Inconsistent Paternal Behavior Predicts Turkish Immigrant and German Children s and Adolescents Mental Health

Objective: Data about Turkish immigrant fathers in Germany is scarce and it is not known how their parenting behavior may affect their children’s outcomes. The aim of this longitudinal study was to test whether Turkish immigrant and German fathers’ inconsistent discipline predicted their children’s mental health. Methods: Turkish immigrant (n=115) and German fathers (n=76) were administered the Inconsistent Discipline Scale items of the Alabama Parenting Questionnaire (APQ). One year later, fathers rated their children’s (N=191; age range 6-16 years; n=97 (50.8%) male) mental health with the Strengths and Difficulties Questionnaire (SDQ). Results: Self-reported levels of inconsistent discipline were the same among Turkish immigrant and German fathers. Turkish immigrant children had higher peer relationship problem scores compared with German children but there were no ethnic differences in total difficulties, emotional symptoms, conduct problems, hyperactivityinattention, and prosocial behavior. Multivariate regression analyses showed that both Turkish immigrant and German fathers’ inconsistent discipline predicted higher total difficulties (ß=0.24; p=0.001), hyperactivity-inattention (ß=0.20; p=0.007), peer problems (ß=0.21; p=0.005), and lower prosocial behavior (ß=-0.21; p=0.004) after statistically controlling for child gender and age, as well as paternal education. There were no effects of inconsistent behavior on emotional symptoms and hyperactivity-inattention. Conclusion: The negative effects of fathers’ inconsistent discipline on children’s and adolescents’ mental health are similar in the Turkish immigrant and German native population. Fathers’ use of consistent parenting behavior may be a potential target for parenting interventions aimed at improving youth mental health.


Introduction
The last three decades have seen increased research interest on father-child relationships worldwide [1,2] yet studies dealing with direct effects of fathers' behavior on their offspring's mental health are still scarce [3][4][5]. The majority of studies investigating effects of specific parenting behaviors on child outcomes have focused on mothers who are traditionally the primary caregivers. This is unfortunate as it has long been accepted that fathers are important environmental agents shaping their children's development [6,7]. Inconsistent parenting is one example of parenting behaviors that may be detrimental to children's outcomes. Mothers and fathers are the most important environmental agents for their children's socialization; parental responsiveness and accountability shapes how young children experience the world around them. If children learn that they can rely on their parents and that their own behavior will cause expected reactions then this may help build their trust in the world; they will learn to take responsibility for their actions and the reactions they cause. On the other hand, if children experience that they cannot predict how their parents may react to certain behaviors, or if deliberate, socially less desirable behaviors do not elicit moderating responses from parents, children may have difficulty defining their role in the world. Accordingly, inconsistent parenting has been linked to emotional and externalizing behavior problems in preschool [8,9] and primary school children [10][11][12], but only for mothers, not fathers [13]. One study has documented direct effects of fathers' inconsistent discipline on their children's and adolescents' conduct problems [14]. Symptoms such as inattention/hyperactivity, emotional, conduct, and peer problems limit general psychological functioning and adversely affect individuals' social relationships, long-term school success as well as life-long developmental outcomes [15][16][17][18]. It is thus timely to scrutinize the effects fathers' inconsistent parenting behavior may have on their children's mental health in order to identify avenues to and target groups for early intervention.
In addition to proximal effects of the family environment, children's mental health profiles may be influenced by their ethnicity [19,20]. In Germany, Turkish immigrants represent the largest ethnic minority [21] yet there is a lack of information about the factors that influence Turkish immigrant children's psychosocial adjustment. Studies conducted in the Netherlands found elevated levels of internalizing and externalizing problems among Turkish immigrant children and adolescents [22][23][24]. We have however recently reported that family adversity and mothers' inconsistent parenting rather than immigrant

Journal of Child & Adolescent Behavior
background were cross-sectionally associated with mother-rated mental health problems of Turkish immigrant and German youths aged 5-15 years [25]. As a result of their immigrant situation, Turkish mothers experience high levels of daily hassles [26], which may be associated with more inconsistent parenting [27]. However, the contributions of Turkish immigrant fathers to their children's mental health have yet to be investigated.
Apart from ethnic background and parenting, several factors have been associated with mental health profiles in children and adolescents. First, there are gender effects: overall, boys usually show more externalizing behavior problems whereas girls show more internalizing symptoms and more prosocial behavior [8,24,[28][29][30]. Second, there are age effects: some have reported that younger children tend to be more affected by behavior problems [28,29], whereas others found higher peer problems in older compared to younger children [30]. Third, there are effects of socio-economic status (SES): low SES children and adolescents tend to have less favorable mental health profiles compared with higher SES peers [28][29][30].
In the context of the available literature, we hypothesized that differences in children's and adolescents' strengths and difficulties would be predicted by fathers' inconsistent discipline rather than Turkish immigrant background after statistically controlling for child gender, age, and fathers' education.

Methods
The SIMCUR project (Social Integration of Migrant Children-Uncovering Family and School Factors Promoting Resilience), is a cohort-sequential longitudinal study on the development of Turkish immigrant children in Germany. In addition to Turkish immigrant families, non-immigrant German families were sampled for comparison purposes. Sampling took place in the Ruhr area, an industrial urban area in the North-West of Germany, in 2010 and 2011. Participants were screened via telephone by bilingual research assistants. Families were considered eligible for the Turkish sample if mother and father or the grandparents were born in Turkey; German participants were eligible if both parents were born in Germany and had German as their first language. In addition, in order to prevent confounding bias, children eligible for participation were born after 32 weeks of gestational age, were not living in a foster family, and did not have a referral to a special needs school. All parents that responded and fulfilled the inclusion criteria were included in the study. Families were allowed to choose if assessments took place in their homes or at the Ruhr-University in Bochum. Families were visited once per year by trained research assistants and fathers were administered questionnaires that were available in Turkish and German. Confidentiality was explained and parents and children signed consent forms. Families received € 25 compensation at each assessment wave.

Participants
A total of 191 fathers participated in the study and provided complete data at both time points (T1=first assessment; T2=second assessment one year later). Table 1 provides demographic information on the Turkish immigrant and the German sample. There were no significant differences with regard to child gender and age distribution, however, according to their ISCED-97 education level classification, the Turkish immigrant fathers were on average less educated than the German fathers.

Instruments
At T1, inconsistent parenting behavior was assessed with four items of the German version of the Alabama Parenting Questionnaire (APQ) [31,32] asking fathers how often the consequences they give for inappropriate behavior depend on their mood or how often they give their children a consequence and don't go through with it, for example. The APQ has high validity and reliability and was previously used for Turkish immigrants in Germany [33][34][35]; internal consistency was acceptable in the current sample (Cronbach's α=0.68). In addition, fathers answered a short demographic questionnaire.
At T2, the German and Turkish parent Strengths and Difficulties Questionnaire (SDQ) versions were administered to participating fathers. The SDQ is a short screening instrument containing 25 items that load on five subscales measuring emotional symptoms, conduct problems, hyperactivity-inattention, peer relationship problems, and prosocial behavior [36]. Each item is scored on a 3-point Likert-type scale (0=not true, 1=somewhat true, 2=certainly true). Except for prosocial behavior, higher scores indicate greater problems. A total difficulties score can be obtained by summing the scores of the four difficulties scales (range 0-40). The factor structure and validity of the German parent SDQ have been documented [28,29]. For this investigation, the relevant SDQ subscale items were subjected to an internal consistency analysis using the R statistical software package; items were treated as strictly categorical in order to ensure statistical accuracy accuracy [37]. In concordance with other studies [8,38,39], internal consistency coefficients (Cronbach's α) were moderate across both groups: Emotional symptoms α=0.71; Conduct problems α=0.77; Hyperactivity-inattention α=0.79; Peer problems α=0.61; and Prosocial behavior α=0.74. The internal consistency of the total difficulties score was α=0.61.

Statistical analyses
Data were analyzed with SPSS 22. Demographic and predictor variable comparisons between Turkish immigrant and German participants were carried out with Student's t-test or Chi-Square test, depending on distribution characteristics of variables. Descriptive statistics (means and standard deviations) for SDQ subscales and the total difficulties score are presented separately for Turkish immigrant and German participants. Multivariate linear regressions were performed on the whole sample (N=191) to analyze the prediction of children's strengths and difficulties by inconsistent discipline and ethnic background after statistically controlling for child gender, age, and paternal education. Table 1 shows that self-reported levels of inconsistent discipline were the same among Turkish immigrant and German fathers. According to the father ratings, Turkish immigrant children had higher peer relationship problem scores compared with German children. There were no ethnic differences in total difficulties, emotional symptoms, conduct problems, hyperactivity-inattention, and prosocial behavior.

Results
Multivariate regression models were computed on the whole sample (N=191) and confirmed our hypothesis: higher levels of fathers' inconsistent discipline predicted higher total difficulties scores one year later whereas there were no effects of Turkish immigrant background, child gender, age and fathers' level of education (please see Table 2). There were no interaction effects, i.e., effects of inconsistent parenting on child problem behavior were the same across Turkish immigrant and German father-child dyads.
On the subscale level we found that fathers' inconsistent discipline predicted higher levels of hyperactivity/inattention and peer problems as well as lower prosocial behavior scores across both groups. Fathers' inconsistency did however not significantly increase emotional symptoms and conduct problems. In addition, we found that girls showed less conduct problems and hyperactivity/inattention, and more prosocial behavior. Older children had, on average, more peer problems, and, except for peer relationship problems, there were no effects of Turkish immigrant background or paternal level of education on child behavior.

Discussion
This is the first study investigating effects of Turkish immigrant and German fathers' inconsistent parenting behavior on their children's strengths and difficulties. The results confirmed our hypothesis that differences in children's mental health are predicted by fathers' inconsistent discipline rather than Turkish immigrant background after statistically controlling for child gender, age, and paternal In accordance with other studies of similar populations [8,24,[28][29][30], boys showed more externalizing behavior problems whereas girls showed more prosocial behavior, and peer problems increased with age [28][29][30]. However, fathers' inconsistent parenting turned out to be the most consistent predictor of children's and adolescents' SDQ scores. Thus, contrary to previous findings, our results suggest that young immigrants' mental health problems may not be explained by their low socioeconomic background [40] but paternal use of inconsistent discipline behavior may contribute to behavioral difficulties, independent of ethnic minority status.
On average, Turkish immigrant fathers rated their children's peer relationship problems higher than German fathers. At first glimpse, this seems not surprising as stereotypes of Turkish immigrant youth in Germany often involve assumptions of aggressive and delinquent behavior patterns. This has however not been confirmed in any scientific study and our findings do not indicate the expected comorbid elevated conduct problems of Turkish immigrant youth. Moreover, Turkish immigrant mothers do not rate their children's peer or conduct problems higher than German mothers [25]. Thus future research will need to further scrutinize potential culturally specific peer relationship problems in Turkish immigrant youth in Germany.
Our results add to the existing literature on Turkish immigrants' parenting behavior: It has been suggested that Turkish immigrant parents tend to indulge and protect their children [31,41,42] and previous evidence has confirmed that Turkish immigrant mothers of 3year olds use more inconsistent parenting behavior than German mothers [27]. Turkish parents may however change and adapt their behavior as their children grow older, as there are no differences in the level of inconsistency between Turkish immigrant and German mothers [25] and fathers (as shown here) of 5-15 year olds. Even if indulging children during the early years is an adaptive and traditional form of parenting in many non-Western cultures [43], we suggest that the negative effects of inconsistency on child outcomes may be independent of a family's ethnic minority background but rather depend on the context of the majority culture a child is growing up in. Future studies should attempt to replicate and further scrutinize the effects of different parenting practices to better understand the mechanisms underlying associations between maternal and paternal behavior and children's psychosocial adjustment in different ethnic groups.

Strengths and Limitations
To our knowledge, this is the first longitudinal study assessing effects of inconsistent behavior on children's outcomes in a large sample of Turkish immigrant fathers and a native comparison group in Germany. We were able to replicate the factor structure of the German and Turkish parent SDQ versions in our sample of fathers who are, in general, an understudied population as they are often harder to reach and motivate to participate in longitudinal studies than mothers. The negative effects of inconsistency on children's mental health were consistent across four of our six outcome measures and across both Turkish immigrant and German father-child dyads who were characterized by a large socioeconomic diversity. Future studies with randomized samples from other populations and contexts are needed to assess the generalizability of our findings. Overall, the variance explained in our regression models was rather small, thus other variables not assessed and tested here may make important contributions in predicting behavioral difficulties.

Conclusion
Our study shows that Turkish immigrant and German fathers may negatively affect their children's mental health if they use high levels of inconsistent discipline. This is important for prevention and intervention efforts aimed at reducing children's internalizing and externalizing problems. In order to develop culturally-sensitive prevention strategies for youth health care practice more research is needed on the understudied group of Turkish immigrant children and adolescents in Germany. Fathers represent important targets for parenting interventions aimed at improving the mental health of the next generation. There are, for example, internationally valid and reliable parent training programs [44,45] but we don't know to what extent they would be accepted by Turkish immigrant fathers in Germany and if they would produce training-induced changes in parenting behavior that would affect children's mental health.