Self-reported maternal expectations and child-rearing practices

Abstract

This study aimed at: (1) disentangling the associations between ethnicity, immigration, educational background, and mothers’ developmental expectations and (self-reported) child-rearing practices; and (2) identifying the cross-cultural differences and similarities in developmental expectations and child-rearing practices. Participants were 111 Dutch and 111 Turkish immigrant mothers in the Netherlands, and 242 Turkish mothers living in Turkey. Dutch and higher-educated mothers had a tendency to believe that children learn certain skills and behaviors at an earlier age than did Turkish and lower-educated mothers, respectively. Turkish mothers, majority group, and higher-educated mothers reported more child-centered parenting practices than Dutch mothers, immigrants, and mothers with less education, respectively. Parent-centered parenting practices were reported mainly by less educated mothers. The analyses on disentangling the associations between sociodemographic background variables and parenting pointed to the relative importance and consistency of maternal education as a predictor of parenting, compared to ethnic background and immigration history. It is concluded that disentangling variables that are often associated with studies comparing immigrant and majority groups is essential for a proper understanding of similarities and differences in developmental expectations and child-rearing practices.

Keywords

culture developmental expectations disentangling Dutch parenting practices Turkish

Parental ethnotheories are belief systems that parents hold regarding the nature of children and development which reflect parents’ developmental expectations, child-rearing values, and perceptions about their children (Harkness, Super, & Mavridis, 2011). It is well known that mothers’ child-rearing beliefs and practices have strong associations with their culture, socioeconomic status, and immigration background (Cheah & Chirkov, 2008; Durgel, Leyendecker, Yagmurlu, & Harwood, 2009; Yagmurlu & Sanson, 2009a); however, most studies examine these factors in isolation without addressing their relations or relative influence on parenting (Atkinson, 1994). For example, comparisons of ethnotheories of immigrant and majority group mothers may confound cultural and educational differences, as the latter group is usually better educated.

The present study set out to examine the associations among ethnicity, immigration, educational background, and mothers’ developmental expectations and child-rearing practices. Disentangling the relative importance of these factors can be expected to significantly increase our understanding of parenting beliefs and practices and to inform policies and programs regarding child development and family dynamics. In addition, we aimed to identify cross-cultural similarities and differences in developmental expectations and parenting practices of ethnic Dutch and Turkish-Dutch mothers living in the Netherlands and Turkish mothers living in Turkey. Finally, the present study investigated the relation between mothers’ age expectations regarding development of certain skills and behaviors and their child-rearing practices.

Maternal beliefs and developmental expectations

Parents’ beliefs and expectations concerning child development shape the ways they guide their children (Harkness et al., 2011). In this study we were interested in mothers’ developmental expectations which refer to maternal ideas regarding at what age a child can develop certain skills and can learn certain behaviors (Goodnow, Cashmore, Cotton, & Knight, 1984). Most of the studies on developmental expectations focus either on mothers’ knowledge regarding child development or mothers’ reports of their own children’s developmental milestones. However, instead of actual knowledge of parents, we were interested in the cultural patterning of mothers’ beliefs and assumptions regarding child development. The earliest cross-cultural studies on parental beliefs showed that mothers who are from collectivistic cultures (Triandis, 1994) expect earlier development in areas such as compliance with adult authority and late acquisition of skills like standing up for own rights and verbal assertion, compared to mothers from individualistic backgrounds (Goodnow et al., 1984; Hess, Kashiwagi, Azuma, Price, & Dickson, 1980). Based on these findings, authors argued that the behaviors which are deemed more important in a culture will be expected to develop earlier in that culture. Willemsen and Van de Vijver (1997) also examined developmental expectations of Dutch, Turkish-Dutch, and Zambian mothers. They found support for domain dependence; for example, large effect sizes were found for the cognitive domain, for which the Dutch mothers expected earlier ages of development. However, also in other domains Dutch mothers reported the youngest ages of development, though with smaller effect sizes. The authors argued that Dutch mothers had the earliest expectations across all domains because these mothers, due to their individualistic background, see children with a global potential to develop earlier than parents from non-Western backgrounds.

It appears that mothers’ expectations vary with their education, occupation, and income (Hess et al., 1980; von der Lippe, 1999). Mothers from higher socioeconomic backgrounds and from Western, individualistic backgrounds tend to indicate earlier ages of mastery than mothers from lower socioeconomic background, regardless of the type of behavior concerned. Research shows that socioeconomic factors explain more variance in mothers’ expectations than culture does (Willemsen & van de Vijver, 1997); however, the roles of culture and socioeconomic factors with parental expectations are only partly overlapping. The background variables that have been studied in the literature are often confounded, which makes it impossible to draw precise conclusions about the relative contribution of cultural and socioeconomic influences to parenting ethnotheories.

Child-rearing practices

Child-rearing practices are specific, goal-directed behaviors through which parents perform their parental duties (Darling & Steinberg, 1993). Child-rearing practices vary with parents’ ideas about child development and their socioeconomic and cultural background (Tajima & Harachi, 2010; Williams, Williams, Lopez, & Tayko, 2000), and influence child development (Wade, 2004). The most frequently studied parenting behaviors include responsiveness, warmth, induction, power assertion, cognitive stimulation, and demandingness (Tamis-LeMonda, Shannon, Cabrera, & Lamb, 2004).

Associations with socioeconomic background of the parents have been reported: Parents with higher levels of education know more about child development, display higher levels of emotional and verbal responsiveness, and provide higher levels of cognitive stimulation and inductive reasoning to their children (Bradley & Corwyn, 2002; Davis-Kean, 2005); less educated mothers display more controlling behaviors and physical disciplining, and less verbal interaction or stimulation of children (Pinderhughes, Nix, Foster, Jones, & The Conduct Problems Prevention Research Group, 2001; von der Lippe, 1999). Cross-cultural differences in child-rearing practices have been found as well. More specifically, non-Western, more collectivistic mothers were found to use more punishment and control (Cardona, Nicholson, & Fox, 2000) and less praising and verbal encouragement (Bradley, Corwyn, & Whiteside-Mansell, 1996) than Western, individualistic mothers.

It can be concluded that, broadly speaking, mothers in individualistic cultures and higher-educated mothers display a parenting pattern that is more child-centered with a greater focus on responsiveness and reasoning, whereas mothers from collectivistic cultures and lower-educated mothers display a more parent-centered pattern with higher levels of power assertion. A more precise delineation of the relative influence of culture and background variables is difficult to give because these variables are often confounded, with Western mothers being more educated than non-Western.

It has been confirmed that parents’ beliefs and expectations have direct and indirect associations with their parenting practices and children’s cognitive, emotional, and social competence (Cheah & Chirkov, 2008; Coleman & Karraker, 2003; Sigel & McGillicuddy De Lisi, 2002). The few studies on the relation between developmental expectations and child-rearing practices showed that parents who expect their children to learn certain skills or behaviors earlier also display the encouraging behaviors that support their children’s development in those areas early (Williams et al., 2000).

Turkish and Dutch parenting patterns

In terms of cultural values, Turkey is traditionally a country that is high on conservatism and hierarchy and low on autonomy and egalitarianism (Schwartz, 1999). The traditional Turkish sociocultural context is characterized by close interpersonal relationships, loyalty, and a focus on kinship (Aygun & Imamoglu, 2002). Rural and lower-educated Turkish parents highly endorse compliance to parents, having close ties with family members, and obedience-oriented parenting, whereas well-educated, urban Turkish parents in urbanized areas display more autonomy-oriented values and child-rearing practices (Kağitçibaşi, 2007; Sunar, 2002; Yagmurlu, Çıtlak, Dost, & Leyendecker, 2009). Obedience-oriented parenting and controlling behaviors are common among Turkish immigrant parents in the Netherlands (Nijsten, 2006).

The Netherlands is a country that is high on egalitarianism and autonomy and low on hierarchy and conservatism (Schwartz, 1999). It is a prototypical culture of independence (Hofstede, 2001). Personal privacy stands out as an important value even when the nuclear family is concerned. Independence and assertiveness are highly endorsed goals in Dutch families, and even infants are expected to be able to play alone and entertain themselves (Pels, 1991; Willemsen & van de Vijver, 1997). Dutch mothers often display parenting styles that support both emotional and material independence of the child (Georgas, Berry, van de Vijver, Kağitçibaşi, & Poortinga, 2006), and they use less controlling and more autonomy-oriented parenting behaviors (Gerrits, Dekovic, Groenendaal, & Noom, 1996).

The present study

This study aimed at identifying the cross-cultural differences and similarities in mothers’ developmental expectations and child-rearing practices while disentangling these variables’ associations with ethnicity, immigration, and educational background. Three main hypotheses were tested. First, we expected that each background variable (i.e., ethnicity, immigration, and educational background) would predict maternal expectations and practices (hypothesis 1); however, there was no expectation about which would be the strongest.

The second hypothesis involved group differences. Dutch and higher-educated mothers were hypothesized to expect earlier development of children than were Turkish and lower-educated mothers, respectively (hypothesis 2a). Moreover, Dutch non-immigrant, and higher-educated mothers were expected to report more use of child-centered parenting such as inductive reasoning (hypothesis 2b) and less use of parent-centered parenting such as obedience-demanding behavior (hypothesis 2c), compared to Turkish, immigrant, and lower-educated mothers.

Third, a significant overall association between mothers’ expectations and parenting practices was expected. Mothers with earlier developmental expectations would report using more child-centered parenting and fewer parent-centered parenting practices (hypothesis 3).

Method

Participants

Participants of the study comprised 111 ethnic Dutch and 111 Turkish immigrant mothers in the Netherlands, and 242 Turkish mothers living in Turkey. All participating mothers had at least one child of preschool age who was the target child in this study. A mother was classified as Turkish immigrant only if her parents were born in Turkey. All participants were recruited through snowball sampling. Participants in Turkey were recruited from diverse areas of Turkey, from megacities such as Istanbul to small rural towns, in order to capture much variation in the population of Turkish mothers. Of the Turkish immigrant sample, 28% of the mothers were born in the Netherlands. The mean age of migration to the Netherlands was 12.14 years (SD = 10.22). Sociodemographic statistics of the groups can be seen in Table 1.

Table 1.

Descriptive characteristics of the three samples

Characteristic	Dutch	Turkish-Dutch	Turkish
Sample size (N of mothers)	111	111	242
Age of child (in months)
M	48.94	49.78	48.50
SD	20.50	22.81	20.71
Age of mother (in years)
M	33.97	32.71	30.99
SD	5.32	7.32	5.06
Mother’s marital status (%)
Married	53	94	97
Married, not living with husband	3	—	1
Partners	18	4	—
Single	18	—	—
Divorced	3	2	1
Widowed	5	—	1
Child’s gender (%)
Girl	51	43	50
Education of mother
M	12.37	11.14	10.44
SD	3.97	4.01	7.37
Child order (% of first child)	73	45	60

An analysis of variance revealed that the three groups of mothers were significantly different from each other in terms of maternal age, years of education, and rank order of the target child. Dutch mothers were significantly older than Turkish-Dutch and Turkish mothers, F(2, 464) = 11.091, p < .001, (partial) η² = .05, and non-immigrant Turkish mothers were significantly less educated than the other mothers, F(2, 464) = 3.953, p < .05, η² = .02. In terms of child order, 73% of the target children in the Dutch group were first-borns and this percentage was significantly higher than those in the Turkish-Dutch and Turkish groups, F(2, 464) = 9.303, p < .001, η² = .04.

Materials

Sociodemographic questionnaire

The background information form had questions about the target child (e.g., date of birth and sex) and family background characteristics (e.g., maternal and paternal age, education, and marital status). Education of mothers was measured as the total number of years of education received.

Child-rearing questionnaire

Mothers completed a modified version of the Child-Rearing Questionnaire developed by Paterson and Sanson (1999). The Turkish version of the Child-Rearing Questionnaire was developed by Yagmurlu and Sanson (2009b). The original Child-Rearing Questionnaire has 30 items where parents indicate the frequency of each behavior on a five-point scale, in which 1 = never and 5 = always. The original scale has four domains: inductive reasoning (e.g., “I try to explain to my child why certain things are necessary”); punishment (e.g., “I use physical punishment, such as smacking, for very bad behavior”); obedience-demanding behavior (e.g., “I expect my child to do what he/she is told to do, without stopping to argue about it”); and warmth (e.g., “My child and I have intimate times together”). The modified version of the scale included two new domains that tap into permissiveness (e.g., “I believe that my child should have his/her way as often as I do”) and stimulation (e.g., “I read books to my child to enhance his/her cognitive development”). Items in these two domains were developed based on existing scales (e.g., Bradley et al., 1996; Keller et al., 2004).

Internal consistencies were .72 for obedience demanding, .74 for punishment, .79 for warmth, .79 for inductive reasoning, .70 for permissiveness, and .82 for stimulation in the total sample. Reliability values in the three samples were fairly similar to those reported for the total sample.

Developmental expectations scale

The Developmental Expectations Scale, which was developed by Durgel and van de Vijver (2008) as a measure of maternal developmental expectations, is based on scales used in previous studies (e.g., Goodnow et al., 1984; Hess et al., 1980; Willemsen & van de Vijver, 1997). This scale aims to capture several domains, from cognitive and physical skills to social behaviors and interrelational aspects of child development. Throughout the procedure of developing the scale, authors ensured the cultural relevance of items in the Turkish and Dutch cultural contexts. Mothers were asked to indicate the age they expect a child to be able to do certain skills/behaviors for the first time. The items were not directly focusing on mothers’ own children, but referred more to any child. The scale consisted of eight subscales: physical skills (e.g., “Stand without support more than 1 minute”); cognitive skills (e.g., “Tell what is left and right”); self-control (e.g., “Keep playing according to game rules even if she loses”); social skills (e.g., “Share toys with other children”); autonomy (e.g., “Decide what to wear”); obedience (e.g., “Give up reading/TV when mother asks for help”); family orientation (e.g., “Like to visit grandparents”); and traditional/moral rules (e.g., “Have a preference for toys according to own gender”). For each subscale, average scores of mothers’ reported age expectations for the particular domain were calculated. Lower scores indicated earlier expectations. Each subscale yielded a strong first factor, which supports the factorial validity of the scale. Alpha coefficients for the total sample were .88 for physical skills, .91 for cognitive skills, .89 for self-control, .91 for social skills, .88 for autonomy, .90 for obedience, .90 for family orientation, and .88 for traditional/moral rules. Again, values for the three samples separately were similar to those reported for the total sample.

Procedure

Mothers were given information on the study and appointments were made for a home visit to their houses. Dutch mothers were interviewed by a Dutch researcher, Turkish non-immigrant mothers were interviewed by a Turkish researcher, and Turkish immigrant mothers were interviewed by a Turkish interviewer in either Dutch or Turkish language, according to their preference. Home visits took about 40 minutes to complete the tasks.

Results

We first investigated the construct equivalence of the scales by addressing to what extent the Developmental Expectations and Child-Rearing questionnaires measure the same psychological constructs in each group. Equivalence was supported, because the agreement between the factor loadings (Tucker’s phi; van de Vijver & Leung, 1997) was higher than the recommended lower limit of .90 for the factors of both scales in all pairwise comparisons.

The predictive role of background variables in developmental expectations and child-rearing practices

We wanted to examine the predictive role of ethnicity (i.e., Turkish vs. Dutch), immigration (i.e., immigrant vs. non-immigrant), and educational background in developmental expectations separately in regression analyses. However, a challenge in the analysis would be the large number of subscales for both developmental expectations and child rearing relative to the small sample sizes per group. Therefore we first used exploratory factor analysis of the subscale scores to reduce their dimensionality. A factor analysis of the eight subscales of the Developmental Expectations Scale yielded a strong first factor which explained 65% of the total variance (in line with earlier studies which found that these expectations are positively interrelated; Willemsen & van de Vijver, 1997). A factor analysis of the child-rearing subscales showed that obedience-demanding and punishment scales form a factor that explained 43% of the variance, whereas warmth, inductive reasoning, permissiveness, and stimulation formed another factor which explained 21% of the total variance. The correlation between the two factors was weak (r = −.15). Thus we created a child-centered parenting variable with mean score of warmth, inductive reasoning, permissiveness, and stimulation and another variable named parent-centered parenting by taking the mean score of obedience demanding and punishment.

Regression analysis was then used to estimate the relative influence of the background variables on developmental expectations and child rearing. Dummy variables were created for ethnicity (0 referring to Dutch and 1 referring to Turkish) and immigration status (0 being non-immigrant and 1 being immigrant). A stepwise regression analysis was conducted: mothers’ age and child order were entered in the first step to account for their confounding effects, and maternal education, ethnicity, and immigration status together were added in the second step. Results showed that parental education and ethnicity were significant predictors of developmental expectations, R ²= .089, F(3, 464) = 14.943, p < .001 (see Table 2). This analysis showed that being an immigrant or majority member did not explain mothers’ developmental expectations, and both ethnicity and maternal education had their own contribution; however, ethnicity had slightly more impact on mothers’ developmental expectations.

Table 2.

Standardized regression coefficients of regression analyses with developmental expectations, child-centered parenting, and parent-centered parenting as dependent variables (N = 464)

	Dependent variable
Predictor	Developmental expectations	Child-centered	Parent-centered
Ethnicity	.203***	.392***	.071
Education	−.136**	.117**	−.268***
Immigration	−.036	−.204***	−.068
Adjusted R ²	.083**	.148**	.074***

Note. Scoring of dummy variables is as follows: For ethnicity 0 = Dutch, 1 = Turkish; for immigration status 0 = majority group, 1 = immigrant.

**p < .01; ***p < .001.

The regression analyses with the child-rearing factors as dependent variables showed that maternal education, ethnicity, and immigration status were significant predictors of child-centered parenting practices, R ²= .148, F(3, 464) = 21.027, p < .001 (see Table 2). Ethnicity was found to be the strongest predictor of child-centered parenting practices of the mothers. Maternal education was the only (and strong) predictor of parent-centered parenting, R ²= .074, F(3, 464) = 19.486, p < .001 (see Table 2).

Overall, in line with hypothesis 1, each background variable uniquely added to the prediction of mothers’ expectations and practices; however, maternal education was found to be more important for parent-centered and cultural background was found to be more important for developmental expectations and child-centered parenting practices.

Group comparisons on developmental expectations and child-rearing practices

To test the group differences in developmental expectations of mothers, a one-way multivariate analysis of covariance (MANCOVA) was conducted with group as independent variable, developmental expectations as dependent variables, and mothers’ age, child order, maternal education, and marital status as covariates (see Table 3). The multivariate test was significant with a medium effect size, Wilks’ lambda = .84, F(2, 464) = 5.198, p < .001, η²= .085. As expected, univariate tests revealed significant group differences for all subscales of developmental expectations, except for psychomotor development (see Table 3). However, effect sizes were found to be small. Dutch mothers expected the earliest achievement in all seven domains compared to the Turkish-Dutch and Turkish mothers. Hypothesis 2a was confirmed based on the earlier regression analysis and MANCOVA results: as expected, Dutch and higher-educated mothers expected earlier ages of achievement in developmental skills than Turkish and lower-educated mothers, respectively.

Table 3.

Developmental expectations and child-rearing practices for three samples: Means, standard deviations, and MANCOVA results

	Dutch		Turkish-Dutch		Turkish		Effect sizes before (η²)	Effect sizes after (η²)
Scale	M	SD	M	SD	M	SD	Effect sizes before (η²)	Effect sizes after (η²)
(a) Developmental expectations
Psychomotor	2.94_a	.58	2.97_a	.61	2.98_a	.84	.000 (n.s.)	.000 (n.s.)
Cognitive	3.73_a	.63	4.03_b	.61	4.23_b	.86	.068***	.046***
Self-control	4.92_a	.95	5.51_b	.94	5.71_b	.90	.107***	.085***
Social	4.29_a	.86	4.78_b	.85	4.76_b	1.06	.043***	.032***
Autonomy	4.69_a	.79	5.02_b	.72	4.98_b	.97	.022**	.014*
Obedience	5.08_a	.98	5.35_b	.91	5.48_b	.97	.028**	.018*
Family	5.13_a	1.04	5.59_b	.95	5.64_b	.94	.045***	.031**
Moral	4.42_a	1.04	4.93_b	.91	4.83_b	1.01	.037***	.025**
Total	4.40_a	.70	4.77_b	.63	4.82_b	.78	.06***	.039***
(b) Child-rearing practices
Obedience demandingness	3.10_a	.66	3.08_a	.72	3.15_a	.83	.002(n.s.)	.001 (n.s.)
Punishment	1.53_a	.49	1.63_a	.51	1.83_b	.60	.051***	.035***
Warmth	4.44_a	.46	4.42_b	.48	4.56_a	.41	.020*	.019*
Inductive reasoning	4.15_a	.56	4.20_a	.63	4.39_b	.51	.037***	.041***
Permissiveness	2.96_a	.56	3.28_b	.56	3.56_c	.56	.160***	.163***
Stimulation	3.50_a	.53	3.90_b	.57	4.13_c	.54	.181***	.199***
Child-centered parenting	3.77_a	.39	3.95_b	.44	4.16_c	.40	.14***	.15***
Parent-centered parenting	2.31_a	.48	2.35_a	.48	2.49_b	.58	.02**	.01*

Notes. (i) Higher mean scores refer to later age expectancies for developmental expectations and to more often displayed for child-rearing practices. (ii) Subscripts of means refer to post-hoc tests. Means with different subscripts are significantly different than each other. (iii) The last two columns refer to effect sizes of culture before and after covariates.

* p < .05; **p < .01; ***p < .001.

A similar one-way MANCOVA was run with child-rearing practices as dependent variables and mothers’ age, child order, maternal education, and marital status as covariates (results can be seen in Table 3). Results showed that the three groups of mothers significantly differed in their child- and parent-centered practices, with a medium effect size, Wilks’ lambda = .82, F(2, 464) = 23.383, p < .001, η²= .09. Child-centered practices were displayed most by Turkish mothers, followed by Turkish-Dutch and Dutch mothers, respectively. Parent-centered practices were reported more by Turkish mothers compared to Turkish-Dutch and Dutch mothers; however, the effect size of this difference was very small. The MANCOVA and regression analyses provided partial support for hypotheses 2b and 2c: In line with expectations, higher-educated mothers displayed child-centered practices more and parent-centered practices less than lower-educated mothers. However, contrary to the hypothesis, child-centered practices were reported more by Turkish mothers than Dutch mothers.

To understand the reasons behind this finding which did not match with our expectations, we scrutinized the sample characteristics. There was a significant educational difference between Dutch and Turkish mothers in our study: 86% of Dutch mothers had at least completed secondary school, whereas that was the case only for 50% of the Turkish sample. Within the Turkish group there was more variety, ranging from mothers who were not educated at all to mothers with a Master’s or PhD degree. In order to better understand the dynamics behind the relations between education, culture, and group differences in child-centered and parent-centered parenting practices, we separated the Turkish sample into higher-educated (completed at least secondary school) and lower-educated (less than secondary school) and compared with the Dutch and Turkish-Dutch samples. A MANCOVA analysis was conducted with group as independent variable and child-rearing factors as dependent variables. A significant and medium effect size was found, Wilks’ lambda = .81, F(3, 464) = 16.416, p < .001, η²= .098. Higher- and lower-educated Turkish mothers reported more use of child-centered practices compared to the Dutch mothers, Turkish immigrant mothers reporting frequency of these practices less than Turkish and more than Dutch mothers. For parent-centered practices, Dutch, Turkish-Dutch, and higher-educated Turkish mothers displayed a similar pattern which was significantly different than lower-educated Turkish parents. So the high level of parent-centered parenting in the Turkish group was restricted to the subsample with a low level of education, and the unexpectedly high reports of child-centered parenting were common in both lower- and higher-educated Turkish mothers.

Relations between developmental expectations and parenting practices

Finally, in order to test the third hypothesis, correlations among the developmental expectations and parenting practices of the mothers were examined (see Table 4). It was found that mothers’ developmental expectations were positively related to their use of parent-centered parenting practices. We found the expected small yet significant overall association between mothers’ expectations and parenting practices: Mothers who believed that children can learn and develop earlier reported to use fewer parent-centered behaviors (hypothesis 3). It is important to note that the associations between expectations and parenting are quite weak. This is most likely because these two concepts are related to one another via other variables which were not examined in this study. This indeed would result in small correlations among the examined variables.

Table 4.

Pearson product-moment correlations among developmental expectations and child-rearing practices (N = 464)

Scale	Obedience demanding	Punishment	Warmth	Inductive reasoning	Permissive	Stimulation	Child-centered	Parent-centered
Psychomotor	.111*	.044	−.131**	−.030	.068	−.086	−.049	.103*
Cognitive	.206***	.187***	−.142**	−.077	.066	−.099*	−.072	.246***
Self-control	.117*	.142**	-.088	−.046	.161***	.009	.022	.159**
Social	.219***	.130**	−.186***	−.114*	.072	−.062	−.080	.226***
Autonomy	.231***	.106*	−.176***	−.139**	−.006	−.123**	−.134**	.221***
Obedience	.139^**	.139^**	−.162^***	−.107^*	.030	−.092^*	−.097^*	.173***
Family orientation	.165***	.089	−.087	−.048	.092*	−.033	−.017	.165***
Moral	.161**	.121**	−.142**	−.123**	.074	−.052	−.068	.179***
Developmental expectation total	.219***	.146**	−.167***	−.108*	.082	−.076	−.075	.234***

*p < .05; **p < .01; ***p < .001.

Discussion

We examined the cross-cultural differences and similarities in (self-reported) developmental expectations and parenting practices in ethnic Dutch and Turkish immigrant mothers living in the Netherlands and Turkish mothers in Turkey. We also aimed at disentangling the associations among ethnicity, socioeconomic factors (notably education), immigration, and developmental expectations and parenting practices. Very few studies in the literature pay attention to this combination of background factors in relation to parenting beliefs and practices. We set out to avoid the common problem in studies that compare majority and immigrant groups of a confounding of ethnicity and socioeconomic status (Hill, 2006). Our findings regarding the relative predictive value of background variables (ethnicity, education, and immigration status) suggested that each of these background variables has a unique contribution to make to developmental expectations and child- and parent-centered parenting practices. Yet maternal education is the most consistent and thus most important predictor of parenting relative to ethnic background and immigration history. Particularly parent-centered child rearing was found to be related to maternal education and not to cultural background. Although Turkish and Dutch groups significantly differed in their power-assertion behaviors, the effect size of the role of culture on parent-centered behaviors was almost equal to zero.

In line with Kohn’s (1959) studies, we can argue that mothers’ socioeconomic status, indicated by their educational level in this study, is very significant for their use and endorsement of authority in their child rearing. When it comes to disciplining techniques, parents who are highly educated may realize that obedience of the child has adverse effects (Chang, Lansford, Schwartz, & Farver, 2004) and is neither functional nor desired in their socioeconomic, usually individualistic context (Kağitçibaşi, 2007). Studies with well-educated Turkish mothers in Turkey also show that they display less obedience-demanding practices despite coming from a cultural context that is described as typically collectivistic (Yagmurlu et al., 2009). Our study demonstrates that what may be conceived as cross-cultural differences in parenting can be largely viewed as educational and socioeconomic variables. These findings have both theoretical and practical implications: To start with the former, it is important in cross-cultural and immigrant studies to disentangle cultural and socioeconomic background. The practical implication is that it is important for both professionals and researchers to be aware of this confounding and to be careful not to over-attribute differences between immigrant and majority group parents to cultural differences.

Child-centeredness is found to be more related to the cultural background than education (with medium- and small-effect size, respectively). As can be expected based on acculturation models (Berry, 2001) and family change theory (Kağitçibaşi, 2007), Turkish immigrant mothers display a pattern in between the Turkish and the Dutch mothers in their child-centered practices. This finding implies that as immigrant mothers experience two different societies, their parenting patterns that are sensitive to culture, in this case child-centeredness, might go through a transformation from the Turkish to the Dutch cultural pattern. It is important to note here that the magnitude of the association between culture and child-centered parenting was small; however, this small association remained significant even after controlling for other background variables. This emphasizes the relevance of cultural background on child-centeredness over and above other background characteristics.

Another aim of this study was to examine group differences in mothers’ expectations. We found that cross-cultural skill differences were not domain specific: Ethnic Dutch mothers expected the earliest development in all domains compared to the Turkish-Dutch and Turkish mothers, with most domains showing small effect sizes. In line with earlier findings pointing to an association between mothers’ age expectancies and their cultural background (Goodnow et al., 1984; Willemsen & van de Vijver, 1997), mothers’ cultural views on child development might be one of the explanations of our results. This finding might suggest that Dutch mothers expect children to learn and develop earlier than non-Western Turkish mothers since they see children as “individuals” with abilities, skills, and a global potential that should be developed. On the other hand, Turkish mothers perceive children more as “kids,” which means that these mothers are more focused on establishing and maintaining good relationships with the children (Kağitçibaşi, 2007). Although related to education, this (small) difference between the Dutch and the Turkish and Turkish-Dutch mothers could not be statistically “explained away” by educational differences.

This parenting pattern of Turkish parents was also found when their parenting practices were examined. Contrary to our expectations based on studies with other cultural groups, the Turkish mothers reported more use of child-centered behaviors compared to the ethnic Dutch. The behaviors that were most different across cultural groups were permissiveness and cognitive stimulation: Turkish and Turkish-Dutch mothers reported these behaviors more than Dutch mothers. As mentioned earlier, Turkish mothers emphasize establishing and maintaining good and strong parent–child relationships, especially during the preschool years, and they see permissiveness as a tool to achieve this dyadic bond (Kağitçibaşi, 2007; Khounani, 2000). Moreover, cognitive stimulation is considered to be important, particularly for immigrant and higher-educated Turkish mothers, since they believe their children’s success at school and in professional life is a necessity in the new, modern context (Durgel et al., 2009; Yagmurlu et al., 2009). Few studies conducted on observed parenting practices of Turkish-Dutch and Dutch mothers also showed that Turkish-Dutch mothers are more cognitively stimulating and interactive than Dutch mothers as they spend time with their preschoolers (Durgel & van de Vijver, 2011). Our finding may be related to the high percentages of our Turkish participants who were stay-at-home mothers (compared to the working Dutch mothers) whose devotion and main task at home was to take care of their preschoolers. Turkish mothers, compared to the ethnic Dutch mothers, may spend more time with their preschoolers and may be more interactive in general, which may reflect their interdependence-oriented style. It could be that they interact more often since they see these as more opportunities to develop and maintain relations and ties with their children.

Our study is not free from limitations. First, this study revealed the associations between culture, education, and mothers’ beliefs and behaviors; however, it is very important to note that most of the associations captured here were small in size. Nevertheless, findings of this study still provide us with valuable information since we were interested in the cultural and educational patterns. Second, this study only examined mother’s expectations and self-reported parenting practices, and does not involve actual child-rearing practices of these mothers. Parents’ self-reports can often be different from how they engage with their children in everyday practice. Hence examination of observed parenting practices of mothers would provide us with a more complete understanding of the parenting dynamics. Moreover, using self-reported data might lead to various biases, such as social desirability which refers to the tendency to answer questions in a way that is viewed favorably by others. For instance, Turkish mothers, contrary to our expectations, reported to display child-centered parenting practices more frequently than the Dutch mothers; this result may be influenced by social desirability bias. Similarly, acquiescence bias might play a role particularly in self-report studies. Acquiescence response style refers to the tendency to respond positively, regardless of the content of the question. Turkish mothers seem to score higher on all developmental expectations and child-rearing categories, which might imply that they tend to agree with questions more than the Dutch mothers. This would be in line with the literature, which shows more acquiescence bias and social desirability in groups higher on collectivism (Smith, 2004) and power distance (van Hemert, van de Vijver, Poortinga, & Georgas, 2002). Observing mothers’ behaviors in a natural environment would be a valuable complement to self-report data; however, the knowledge that one is being observed may also bring some degree of bias in behavior due to social desirability effects. We would not argue in favor of abandoning self-report data in the domain of parenting. It was previously established that self-reported and observed parenting behaviors correlated with each other (Kochanska, Kuczynski, & Radke-Yarrow, 1989). It is also true that questionnaires tap numerous behaviors that otherwise cannot be assessed in behavioral observations (e.g., how the parent handles everyday conflicts, rhythmicity of the child’s daily activities; Rothbart & Bates, 1998).

Another limitation of this study is the use of snowball sampling (Goodman, 1961), which may lead to the increased risk of sample bias as well as problems with generalizing our findings to the larger population (van de Vijver & Leung, 1997). Snowball samples are highly biased towards inclusion of participants who have many interrelations, hence they decrease the generalizability of the findings of the study to the general population. Nevertheless, compiling random samples of mainstream and immigrant families with young children is very difficult in the Netherlands given the unavailability of registers; thus snowballing is the most commonly used sampling method in this context, since this type of recruitment is especially useful if the target groups are difficult to contact (Arends-Tóth & van de Vijver, 2008; Boehnke, Lietz, Schreider, & Wilhelm, 2011). When examining hard-to-reach, hidden populations for whom sampling frames are not feasible, snowball sampling may be the only method available. On the other hand, the Turkish data collected in Turkey were made sure to represent several facets of the general population from lower to higher education, rural to urban backgrounds, and so on.

Finally, we address suggestions for future studies. This study has found that maternal education and culture are significantly predicting Dutch and Turkish mothers’ parenting; further studies should include more background variables (e.g., family income, mothers’ employment status) to identify and unpackage cross-cultural differences better (Whiting, 1976). Moreover, expanding this research to various other cultural groups would increase the generalizability of our findings. Besides, examining the changes in maternal expectations and practices of immigrant groups in relation to their acculturation orientation would be a great addition to the literature in understanding the aspects of parenting that are culture-sensitive and resistant to change. Also, it is important to note that the present research is a cross-sectional study which does not inform us about the causality between variables studied. Therefore, in order to talk about causal relationship, longitudinal studies must be conducted.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Arends-Tóth

J. V.

van de Vijver

F. J. R.

(2008). Family relationships among immigrants and majority members in the Netherlands. Applied Psychology: An International Review, 57, 466–487. doi: 10.1111/j.1464-0597.2008.00331.x

Atkinson

(1994). Rural and urban families’ use of child care. Family Relations, 43, 16–22. Retrieved from: https://dx-doi-org.web.bisu.edu.cn/10.2307/585137

Aygun

Z. K.

Imamoglu

(2002). Value domains of Turkish adults and university students. Journal of Social Psychology, 142, 333–351. doi: 10.1080/00224540209603903

Berry

J. W.

(2001). A psychology of immigration. Journal of Social Issues, 57, 615–631. Retrieved from: https://dx-doi-org.web.bisu.edu.cn/10.1111/0022-4537.00231

Boehnke

Lietz

Schreier

Wilhelm

(2011). Sampling: The selection of cases for culturally comparative psychological research. In Matsumoto

Van de Vijver

F. J. R.

(Eds.), Cross-cultural research methods in psychology (pp. 101–129). Cambridge, UK: Cambridge University Press.

Bradley

R. H.

Corwyn

R. F.

(2002). Socioeconomic status and child development. Annual Review of Psychology, 53, 371–399.

Bradley

R. H.

Corwyn

R. F.

Whiteside-Mansell

(1996). Life at home: Same time, different places. An examination of the HOME inventory in different cultures. Early Development and Parenting, 5, 251–269. doi: 10.1002/(SICI)1099-0917(199612)5:4<251: AID-EDP137>3.0.CO;2-I

Cardona

P. G.

Nicholson

B. C.

Fox

R. A.

(2000). Parenting among Hispanic and Anglo-American mothers with young children. Journal of Social Psychology, 140, 357–365. doi: 10.1080/00224540009600476

Chang

Lansford

J. E.

Schwartz

Farver

J. M.

(2004). Martial quality, maternal depressed affect, harsh parenting, and child externalizing in Hong Kong Chinese families. International Journal of Behavioral Development, 28, 311–318. Retrieved from: https://dx-doi-org.web.bisu.edu.cn/10.1080/01650250344000523

10.

Cheah

C. S. L.

Chirkov

(2008). Parents’ personal and cultural beliefs regarding young children: A cross-cultural study of Aboriginal and Euro-Canadian mothers. Journal of Cross-Cultural Psychology, 39, 402–423. Retrieved from: https://dx-doi-org.web.bisu.edu.cn/10.1177/0022022108318130

11.

Coleman

P. K.

Karraker

K. H.

(2003). Maternal self-efficacy beliefs, competence in parenting, and toddlers’ behavior and developmental status. Infant Mental Health Journal, 24, 126–148. doi: 10.1002/imhj.10048

12.

Darling

Steinberg

(1993). Parenting style as context: An integrative model. Psychological Bulletin, 113, 487–496. doi: 10.1037/0033-2909.113.3.487

13.

Davis-Kean

P. E.

(2005). The influence of parent education and family income on child achievement: The indirect role of parental expectations and the home environment. Journal of Family Psychology, 19, 294–304. doi: 10.1037/0893-3200.19.2.294

14.

Durgel

E. S.

Leyendecker

Yagmurlu

Harwood

(2009). Sociocultural influences on German and Turkish immigrant mothers? Long-term socialization goals. Journal of Cross-Cultural Psychology, 40, 834–852. Retrieved from: https://dx-doi-org.web.bisu.edu.cn/10.1177/0022022109339210

15.

Durgel

E. S.

van de Vijver

(2008, July). Developmental expectations of Dutch and Turkish-Dutch mothers of preschoolers. Paper presented at the 19th International Congress of the International Association for Cross-Cultural Psychology, Bremen, Germany.

16.

Durgel

E. S.

van de Vijver

(2011, July). Parenting practices of Turkish immigrant and Dutch mothers in the Netherlands. Paper presented at the Regional Conference of the International Association for Cross-Cultural Psychology, Istanbul, Turkey.

17.

Georgas

Berry

J. W.

Van de Vijver

F. J. R.

Kağitçibaşi

Poortinga

Y. H.

(2006). Families across cultures: A 30-nation psychological study. Cambridge, UK: Cambridge University Press.

18.

Gerrits

Dekovic

Groenendaal

Noom

(1996). Opvoedingsgedrag [Parenting behavior]. In Rispens

Hermanns

J. M.

Meeus

W. H.

(Eds.), Opvoeden in Nederland [Parenting in the Netherlands] (pp. 41–69). Assen, the Netherlands: Van Gorcum.

19.

Goodman

L. A.

(1961). Snowball sampling. Annals of Mathematical Statistics, 20, 572–579. Retrieved from: https://dx-doi-org.web.bisu.edu.cn/10.1214/aoms/1177729949

20.

Goodnow

Cashmore

Cotton

Knight

(1984). Mothers’ developmental timetables in two cultural groups. International Journal of Psychology, 19, 193–205. doi: 10.1080/00207598408247526

21.

Harkness

Super

C. M.

Mavridis

C. J.

(2011). Parental ethnotheories about children’s socioemotional development. In Chen

Rubin

K. H.

(Eds.), Socioemotional development in cultural context (pp. 73–98). New York, NY: Guilford Press.

22.

Hess

Kashiwagi

Azuma

Price

Dickson

W. P.

(1980). Maternal expectations for mastery of developmental tasks in Japan and the United States. International Journal of Psychology, 15, 259–271. doi: 10.1080/00207598008246996

23.

Hill

N. E.

(2006). Disentangling ethnicity, socioeconomic status, and parenting: Interactions, influences, and meaning. Vulnerable Children and Youth Studies Journal, 1, 114–124. doi: 10.1080/17450120600659069

24.

Hofstede

(2001). Culture’s consequences: Comparing values, behaviors, institutions, and organizations across nations (2nd ed.). Thousand Oaks, CA: SAGE Publications.

25.

Kağitçibaşi

(2007). Family, self, and human development across cultures: Theory and applications (2nd ed.). Mahwah, NJ: Lawrence Erlbaum.

26.

Keller

Lohaus

Kuensemueller

Abels

Yovsi

Voelker

Mohite

. (2004). The bio-culture of parenting: Evidence from five cultural communities. Parenting: Science and Practice, 4, 25–50. doi: 10.1207/s15327922par0401_2

27.

Khounani

P. M.

(2000). Binationale Familien in Deutschland und die Erziehung der Kinder: eine Vergleichsuntersuchung zur Familären Erziehungsituation in mono- und bikulturellen Familien im hinblick auf multikulturelle Handlungsfähigkeit [Binational families in Germany and the socialization of children: A comparative study of the familiar socialization situation in mono- and bicultural families with a view of multicultural skills]. Frankfurt, Germany: Lang.

28.

Kochanska

Kuczynski

Radke-Yarrow

(1989). Correspondence between mothers’ self-reported and observed child-rearing practices. Child Development, 60, 56–63.

29.

Kohn

M. L.

(1959). Social class and parental values. American Journal of Sociology, 64, 337–351. Retrieved from: https://dx-doi-org.web.bisu.edu.cn/10.1086/222493

30.

Nijsten

C. C.

(2006). Coming from the East: Child rearing in Turkish families. In Deković

Pels

Model

(Eds.) Unity and diversity in child rearing: Family life in a multicultural society (pp. 25–58). Ceredigion, UK: The Edwin Mellen Press.

31.

Paterson

Sanson

(1999). The association of behavioral adjustment to temperament, parenting and family characteristics among 5-year-old children. Social Development, 8, 293–309.

32.

Pels

(1991). Developmental expectations of Moroccan and Dutch parents. In Bleichrodt

Drenth

(Eds.), Contemporary issues in cross-cultural psychology (pp. 64–71). Amsterdam, the Netherlands: Swets and Zeitlinger.

33.

Pinderhughes

E. E.

Nix

Foster

E. M.

Jones

, & The Conduct Problems Prevention Research Group. (2001). Parenting in context: Impact of neighborhood poverty, residential stability, public services, social networks, and danger on parental behaviors. Journal of Marriage and the Family, 63, 941–953. doi:10.1111/j.1741-3737.2001.00941.x

34.

Rothbart

M. K.

Bates

J. E.

(1998). Temperament. In Damon (Series Ed.)

Eisenberg

(Vol. Ed.), Handbook of child psychology: Vol. 3. Social, emotional and personality development (5th ed., pp. 105–176). New York, NY: Wiley.

35.

Schwartz

S. H.

(1999). Cultural value differences: Some implications for work. Applied Psychology: An International Review, 48, 23–47. doi: 10.1111/j.1464-0597.1999.tb00047.x

36.

Sigel

I. E.

McGillicuddy De Lisi

A. V.

(2002). Parent beliefs are cognitions: The dynamic belief systems model. In Bornstein

M. H.

(Ed.), Handbook of parenting: Being and becoming a parent (pp. 485–508). Mahwah, NJ: Lawrence Erlbaum.

37.

Smith

P. B.

(2004). Acquiescent response bias as an aspect of cultural communication style. Journal of Cross-Cultural Psychology, 35, 50–61.

38.

Sunar

(2002). Change and continuity in the Turkish middle class family. In Ozdalga

Liljestrom

(Eds.), Autonomy and dependence in family: Turkey and Sweden in critical perspective (pp. 217–238). Istanbul, Turkey: Swedish Research Institute.

39.

Tajima

E. A.

Harachi

T. W.

(2010). Parenting beliefs and physical discipline practices among Southeast Asian immigrants: Parenting in the context of cultural adaptation to the United States. Journal of Cross-Cultural Psychology, 41, 212–235. Retrieved from: https://dx-doi-org.web.bisu.edu.cn/10.1177/0022022109354469

40.

Tamis-LeMonda

C. S.

Shannon

J. D.

Cabrera

Lamb

(2004). Fathers and mothers at play with their 2- and 3-year-olds: Contributions to language and cognitive development. Child Development, 75, 1806–1820. doi: 10.1111/j.1467-8624.2004.00818.x

41.

Triandis

H. C.

(1994). Culture and social behavior. New York, NY: McGraw-Hill.

42.

Van de Vijver

F. J. R.

Leung

(1997). Methods and data analysis for cross-cultural research. Newbury Park, CA: SAGE Publications.

43.

Van Hemert

D. A.

van de Vijver

F. J. R.

Poortinga

Y. H.

Georgas

(2002). Structural and functional equivalence of the Eysenck personality questionnaire within and between countries. Personality and Individual Differences, 33, 1229–1249.

44.

Von der Lippe

(1999). The impact of maternal schooling and occupation on child-rearing attitudes and behaviours in low income neighbourhoods in Cairo, Egypt. International Journal of Behavioral Development, 23, 703–729. doi: 10.1080/016502599383766

45.

Wade

S. M.

(2004). Parenting influences on intellectual development and educational achievement. In Hoghughi

Long

(Eds.), Handbook of parenting: Theory and research for practice (pp. 198–212). London, UK: SAGE Publications.

46.

Whiting

B. B.

(1976). The problem of the packaged variable. In Riegel

K. F.

Meacham

J. A.

(Eds.), The development of individuals in a changing World. Chicago, IL: Aldine.

47.

Willemsen

M. E.

van de Vijver

F. J. R.

(1997). Developmental expectations of Dutch, Turkish-Dutch, and Zambian mothers: Towards an explanation of cross-cultural differences. International Journal of Behavioral Development, 21, 837–854. doi: 10.1080/016502597384695

48.

Williams

P. D.

Williams

A. R.

Lopez

Tayko

N. P.

(2000). Mothers’ developmental expectations for young children in the Philippines. International Journal of Nursing Studies, 37, 291–301. doi: 10.1016/S0020-7489(00)00004-3

49.

Yagmurlu

Çıtlak

Dost

Leyendecker

(2009). Child socialization goals of Turkish mothers: An investigation of education related within-culture variation. Turkish Journal of Psychology, 24, 1–15.

50.

Yagmurlu

Sanson

(2009a). Acculturation attitudes and parenting in Turkish-Australian mothers. Journal of Cross-Cultural Psychology, 40, 361–380. doi: 10.1177/0022022109332671

51.

Yagmurlu

Sanson

(2009b). Parenting and temperament as predictors of prosocial behavior in Australian and Turkish-Australian children. Australian Journal of Psychology, 61, 77–88. doi: 10.1080/00049530802001338