A longitudinal study of Estonian mothers’ self-reported language teaching practices and children’s language skills

Abstract

This study explored associations between mothers’ language teaching practices and children’s language skills concurrently and longitudinally, while also taking into account the children’s sex and mothers’ education. Estonian mothers of 76 children reported their language teaching practices at child ages 3;0 and 4;0. Children’s language comprehension and production were measured via the examiner-administered New Reynell Developmental Language Scales (NRDLS). The results indicated that at Wave 1, girls scored higher on language comprehension and production than boys. Mothers’ higher education predicted higher scores on language production. Maternal corrective feedback was a negative predictor of children’s concurrent language comprehension. At Wave 2, neither maternal teaching practices, maternal education nor child’s sex predicted the language measures of interest. The longitudinal results showed that later language production scores were negatively predicted by mothers’ corrective feedback. Yet, the strongest predictors of both language scores were the scores of language comprehension and production measured 1 year earlier. Moreover, maternal language teaching practices at Wave 2 were predicted by mothers’ early language teaching practices, but not by children’s earlier language skills.

Keywords

Language development language teaching practices maternal speech vocabulary language comprehension language production

Individual differences in early language skills have long-lasting effects on children’s later development and their school and life achievement (Golinkoff et al., 2019; Rescorla, 2009). A great deal of theorizing and empirical research on language development has been done in the social-interactionist framework, which holds that children acquire language through conversational exchanges with adults (Bruner, 1990; Hoff, 2006; Vygotsky, 1978). Although some of the variability in early language skills is rooted in heritability (Stromswold, 2001), children’s language learning experiences in parent–child conversations – the quantity and quality of the language input children receive as well as children’s involvement in one-to-one dyadic conversations – play a crucial role in language development (Golinkoff et al., 2019; Hart & Risley, 1995; Hoff, 2006; Rowe & Snow, 2020). These theoretical views are based on research that has been mostly conducted in the Western English-speaking world. Much less is known about children’s language development in other settings with somewhat different family interaction patterns. For instance, parents with non-Western backgrounds (i.e. East-Asia) as well as from Nordic regions in Europe (i.e. Estonia and Finland) have been found to talk less with children and to expect less verbalization from their children (see Tulviste, 2019). In order to understand the mechanisms of language learning, more knowledge about the differential effects of various features of parent–child conversation on language development at different child ages across cultures is needed.

Links between child language and mothers’ language teaching practices

Most studies have assessed parents’ speech style directly by the time-consuming transcription and analysis of parent–child real-life interactions. Only few have used parental self-reports as a source of information about parents’ ‘language teaching practices’. In this study, we address the two aspects of parental self-reported language teaching practices identified in previous studies (Dale et al., 2015; Tulviste & Tamm, 2019). Informal language stimulation was defined as parents’ language practices that give children more talking experiences as parents tend to speak a lot to children and engage children frequently in conversations in terms of reading books and looking at pictures with children, asking them to talk about recent events, talking when doing household chores, and so on (Dale et al., 2015; Tulviste & Tamm, 2019). Providing corrective feedback was described as the tendency to directly point out children’s language errors and correct pronunciation, sentence structure, and vocabulary (Dale et al., 2015; Tulviste & Tamm, 2019).

A large twin study with 3- to 4-year-olds in the United Kingdom found that parental self-reported informal language stimulation was positively, whereas providing corrective feedback was weakly and negatively, related to parent-reported expressive vocabulary and grammar (Dale et al., 2015). The parental speech styles and children’s language abilities had partially genetic bases (Dale et al., 2015). Similar associations between mothers’ self-reported language teaching practices and children’s language skills were found in Estonia at children’s age 3;0 in the analysis of the Wave 1 data of the current study (Tulviste & Tamm, 2019). Estonian middle-class mothers have been observed to talk less, elicit less verbalization from children, and be more directive than parents with a similar educational level but from different cultures (Tulviste et al., 2003, 2016). Yet, Estonian mothers’ self-reported use of informal language stimulation was associated positively with mother-reported expressive vocabulary and experimenter-assessed language comprehension, whereas corrective feedback was negatively related to language comprehension and production (Tulviste & Tamm, 2019). The study demonstrated that mothers’ language teaching practices, in turn, were related to child-rearing value orientation: Mothers who expected more autonomy and self-direction from children tended to report more frequent use of both speech styles (Tulviste & Tamm, 2019).

A study conducted with 2- to 4-year-olds in China (Zhang et al., 2008) indicated that although Chinese parenting has been generally characterized as including more teaching than European-American parenting, mothers who reported a higher frequency of conversation eliciting and storytelling ‘language teaching practices’ had children with larger parent-reported expressive vocabulary. The practice of eliciting imitation, however, was negatively related to parent-reported vocabulary and grammatical development (Zhang et al., 2008). Thus, despite some cultural differences in the pattern of British, Chinese, and Estonian family interactions, and somewhat different methods used in the studies, the facilitative features of parents’ practices of talking with children in order to teach them language seem to be rather similar across cultures.

Age-related differences in the effects of language teaching practices

There is research evidence suggesting that the effectiveness of specific features of parental speech depends on the age/language skills of the child (Rowe, 2012; Rowe & Snow, 2020). Thus, associations between parents’ language teaching practices and children’s language skills might change when children become older. Studies suggest that the positive effect of children’s engagement in conversations does not decline with age (Rowe & Snow, 2020), but the negative effect of corrective feedback might disappear. It is known that children’s early language acquisition depends on implicit learning, but explicit instructions and corrections become more important in later language development, for example, in the acquisition of academic language in instructed settings (Snow, 2014; Yang & Li, 2012). Such language learning requires some level of metalinguistic awareness (i.e. the ability to shift attention away from the meaning to some features of language itself), which starts to develop at the age of 3–4 years (Chaney, 1992). During the fourth year of life, children’s language abilities have been shown to be predicted rather by mothers’ use of decontextualized language than the quantity of input (Rowe, 2012). Thus, at older ages, mothers’ overt corrective feedback may help to promote language learning or at least be less negatively related to children’s language skills than found among Estonian children at age 3;0 (Tulviste & Tamm, 2019).

The present study

The studies described above have identified concurrent associations between mothers’ self-reported language teaching practices and children’s language skills (Dale et al., 2015; Tulviste & Tamm, 2019; Zhang et al., 2008). The present two-wave longitudinal study focuses on Estonian mothers’ self-reported ways of using language in conversations with children and its concurrent and longitudinal associations with children’s language skills. The analyses based on the Wave 1 data have been published in Tulviste and Tamm (2019). The current study aims to compare concurrent associations between maternal language teaching practices and children’s language comprehension and production at children’s ages 3;0 and 4;0. Moreover, the study investigates longitudinal and bidirectional associations between mothers’ language teaching practices and children’s language skills. It is known that mothers adjust their speech to the level of children’s language comprehension and production and that they speak more and use more advanced language with children whose language skills are better (Snow, 2014). This emerged also from the analysis of our first wave data: The Estonian mothers of 3-year-olds with better language skills reported a more frequent use of informal language stimulation, whereas mothers of children with poorer language skills used direct corrective feedback more frequently (Tulviste & Tamm, 2019). Therefore, we assess the directionality of the association: We use mothers’ language teaching practices at 3;0 to predict child language skills at 4;0 and children’s language skills at 3;0 to predict mothers’ language teaching practices at 4;0.

Mothers’ language teaching style was measured by asking mothers to indicate their frequency of employing different practices when talking with children in order to teach them language (see Tulviste & Tamm, 2019). Children’s language comprehension and production were directly measured via an examiner-administered New Reynell Developmental Language Scales IV (NRDLS, Edwards et al., 2011) to find out whether the two language teaching practices play somewhat different roles in the development of language comprehension and production.

When investigating longitudinal and bidirectional associations, we also take the child’s sex and mothers’ education into account. There is evidence that children’s language performance is strongly influenced by parents’ SES, especially their level of education (Fenson et al., 2007). Mothers with higher education have been found to speak in ways that support children’s language development more strongly: They tend to talk more with children, use a greater variety of words, fewer directives and engage children in back-and-forth conversations more (Fernald et al., 2013; Hart & Risley, 1995; Hoff, 2006; Rowe, 2018). Girls have been found to outperform boys in language performance (Bornstein et al., 2004; Eriksson et al., 2012; Fenson et al., 2007). Moreover, some studies show that parents talk more and are more responsive with girls (Leaper & Smith, 2004). For instance, Estonian 4-year-old girls heard a larger quantity of adult words per day than did boys (Tulviste & Tamm, 2021).

This longitudinal study was guided by the following research questions:

Research question 1. How similar are the associations between children’s language skills and maternal self-reported language teaching practices at ages 3;0 and 4;0 years?

Research question 2. Do maternal language teaching practices predict children’s language skills 1 year later when the child’s sex, maternal education, and the same prior language skill are controlled for?

We also checked the possibility that children’s earlier language skills predict later maternal language teaching practices.

Method

Sample

In this study, 76 mother–child dyads participated. One child was from an Estonian-Russian bilingual family, but Estonian was the dominant language in all families. The Estonian language is a Finno-Ugric language and is spoken by approximately 1 million people. Estonian is an agglutinative language, and it is characterized by a large number of cases (14 productive cases), no grammatical gender (either of nouns or personal pronouns), and no articles (see Tulviste & Schults, 2020). Children’s mean age at Wave 1 was 35.76 months (SD = 0.76) and at Wave 2, it was 48.35 months (SD = 0.58). There were slightly more girls (56.6%) than boys (43.4%). All children were reported by mothers to be developing typically. Family’s background information was collected only at Wave 1. Mother’s mean age was 32.07 years (SD = 6.33). Over half of mothers (65.2%) had higher education, 27.5% had secondary education, and the rest (7.2%) had basic education.

Materials

Language teaching practices

Maternal language teaching practices were assessed with 12 items describing various ways of communicating with a child (see Tulviste & Tamm, 2019). Eight items measured the use of informal language stimulation, and four items measured the use of direct corrective feedback to semantic, syntactic, and phonetic aspects of language. Mothers indicated how often they communicated with their child in ways described on a 4-point scale (1 = never, 2 = a couple of times per week, 3 = every day, and 4 = three times a day). Cronbach’s alphas for informal language stimulation at the two Waves were .76 and .75, respectively, and of the corrective feedback subscale .88 and .90, respectively.

Informal language stimulation

Read nursery rhymes and sing songs with the child.

Read books and look at pictures with the child.

Teach about directions and location.

Talk when doing household chores.

Repeat and extend child speech.

Follow the child’s lead when communicating.

Ask to talk about recent events.

Narrate our activities while playing.

Corrective feedback

9. Correct if uses wrong word for something.

10. Correct pronunciation.

11. Correct wrong inflectional endings.

12. Correct if omits a word.

Language comprehension and production

Children’s language comprehension and production were assessed with the New Reynell Developmental Language Scales IV (NRDLS, Edwards et al., 2011) by trained research assistants. Language comprehension was measured with 72 items and language production with 64 items. The Comprehension scale requires the child to demonstrate a comprehension of increasingly difficult verbal expressions ranging from labeling objects to complex sentences by following simple directions and more complex directions in play-based activities. The Production scale examines a production of the same vocabulary items and grammatical features. The average testing time was 45 minutes. NRDLS IV is for 2- to 7-year-old children, but the Estonian version of the test is normed for 3- to 4-year-old children. Cronbach’s alphas for different tasks in the Comprehension scale ranged from .68 to .74 and in the Production scale, from .79 to .88, the correlation between the scales was high, r = .81.

Procedure

The study was approved by the Research Ethics Committee of the University of Tartu. Participants were recruited through kindergartens and child care centers. Parents whose child belonged to the age group of our study were given information about the longitudinal study and asked to sign the informed consent forms if they decided to take part. The research assistants contacted the parents before the child’s third (Wave 1) and fourth (Wave 2) birthday to arrange the data collection at the participant’s home. The procedure of the data collection was the same at Waves 1 and 2. The research assistants tested children’s language development with the NRDLS at their home. Mothers were asked to fill in the questionnaires online. Alternatively, they could complete the questionnaires on article and return them to the research assistants in sealed envelopes.

Data analysis

Correlation analyses and comparisons of mean scores were conducted in SPSS 23.0. The variables were normally distributed. To test the hypotheses, we carried out a path analysis without latent factors in AMOS 23.0 (see Figure 1). We examined whether mothers’ reported use of informal language stimulation and corrective feedback predict concurrently and longitudinally children’s language scores (comprehension and production). We also checked whether there are longitudinal bidirectional associations between children’s language scores and mothers’ language teaching practices. Children’s sex and mothers’ education were used as predictors of children’s language and mothers’ language teaching practices. Due to the small sample, we used bootstrapping with maximum likelihood (2000 iterations, 95% confidence intervals). The fit of the model was evaluated by the comparative fit index (CFI; > 0.90 indicates good fit), the root mean square error of approximation (RMSEA; < 0.05 indicates good fit), and the χ²/df ratio (3 or less indicates good fit; Hoe, 2008).

Figure 1.

Associations Examined in the Path Model.

Results

The descriptive statistics and correlations between children’s language scores and maternal language teaching practices at Wave 1 and Wave 2 are shown in Tables 1 and 2. The scores of language comprehension, t(82) = 10.95, p < .001, Cohen’s d = 1.20, and language production, t(82) = 14.50, p < .001, Cohen’s d = 1.52 at Wave 2 were significantly higher than at Wave 1. There was a high degree of consistency in mother-reported language teaching practices: No significant differences were found in Wave 1 and Wave 2 scores.

Table 1.

Mean scores of language teaching styles and language scores at Wave 1 and Wave 2.

	Max	Wave 1		Wave 2		p
	Max	M (SD)	Range	M (SD)	Range	p
Corrective feedback	4	3.03 (0.83)	1.25–4	2.98 (0.84)	1.25–4	ns
Informal language stimulation	4	2.89 (0.46)	1.88–4	2.82 (0.44)	1.75–3.88	ns
Comprehension	72	47.62 (12.25)	3–65	58.71 (8.41)	27–72	***
Production	64	30.86 (14.91)	0–55	46.54 (11.03)	12–64	***

ns = insignificant differences between Wave 1 and Wave 2 scores.

***

p < .001.

Table 2.

Correlations between language teaching styles and language scores at Wave 1 and Wave 2.

	1	2	3	4
1. Comprehension	.65***	.61***	−.002	−.25*
2. Production	.84***	.78***	−.01	−.15
3. Informal language stimulation	.07	.03	.61***	.52***
4. Corrective feedback	−.19	−.24*	.46***	.70***

Wave 1 correlations are above the diagonal; Wave 2 correlations are below the diagonal; correlations between the same measure at Wave 1 and Wave 2 are on the diagonal in bold.

p < .05. ***p < .001.

Predictors of language scores

The results of the path model are shown in Tables 3 and 4, and in Figure 2. The model fit the data well: CFI = 1.0, RMSEA = .00 (90% CI .00, .17), χ²/df = 0.023. At Wave 1, the background variables and mothers’ language teaching practices explained 18.5% of variance in children’s language comprehension and 24.1% of variance in language production (see Table 3). Compared with boys, girls had higher language comprehension and production scores. Maternal education was positively linked to children’s language production. Maternal use of corrective feedback was related to children’s lower language comprehension. At Wave 2, the background variables and mothers’ language teaching practices did not predict children’s language scores.

Table 3.

Predictors of child language comprehension and production.

	Comprehension		Production
	Wave 1	Wave 2	Wave 1	Wave 2
	B (SE)	B (SE)	B (SE)	B (SE)
Child’s sex (ref. girls)	.27 (.11)*	−.02 (.10)	.24 (.11)*	.07 (.09)
Mother’s education	.25 (.14)	.06 (.10)	.42 (.11)**	.02 (.08)
Informal language stimulation W1	.23 (.12)	−.02 (.12)	.16 (.13)	.07 (.13)
Corrective feedback W1	−.31 (.11)*	−.08 (.15)	−.17 (.12)	−.22 (.12)*
Comprehension W1	−	.41 (.20)**	−	.25 (.14)*
Production W1	−	.46 (.20)*	−	.64 (.14)**
Informal language stimulation W2	−	.05 (.10)	−	.03 (.12)
Corrective feedback W2	−	.11 (.15)	−	.16 (.11)
R ²	18.5%	60.9%	24.1%	68.7%

B = standardized regression coefficients.

p < .05. **p < .01.

Table 4.

Predictors of mothers’ language teaching practices.

	Informal language stimulation		Corrective feedback
	Wave 1	Wave 2	Wave 1	Wave 2
	B (SE)	B (SE)	B (SE)	B (SE)
Child’s sex (ref. girls)	−.09 (.12)	.02 (.09)	.08 (.13)	−.11 (.09)
Mother’s education	−.04 (.15)	−.08 (.13)	−.03 (.14)	−.15 (.08)
Comprehension W1	−	.19 (.12)	−	−.04 (.14)
Production W1	−	−.16 (.14)	−	−.13 (.13)
Informal language stimulation W1	−	.56 (.12)**	−	.04 (.12)
Corrective feedback W1	−	.09 (.12)	−	.67 (.11)**
R ²	0.9%	41%	0.8%	57.4%

B = standardized regression coefficients.

p < .01.

Figure 2.

Significant Associations in the Path Model.

The strongest longitudinal predictors of Wave 2 language scores were the Wave 1 comprehension and production scores. While informal language stimulation at Wave 1 did not predict children’s language scores at Wave 2, mothers’ reported use of corrective feedback at Wave 1 was negatively linked to children’s language production at Wave 2. The background variables, language scores at Wave 1, and mothers’ language teaching practices at Wave 1 explained 60.9% of variance in children’s language comprehension at Wave 2 and 68.7% of variance in language production at Wave 2 (see Table 4).

Predictors of language teaching practices

There were no significant links between background variables and mothers’ language teaching practices at Wave 1 and Wave 2. The strongest predictor of Wave 2 language teaching style was the same score at Wave 1. Children’s language scores at Wave 1 did not predict mothers’ language teaching practices at Wave 2.

Discussion

The present two-wave longitudinal study addressed the concurrent and longitudinal associations between mothers’ self-reported language teaching practices and language skills of children at 3;0 and 4;0 years, while also taking children’s sex and mothers’ education into account.

Relationships between maternal language teaching practices and child language

The results indicated that at age 3;0 years, girls and children of mothers who reported less frequent use of direct corrective feedback had higher language comprehension scores, and girls and children of mothers with higher education obtained better production scores (see more details in Tulviste & Tamm, 2019). The findings provide support to the notion that children of parents with higher education and girls tend to outperform boys in language development (Bornstein et al., 2004; Eriksson et al., 2012; Fenson et al., 2007; Fernald et al., 2013; Hart & Risley, 1995; Rowe, 2018), but only at children’s age 3;0 (not at 4;0). Unlike in children at age 3;0 years, none of the factors considered here predicted concurrently children’s language comprehension and production at age 4;0 years. It has been relatively well-established that gender differences in language development are small, decrease with age, and do not persist beyond early development in many language domains (e.g. see Kidd & Donnelly, 2020; Rinaldi et al., 2021).

At the same time, maternal language teaching style at 3;0 did matter longitudinally. Mothers who reported a high frequency of using corrective feedback at children’s age of 3;0 had children with lower language production scores 1 year later. An important finding of this longitudinal study is that the strongest predictors of subsequent language skills turned out to be the language skills measured 1 year earlier. Specifically, prior language comprehension and production scores predicted later comprehension scores, as well as later production scores. There is plenty of empirical evidence that earlier language skills predict subsequent language proficiency (Golinkoff et al., 2019; Hart & Risley, 1995; Rescorla, 2009). Taking into account that recent theories emphasize that children’s active participation in social interactions supports their language development, poorer language skills may reduce children’s engagement in interactions that, in turn, results in lower language skills over time. In contrast, good language skills would likely promote social interaction. It is also possible that children with better language skills are more likely involved in interaction than those with lower language skills, and that may lead to their faster development. The reason why the scores on language comprehension and production were predicted by both language scores obtained 1 year earlier likely is that the NRDLS assesses first the comprehension and then the production of the same vocabulary items and grammatical features, and the two scores are strongly correlated (Edwards et al., 2011).

Maternal language teaching practices seem to matter in predicting language skills of younger children (at 3;0), but not of older children (at 4;0) who are more competent language users. The current study does not allow us to say whether it holds true also with children under 3 years of age. Concurrent associations at age 4;0 years did not replicate the prior findings of correlational studies that informal language stimulation has a positive, but overt corrective feedback has a negative effect on child language development (Dale et al., 2015; Tulviste & Tamm, 2019; Zhang et al., 2008). The different links between maternal language teaching practices and child language across ages raise an intriguing question: What teaching practices account for variation in children’s language skills at 4;0 and older? It is possible that mothers’ language teaching practices after 3 years of age are not as influential as at earlier ages. The variety of the language learning experiences that the child is exposed to grows when children start to spend more time outside the family and interact more frequently with peers, kindergarten teachers and other non-family members, and start to use digital media devices more frequently. It is important to note here that maternal language teaching practices measured in the present study were rather general and did not include all the various practices that might affect language learning at age 4;0 and older. For instance, the informal language stimulation scale did not contain items about talking during routine activities like bathing and eating, co-viewing TV, hearing parents’ talking, and some items about the use of decontextualized language, such as providing explanations, using rare words, and asking wh-questions. (Rowe, 2012). However, the longitudinal data are consistent with a growing longitudinal research evidence that early language skills play a crucial role in later language development, and because of that deserve special attention (Rescorla, 2009).

We also tested bidirectional associations between mothers’ language teaching practices and children’s language skills because it is likely that children’s poor language skills evoke mothers’ correction of language mistakes, while good language skills would elicit more informal language teaching. In this study, no effects of children’s language skills at 3;0 on mothers’ self-reported language teaching style used 1 year later were found. Thus, our results provide evidence that maternal language teaching practices shape children’s subsequent language skills, not the other way around.

The stability of mother-reported language teaching practices

The frequency of using the two language teaching practices depended on how frequently mothers reported to be using the same practices a year earlier. Neither child’s sex nor maternal education related to the frequency of maternal language teaching practices. The present study broadens this knowledge base by demonstrating the stability of maternal use of the two language teaching practices.

The finding of consistency in the reported frequency of using both language teaching styles at ages 3;0 and 4;0 suggested, among other things, that the self-report measure used in the study is reliable, and could be used in studies about language teaching practices. Most prior studies about parental speech features have used rather direct observation of real-life parent–child interactions than parental self-reports. A concern about self-report is that it can be difficult for mothers to answer accurately how frequently they use certain talking practices. In addition, maternal reports may be sensitive to their beliefs and knowledge about child language development and be related to mothers’ educational level and cultural background. At the same time, self-reports have some advantages, such as being less time-consuming and allowing to investigate bigger samples and get information about such language teaching practices that may not occur during observed conversations.

The effect of providing corrective feedback

To date, few studies have directly addressed the (negative) effect of providing corrective feedback on early language learning (Dale et al., 2015; Tulviste & Tamm, 2019). According to our data, the two aspects of language teaching are used with the same frequency and simultaneously as shown by their strong correlation with each other. Mothers who used one type of language teaching practices more also used the other one more frequently. It is likely that mothers provide feedback about the child’s errors also during informal language stimulation, but they do it in a way that does not necessarily interrupt the conversation (e.g. they may reformulate erroneous child utterances in the next conversational turn; Clark, 2018). Compared with informal language stimulation, providing direct corrective feedback breaks the ongoing conversation. Moreover, to learn from overt corrective feedback, metalinguistic abilities are needed to focus on language itself separately from the reality it is referring to. It could also be that some mothers place more importance to supporting their children’s language development than others. It is also possible that parental reports are biased: They may have provided socially desirable responses.

It is likely that the effect of overt corrections depends also on some other factors not addressed in our study (e.g. interactional and cultural context, the child’s temperament, and parents’ conversational style). It is also possible that the frequent use of overt corrective feedback and its negative associations with children’s language skills observed in our study reflect Estonian mothers’ tendency to be more directive than mothers with some other cultural background (Tulviste et al., 2003). However, more research including those with intervention designs should be done before changes in language teaching practices concerning providing less corrective feedback can be suggested. At the same time, the results clearly show that earlier language skills matter the most in predicting later language skills.

A limitation of the current longitudinal study is that it covers only 1 year. Because of that, the findings cannot be generalized to the entire preschool period. It is likely that in the end of preschool, children’s better cognitive skills allow them to learn also from explicit language teaching practices, such as corrective feedback. Another limitation is that for assessing mothers’ language practices only self-reports were used instead of observations of how mothers actually speak with children to teach them language.

Conclusion

Despite the limitations, the study adds new knowledge that helps to understand the mechanisms of language learning of children of different ages. Namely, maternal earlier self-reported corrective feedback negatively predicts children’s later language production, but previous language comprehension and production scores are the strongest contributors of later language skills. The finding highlights the importance of supporting language development during the first years of life.

Footnotes

Author contribution(s)

Tiia Tulviste: Conceptualization; Funding acquisition; Investigation; Methodology; Project administration; Supervision; Writing – original draft.

Anni Tamm: Formal analysis.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Research for this article was supported by the Estonian Research Council (grant no. PUT1359) and by the Netherlands Institute for Advanced Study in the Humanities and Social Sciences (NIAS-KNAW).

ORCID iD

Tiia Tulviste

References

Bornstein

Hahn

C.-S.

Haynes

(2004). Specific and general language performance across early childhood: Stability and gender considerations. First Language, 24(3),267–304. https://doi.org/10.1177/0142723704045681

Bruner

(1990). Acts of meaning. Harvard University Press.

Chaney

(1992). Language development, metalinguistic skills, and print awareness in 3-year-old children. Applied Psycholinguistics, 13, 485–514.

Clark

E. V.

(2018). Conversation and language acquisition: A pragmatic approach. Language Learning and Development, 14(3), 170–185.

Dale

P. S.

Tosto

M. G.

Hayiou-Thomas

M. E.

Plomin

(2015). Why does parental language input style predict child language development? A twin study of gene-environment correlation. Journal of Communication Disorders, 57, 106–117. https://doi.org/10.1016/j.jcomdis.2015.07.004

Edwards

Letts

Sinka

(2011). The New Reynell Developmental Language Scales. GL Assessment.

Eriksson

Marschik

P. B.

Tulviste

Almgren

Pérez Pereira

Wehberg

. . .Gallego

(2012). Differences between girls and boys in emerging language skills: Evidence from 10 language communities. British Journal of Developmental Psychology, 30, 326–343. https://doi.org/10.1111/desc.12019

Fenson

Marchman

V. A.

Thal

D. J.

Dale

P. S.

Reznick

Bates

(2007). MacArthur-Bates Communicative Development Inventories (2nd ed.). Brookes.

Fernald

Marchman

V. A.

Weisleder

(2013). SES differences in language processing skill and vocabulary are evident at 18 months. Developmental Science, 16, 234–248. https://doi.org/10.1111/desc.12019

10.

Golinkoff

R. M.

Hoff

Rowe

M. L.

Tamis-LeMonda

C. S.

Hirsch-Pasek

(2019). Language matters: Denying the existence of the 30-million-word cap has serious consequences. Child Development, 90(3),985–992. https://doi.org/10.1111/cdev.13128

11.

Hart

Risley

T. R.

(1995). Meaningful differences in the everyday experience of young American children. Brookes.

12.

Hoe

S. L.

(2008). Issued and procedures in adopting structural equation modeling technique. Journal of Applied Quantitative Methods, 3, 76–83.

13.

Hoff

(2006). How social contexts support and shape language development. Developmental Review, 26, 55–88. https://doi.org/10.1016/j.dr.2005.11.002

14.

Kidd

Donnelly

(2020). Individual differences in first language acquisition. Annual Review of Linguistics, 6(1),319–340. https://doi.org/10.1146/annurev-linguistics-011619-030326

15.

Leaper

Smith

T. E.

(2004). A meta-analytic review of gender variations in children’s language use: Talkativeness, affiliative speech, and assertive speech. Developmental Psychology, 40(6),993–1027. https://doi.org/10.1037/0012-1649.40.6.993

16.

Rescorla

(2009). Age 17 language and reading outcomes in late-talking toddlers: Support for a dimensional perspective on language delay. The Journal of Speech, Language, and Hearing Research, 52, 16–30. https://doi.org/10.1044/1092-4388(2008/07-0171)

17.

Rinaldi

Pasqualetti

Volterra

Caselli

M. C.

(2021). Gender differences in early stages of language development. Some evidence and possible explanations. Journal of Neuroscience Research, 1–11. Advance online publication. https://doi.org/10.1002/jnr.24914

18.

Rowe

M. L.

(2012). A longitudinal investigation of the role of quantity and quality of child-directed speech in vocabulary development. Child Development, 83(5),1762–1774. https://doi.org/10.1111/j.1467-8624.2012.01805.x

19.

Rowe

M. L.

(2018). Understanding socioeconomic differences in parents’ speech to children. Child Development Perspectives, 12(2),122–127. https://doi.org/10.1111/cdep.12271

20.

Rowe

M. L.

Snow

C. E.

(2020). Analyzing input quality along three dimensions: Interactive, linguistic, and conceptual. Journal of Child Language, 47, 5–21. https://doi.org/10.1017/S0305000919000655

21.

Snow

C. E.

(2014). Input to interaction to instruction: Three key shifts in the history of child language research. Journal of Child Language, 41, 117–123. https://doi.org/10.1017/S0305000914000294

22.

Stromswold

(2001). The heritability of language: A review and metaanalysis of twin, adaption, and linkage studies. Language, 77(4),647–723. https://doi.org/10.1353/lan.2001.0247

23.

Tulviste

(2019). Parenting: Talking with children across cultural contexts. In Tulviste

Best

Gibbons

(Eds.), Children’s social worlds in cultural context (pp. 135–147). Springer. https://doi.org/10.1007/978-3-030-27033-9_10

24.

Tulviste

Mizera

De Geer

Tryggvason

M.-T.

(2003). A comparison of Estonian, Swedish, and Finnish mothers’ controlling attitudes and behavior. International Journal of Psychology, 38, 11–18. https://doi.org/10.1080/00207590244000278

25.

Tulviste

Schults

(2020). Parental reports of communicative development at the age of 36 months: The Estonian CDI-III. First Language, 40(1),64–83. https://doi.org/10.1177/0142723719887313

26.

Tulviste

Tamm

(2019). Informal language stimulation rather than corrective feedback matters in Estonian children’s language performance. Learning and Instruction, 63, Article 101221. https://doi.org/10.1016/j.learninstruc.2019.101221

27.

Tulviste

Tamm

(2021). Is silence golden? A pilot study exploring associations between children’s language environment and their language skills in Estonian-speaking families. Journal of Experimental Child Psychology, 207, Article 105096. https://doi.org/10.1016/j.jecp.2021.105096

28.

Tulviste

Tõugu

Keller

Schröder

De Geer

(2016). Children’s and mothers’ contribution to joint reminiscing in different sociocultural contexts: Who speaks and what is said. Infant and Child Development, 25(1),43–63. https://doi.org/10.1525/eth.2004.32.1.34

29.

Vygotsky

L. S.

(1978). Mind in society: The development of higher psychological processes. Harvard University Press.

30.

Yang

(2012). Brain networks of explicit and implicit learning. PLoS ONE, 7(8), e42993. https://doi.org/10.1371/journal.pone.0042993.

31.

Zhang

Y. W.

Jin

X. M.

Shen

X. M.

Zhang

J. M.

Hoff

(2008). Correlates of early language development in Chinese children. International Journal of Behavioral Development, 32(2),145–151. https://doi.org/10.1177/0165025407087213