Maternal parenting styles, homework help, and children’s literacy development in language minority and Finnish-speaking families

The aim of this study was to examine the role of mothers’ (language minority mothers, LM, n = 49, and Finnish-speaking mothers, MP, n = 368) parenting styles and maternal help with their children’s homework in the children’s (mean age 11.43 years) literacy skills at fourth grade in Finland. In addition, the moderating effect of a child’s gender on this relationship was investigated. The results showed that the LM mothers used psychological control more than MP mothers. Furthermore, the more LM mothers used warmth and psychological styles of parenting, the more they helped their daughters, not sons, with homework. MP mothers’ parenting styles did not relate to their children’s reading and spelling skills. LM maternal behavioral control parenting styles were positively related, and psychological control was negatively related to children’s reading skills through help with homework. Maternal help negatively related to children’s reading skills in both groups. This research provides also new information to teachers and educators who work in multicultural teams. Because our findings could be applied in the educational settings of multilingual students, they are likely to be of great interest to the visionary scientists, researchers, teachers, and trainees.


Introduction
Interaction between parents and children is an important interpersonal aspect of children's achievement and learning in school (Pianta et al. 1997). Parental involvement at home affects children's school success (Castro et al. 2015). Talking with their child about school, supervision of homework, meeting with teachers, visiting their child's class, and collecting money for school trips are typical examples of involvement for both parents. These behaviors have been shown to predict positive outcomes for children, including academic skills (Hill and Taylor 2004;Pomerantz et al. 2007). Different parenting styles also affect children's acculturation (Prady et al. 2013) and integration in the classroom (Nguen et al. 2014), as well as in their new home country and its culture. In addition, mothers and fathers have different roles in parenting, and these roles vary in many countries (Newman et al. 2015). Home involvement does not always affect a child positively and may matter more for some children than for others (Pomerantz et al. 2007;Silinskas et al. 2013b).
Different parenting styles have been studied extensively and their effects demonstrated, but with mixed results. There has been less study on the parenting styles of parents in different language groups and their influence on children's academic skills. This study contributes to the existing literature by comparing the influences of two groups of mothers and children: language minority (LM) and majority population (MP). It examines the relations of parenting styles and maternal help with homework on the children's fourth-grade literacy skills: named decoding, reading comprehension, and pseudoword spelling in Finnish.

Parenting styles and children's academic skill development
Differences in parenting styles are largely understood based on a theory developed by Baumrind (1966). Parenting styles can be categorized into three types: authoritative, authoritarian, and permissive. These three types of parenting differ in two basic dimensions: affection (warmth) and control. Authoritative parenting is characterized by parental affection and behavioral control, which have been shown to be positively correlated with outcomes of children of various ages (Baumrind 1989;Hart et al. 2003;Maccoby and Martin 1983). Authoritarian parenting is marked by high behavioral and punitive control (Hart et al. 2003) and low affection. Permissive parenting is defined by high affection and lower behavioral control (Baumrind 1989).
Studies have shown that authoritative parenting has positive effects on children's school performance (Steinberg et al. 1989) and spelling skills in the first and second grades (Chao 2000;Karreman et al. 2006;Kiuru et al. 2012;Wang et al. 2006). Warm relationships support and encourage children toward autonomy at a young age and strengthen their social and academic functioning in early childhood (Landry et al. 2006). Fletcher et al. (2008) reported that the highest levels of positive academic achievement are linked to authoritative parenting, while the highest levels of social problems in children are linked to authoritarian parenting. Along with permissive parenting, authoritarian parenting has negative effects on school success (Aunola and Nurmi 2005;Chen et al. 2014;Coley et al. 2011).
In our view, the factors that affect behavior and, especially, academic skills in children from immigrant families are among the most interesting research targets (Hill 2001). Sparkman et al. (2015) found significant differences in the parenting styles of Hispanic and African-American parents in the USA: African-American parents were more likely to use warmth and support autonomy than Hispanic parents, while Hispanic parents were likely to use more coercion and rejection in their parenting. Chao (2000) investigated the parenting styles, home involvement, and socialization goals of Chinese immigrant and European-American mothers in relation to their first-and third-grade children's school achievements. Chao found that Chinese mothers endorsed more authoritarian and permissive parenting styles than European-American mothers. However, few studies have reported on the effects of parenting styles and maternal help on children's achievement in the LM context.
Common to the previous studies is that LM parents tend to be less educated than MP parents, and this factor may affect parenting. Nam and Park (2015) studied the thoughts and practices of 106 immigrant parents and their children regarding their schools' efforts to strengthen three aspects of parental involvement: parenting, communication, and learning at home. They found that mothers' educational levels correlated significantly to their children's English academic skills. The more the mothers were educated, the more they communicate and teach their child at home. Additionally, parents with a higher socioeconomic background (SES) have been shown to demonstrate more warmth in their parenting practices and to engage in more developmental activities (Melhuish et al. 2008).
In the current study, Finnish mothers (MP) and LM mothers' education and home languages are controlled for, and the relations of parenting styles and help with homework on reading and spelling skills are used to control for academic skill development. Little has been studied about the relationship between parenting styles and children's gender. Mileva-Seitz et al. (2014) noted that boys were more vulnerable to effects of early parenting and boys were more sensitive to warm parenting than girls. Another research aim of the current study is to examine the different relations of mothers' parenting styles and influence of homework help on girls or boys.
We appreciate that there is no one right way to parent or control a child. However, a parenting style with high warmth and caring control is generally associated with healthy and balanced child development (Darling and Steinberg 1993;Joussemet et al. 2008). On the contrary, children whose parents use a controlling parenting style were likely to require external control more than internal motivations (Lepper 1983). More information is needed about parenting styles and maternal help across equal socioeconomic language backgrounds and different cultural perspectives.

Home involvement and children's literacy skill development
In this study, children's literacy skills defined decoding, reading comprehension, and pseudoword spelling. The process of reading primarily requires performance of the alphabetic principle formed by letter-sound knowledge and phonemic skills, followed by maintenance of the detected phoneme in short-term memory, synthesis of all the read phonemes; and, finally, recognition of word meaning after generating the phoneme sequence (Goswami 2002). Literacy research has shown that spelling skills depend on linguistic factors such as word frequency (frequency effect) and lexical value (lexicality effect) (Defior et al. 2002;Snowling 2000). In Finland, the phoneme-grapheme correspondence method is commonly used in teaching reading and writing; Finnish language is orthographically very transparent, as a phoneme is almost always spelled or pronounced the same way. On average, most Finnish children read accurately before the end of the first grade (Holopainen et al. 2000).
Finnish education includes the preparatory Finnish language learning (Finnish 2) system. At least 900 h/year is provided for immigrant/LM children aged 6 to 10 and 1000 h/year for older children who come to Finland. After that, children begin basic Finnish education (Finnish National Board of Education 2015). These Finnish 2 language lessons are for those whose mother language is not Finnish or Swedish and whose parents speak a language other than Finnish. In Finnish 2 lessons, children have the opportunity to learn everyday Finnish vocabulary, reading, and writing skills and to practice the Finnish language naturally in different situations.
In educational research, the term "home involvement" includes either both parents' involvement (parental involvement) or mothers' sole involvement (maternal home involvement). Different home involvement practices have been identified as critical to defining children's social development in diverse cultural and ethnic groups (Aunola and Nurmi 2005;Chao 2000;Lansford et al. 2005;Steinberg et al. 1992). Tight, controlled parenting styles have been shown to negatively affect children, whereas effective socialization behaviors, such as the display of warmth and acceptance and suitable levels of control, have been shown to be associated with cognitive and behavioral competence in children (Baumrind 1966;Darling and Steinberg 1993;Hill 2001). Effective home involvement, such as help with reading, has proven to have a significant effect on children's academic skills (Sénéchal 2006) and motivation for learning (Blanch et al. 2013).
However, findings are contradictory. Some studies have shown a positive influence of home involvement on children's success in reading performance (Manolitsis et al. 2011;Pomerantz et al. 2007;Sénéchal and LeFevre 2002), rapid naming, and vocabulary (Torppa et al. 2006). However, other studies have shown that parental teaching has a negative effect on reading; that is, the more parents helped, the worse their children performed at school (Silinskas et al. 2010). Parenting styles have also been associated with children's school performance. Behavioral control increased the occurrence of task-avoidant behavior, especially in boys, and a warm parenting style reduced anxiety in girls regarding school tasks during their first school year (Kervinen and Aunola 2013). Parents tend to react to their children's poor performance by increasing the frequency of help, and children who perform poorly are more likely to ask for help from their parents (Green et al. 2007). Correspondingly, parents whose children perform well at school have been shown to help their children less than those whose children show poor reading skills (Ki and Soon 2012;Silinskas et al. 2013a). This study aims to determine the relation of different parenting styles and maternal help with homework on children's literacy skills, named decoding, reading comprehension, and pseudoword spelling in Finnish.

The present study
In the present study, the relations of mothers' parenting styles and home involvement on their children's literacy skills were assessed in two different groups: the LM group and the MP group. The following research questions were formulated: 1. To what extent do LM and MP mothers differ in their parenting styles and homework help? To what extent do the relations of maternal parenting styles and help with homework on children's literacy skills (named decoding, reading comprehension, and pseudoword spelling) differ between the LM and MP groups? 2. To what extent do the child's gender and mother's education influence the association between maternal parenting styles, maternal help with homework, and children's literacy skills (named decoding, reading comprehension, and pseudoword spelling)?

Method
Participants and procedure The current study is part of an extensive longitudinal study (AUTHORS 2006) in which an age cohort sample of Finnish children was followed from kindergarten (6 years of age, N = 1831) to grade 4 (11 years of age, N = 1887) to investigate the development of academic literacy skills (named decoding, reading comprehension, and pseudoword spelling), home involvement, and their associations with parent practices. Participants were collected from three medium-sized towns and one municipality whose administration is based on residents' self-government: two in Central Finland, one in Eastern Finland, and one in Western Finland. At the outset of the study, children's parents and teachers were asked for their written consent at the beginning of kindergarten.
A subsample of the present study comprised 417 fourth-grade children and their mothers. This sample of Finnish children was reduced systematically so group sizes were comparable to improve the statistical reliability of the study. As not enough fathers responded, we used only the mothers' answers in this study.
Children's reading and spelling skills were measured in a group test in April 2011 at the end of fourth grade by trained testers studying teacher education or psychology. The children's mothers were asked to complete questionnaires during the spring term of fourth grade (April 2011).

Majority population
The comparison group consisted of 368 Finnish-speaking children (190 boys, 178 girls, mean age of 11.49 years in fourth grade).

Measures
Decoding A group test was used to assess fourth-grade students' ability to decode a word-chain task (Nevala and Lyytinen 2000). The word-chain task was a time-limited test with ten rows, involving four to six words written together without any spaces in between. The children silently read the words comprising the chains and then noted the word boundaries by drawing a division line between the words. The score comprised the number of correct responses (maximum of 40) within the time limit set (1 min, 15 s). Cronbach's α for the tasks was 0.78.
Reading comprehension Children's reading comprehension and reading fluency were evaluated using a Test of Silent Reading Efficiency and Comprehension test (TOSREC) (Wagner et al. 2009; Finnish version by Lerkkanen and Poikkeus 2009). The group test was administered in the spring of fourth grade. The test contained 60 statements or sentences. Each child had 3 min to choose from two alternatives (true or false) and circle the correct option. Children received one point for each correct answer. The maximum score was 60, and Cronbach's α was 0.85.
Pseudoword spelling Spelling was assessed in grade 4 using a task with eight items consisting of one-syllable pseudowords (two items: kirk, vuil), three-syllable pseudowords (three items: märstö, saihdin, and raalsku), and four-syllable pseudowords (three items: voistoa, hiuruutti, and seivolssi) from the ARMI reading test battery . Each item was read orally twice. The score was the number of correct items (maximum value of 8). Cronbach's α for the tasks was 0.65.
Parenting styles Mothers' parenting styles were assessed using a Finnish version (Aunola and Nurmi 2005) of the Blocks' Child-Rearing Practices Report (CRPR; Roberts et al. 1984), which consisted of 17 items on a five-point scale (1 = does not fit me at all; 5 = fits me very well). The final summary variables are as follows.
(1) Warmth included ten items reflecting a positive relationship with the child ("I often tell my child that I appreciate what she/he tries or achieves"); Cronbach's α was 0.84. (2) Behavioral control included three items that reflected clear expectations of the child's behavior and the valuing of obedience in children ("My child should learn that we have rules in our family"); Cronbach's α was 0.69. (3) Psychological control included four items that reflected maternal attitudes toward guilt, pride, and expressing disappointment ( "I believe the child should be aware of how much I have done for him/her"); Cronbach's α was 0.75.

Maternal help
Mothers' help with their children's homework was measured using three questions: (1) Do you teach your child with his/her homework? (2) Do you help or guide your child with his/her homework? (3) Do you help your child with his/her reading-related homework? Answers were provided on a five-point scale (1 = never, 5 = always). The four-item scale was originally published by Silinskas et al. (2013b) and showed good internal consistency (0.88-0.91). In this study, we used three items for reading and writing, omitting the math item (Cronbach's α 0.76).

Results
The means and standard deviations of all variables for the LM (N = 49) and MP groups (N = 368) are shown in Table 1. An independent-sample t test (SPSS, Version 21) was first used to determine the difference between LM and MP mothers' parenting styles and maternal help with homework.
As can be seen in Table 1, while the LM mothers used more psychological control than did MP mothers, no significant differences were found between the groups in other measurements.
The correlations between all variables are shown separately for LM and MP in Table 2. In the LM group, the students' reading comprehension tests and mothers' help correlated negatively and positively between the decoding test and the reading comprehension test. Psychological and behavioral control also correlated positively. In the MP group, intercorrelations were low, except for the positive correlations between the reading comprehension test and the decoding test and between the uses of behavioral and psychological control.
A path model was constructed to answer the research questions on the relationship between maternal parenting styles, maternal homework help, and children's literacy skills. In the path model, all three literacy outcomes were regressed regarding maternal help with homework. Maternal help with homework was regressed on all three variables for parenting style. Correlations among literacy skills (reading comprehension, pseudoword spelling, and decoding) were determined. Maternal education was the control variable in all analyses.
Concurrent associations were allowed between all independent variables. These relations were analyzed separately for two groups in two distinct models, LM and MP, after a multigroup differentiate test (p < .005) showed it to be necessary. The indirect paths were estimated from each of three parenting styles to each of the literacy skills via maternal help. Finally, we applied a multigroup method in both models (LM and MP) to investigate the effect of children's gender and after that the effect of mother's education. Additional information was thus obtained on the relationship between maternal parenting styles, maternal help, and children's literacy skills. For the final models, non-significant paths were fixed to zero.
The The parameters of the models were estimated using a full-information maximum likelihood estimation with standard errors that were robust against non-normal distributions (Muthén andMuthén 1998-2012). As distributions of the variables were skewed, the model parameters were estimated using the MLR estimator, which produces standard errors and chi-square test statistics for missing data with non-normal outcomes by means of a sandwich estimator. For all models, goodness of fit was evaluated using five indicators: the χ 2 test, the comparative fit index (CFI), the Tucker-Lewis index (TLI), the root-mean-square error of approximation (RMSEA), and the standardized root-mean-square residual (SRMR). According to Hu and Bentler (1999), a non-significant χ 2 test, TLI and CFI values above 0.95, RMSEA values below 0.06, and SRMR values close to 0.08 can be considered indicators of good model fit to the data. CFI and TLI values below 0.90 and RMSEA and SRMR values above 0.10 are indications of a poor model fit (Kline 2005).

Mothers' parenting styles and maternal help in the language minority group
For the LM group with gender as a moderator, after fixing non-significant paths to zero, the model fit the data well: χ 2 (26 N = 47) = 38.74, p = .26, CFI = 0.92, TLI = 0.90, RMSEA = 0.08, and SRMR = 0.10. Results (Fig. 1) showed that mothers' warmth and psychological parenting styles correlated positively and behavioral control negatively to helping girls. The more warmth and behavioral control LM mothers reported concerning their relationship with daughters, the more they helped their daughters with homework. The results also showed that the less LM mothers used psychological control, the less they helped their daughters. In addition, the less LM mothers helped their children, whether girls or boys, the better their children succeeded on both the decoding and reading comprehension tests. Mothers' help and parenting styles did not influence success on the pseudoword spelling test.
The results also showed a significant positive indirect relation of LM mothers' behavioral control on reading skills (0.35, p < .01) and decoding skills (0.46, p < .01) with maternal help. There was also a significant negative indirect relation of mothers' psychological control on reading comprehension skills (−0.41, p < .01) and decoding skills with maternal help (−0.56, p < .001).

Mothers' parenting styles and maternal help in the majority population group
For the MP group, after fixing non-significant paths to zero, the model fit the data well: χ 2 (29 N = 269) = 25.75, p = .84, CFI = 1.00, TLI = 1.10, RMSEA = 0.00, and SRMR = 0.05. The results showed that the less MP mothers helped their children with homework, the better scores they had in reading comprehension, decoding, and pseudoword spelling. As for the research question "to what extent do children's gender and mothers' education influence the association between maternal parenting styles, maternal help with homework, and children's literacy skills (named decoding, reading comprehension, and pseudoword spelling)," the results showed that the higher the mothers' education level, the less they helped their daughters (Fig. 2).

Discussion
In the present study, we examined the relations of mothers' parenting styles and maternal help on their children's reading comprehension, decoding, and pseudoword spelling skills in LM and MP children. The primary result was that mothers' parenting styles and maternal help in these two groups turned out very differently and partly depended on the gender of the child. First, LM mothers used significantly more psychological control in their parenting than did MP mothers, and the relations of these differing parenting styles contributed to LM girls' reading comprehension and decoding skills. Second, behavioral control and warm parenting styles positively influenced maternal help, which in turn positively influenced reading comprehension skills, especially that of girls, in the LM group. Third, the higher the mothers' education, the less they helped their daughters. Overall, the average level of literacy skills for both groups at the end of fourth grade was high. This is partly due to the high quality of the Finnish 2 language system and Finnish education in general. Parenting styles and maternal homework helped influence children's reading comprehension, decoding, and spelling skills, showing that mothers' psychological control negatively influenced both maternal help and children's reading comprehension and decoding skills. This is consistent with the results of previous studies, which showed that psychological control appears to negatively influence children's school performance (Aunola and Nurmi 2005;Chen et al. 2000;Chen et al. 2014;Coley et al. 2011;Ginsburg and Bronstein 1993). Moreover, a warm mother-child relationship encourages and supports children and has a positive relation on children's academic skills (Landry et al. 2006).
The result confirms earlier research that demonstrated the positive relations of relative warm and authoritative parenting with moderate control, which also offers a positive outcome in school achievement (Steinberg et al. 1989;Cox and Paley 2003;Landry et al. 2006;Fletcher et al. 2008;Newman et al. 2015). The parenting style of LM mothers can be explained in terms of warm mother-daughter relationships and mothers' trust in their children's abilities (see Sikiö et al. 2016), which strengthen intrinsic motivation. This positive cycle enables a child's success with literacy.
What explanations can be found for the differences in the LM group? Previous findings suggest that the effects of psychological control can vary between different cultures, as mothers engage in dissimilar levels of parenting. LM parents (see Renzaho et al. 2011) may demand obedience, hold high expectations, scrutinize their children's behavior, and use discipline. The native Finnish population, on the other hand, seems to prefer a positive and warm parenting style. This cultural difference may manifest itself in questionnaire answers regarding psychological control; in this way, it could appear to negatively influence educational thinking. In this study, LM group children comprised 18 different countries, and their parenting styles differed from Finnish parenting styles, indicating the relations of different cultures on parenting styles. The explanation for different parenting styles could also be found in cultural differences surrounding raising children. Mothers' sensitivity to their children's learning Finnish and its difficulties could cause them to exercise more psychological control. Like the LM mothers in this study, mothers who cannot speak native Finnish properly have more difficulty helping their children with Finnish literacy tasks (Aram 2007). Different parenting styles may also be related to acculturation (Prady et al. 2013). The new home country, in this case Finland, could offer a limited or even poor culture that families do not want to identify with. This could influence both LM parents' knowledge of the MP school system and their ability to help with homework. For example, teachers give daily homework assignments to their pupils according to grade level. Homework is not required for preschool, but first and second grades usually require 30 min to 1 h of homework each day and 1 to 2 h for a child in third or fourth grade. Most homework tasks are designed to reinforce what has been taught in class, but in the upper grades, children are increasingly expected to acquire additional information at home by means such as the internet. Parents with insufficient knowledge of the Finnish language or educational system may experience difficulties in helping their children.

Warmth
Integration may facilitate the adaptation of LM groups to the MP (Nguen et al. 2014). The results of this study indicate that LM children were integrated to the Finnish school system from the beginning of their schooling, which likely explains their success on reading comprehension, decoding, and pseudoword spelling tests. A multicultural environment could also benefit different families and parenting styles by integrating new families into the Finnish school culture and homework traditions. For example, it is notable that psychological control is a norm in Asian culture (Pomerantz and Wang 2009). LM children from Asia are accustomed to hard discipline because it is a part of their culture. Mothers having psychological control is a well-respected and accepted parenting style in a large part of the world (Chao and Kanatsu 2008;Kim and Hong 2007). In this day and age, when children from different cultures increasingly move around the world, it is important to understand their background and how it influences parenting.
The results of maternal help regarding children's academic skills showed a negative impact on reading comprehension and decoding skills for both groups, and on the pseudoword spelling for the MP group. This means that the less mothers helped their children, the better their children's success in reading tests. This result is consistent with previous studies (Sikiö et al. 2016;Silinskas et al. 2013b), indicating that if the children are competent, there is no need to help them at home.
It is important to discuss the role of children's gender, mothers' education and parenting styles, and mothers' help with homework on children's reading and spelling skills. Different parenting styles influenced mothers' home involvement with girls in the LM group; the less LM mothers helped, the better the girls succeeded in both reading comprehension and decoding tests. In addition, the MP mothers' home involvement negatively influenced girls' performance on the pseudoword test.
Regarding the results of mothers' education, the higher the education of MP mothers, the less they helped their children (see also Ki and Soon 2012). In our study, an explanation for this could be that Finnish mothers tend to place high trust in the school system (Lerkkanen et al. 2013) and/or the level of mothers' education. The higher they were educated, the more they trusted their children's skills (Sikiö et al. 2016).

Limitations
There are some limitations in the current study that must be taken into consideration. The LM sample size was small, which could have affected the analysis and generalizability of the results. However, the total number of immigrant children is limited in Finland overall; 5.5% of all school children are immigrants, and 3.7% of immigrant children are immigrants who do not speak Finnish, which means that the proportion of LM children in this study is statistically representative of the Finnish context (Statistic Finland 2015).
In addition, the LM group was heterogeneous; the families represented 18 different languages, and therefore no generalizations can be made for any one language. It is notable that in this study, LM mothers used languages other than Finnish at home, but we did not record their Finnish language proficiency. We should have asked whether they used other languages (including Finnish) as well. Moreover, only mothers reported their parenting styles and levels of help with homework. Research demonstrates that mothers and fathers have different roles in parenting and that these roles vary by country (Newman et al. 2015). This aspect would have been interesting to study alongside the impact of fathers' parental involvement, but their response rate was too low. Finally, the mothers' support may have been biased either for or against the development of children's academic skills.

Practical implications
The number of immigrants and LM children in Finland has historically been quite low but has recently increased; this makes it even more important to examine the questions raised in this study in order to compare Finland's situation with that of other countries. We hope that the results will be taken into account when guiding and supporting relationships between parents, students, and teachers. Although parenting issues are complex and difficult, thanks to meeting with parents and teachers, the results of our study could raise discussions about parenting styles and their impact on academic achievement. The knowledge of orthographical differences is important for teachers when teaching children to read and spell. The change from deep orthographic system to transparent letter sound mapping system may cause difficulties for some LM children.

Conclusion
Mothers in the LM group used significantly more psychological control in their parenting than did MP mothers. Behavioral control influenced positively and psychological control negatively indirectly through the maternal help to reading skills in the LM group. The higher the mother's education, the less they helped their daughters. This study opens up a new way of thinking about parenting styles and their different influences on school success.