: A meta-analysis. Journal of Child and Family Studies, 30(12), 2938–2951. https://doi.org/10.1007/s10826-021-02113-zCrossRef
Pinquart, M., & Pfeiffer, J. P. (2020). Longitudinal associations of the attainment of developmental tasks with psychological symptoms in adolescence: A meta-analysis. Journal of Research on Adolescence, 30(S1), 4–14. https://doi.org/10.1111/jora.1246CrossRefPubMed
