Influence of Maternal Praise on Developmental Trajectories of Early Childhood Social Competence

Ryoji Shinohara, Yuka Sugisawa, Lian Tong, Emiko Tanaka, Taeko Watanabe, Yoko Onda, Yuri Kawashima, Maki Hirano, Etsuko Tomisaki, Yukiko Mochizuki, Kentaro Morita, Gan-Yadam Amarsanaa, Yuko Yato, Noriko Yamakawa, Tokie Anme Research Institute of Science and Technology for Society, Japan Science and Technology Agency, Tokyo, Japan Center for Birth Cohort Studies, Interdisciplinary Graduate School of Medicine and Engineering, University of Yamanashi, Yamanashi, Japan Graduate School of Comprehensive Human Sciences, University of Tsukuba, Ibaragi, Japan College of Letters, Ritsumeikan University, Kyoto, Japan Clinical Research Institute, Mie-Chuo Medical Center National Hospital Organization, Mie, Japan Email: rshinohara@yamanashi.ac.jp


Introduction
Many studies suggest that positive parenting behavior is associated with the development of social competence in children.A secure parent-child attachment, which partly flows from the sensitivity and responsiveness of the caregiver, provides children with personal resources, such as emotional security and a sense of autonomy.Subsequently, these resources empower children to explore peer relationships (Ainsworth et al., 1978;Belsky et al., 1984;Lamb et al., 1985).Children derive working models from attachment relationships to their parents, that is, they internalize an enduring blueprint or a set of expectations concerning the operation of interpersonal relationships in general (Bowlby, 1969;Sroufe & Fleeson, 1985).In addition, parental warmth and engagement are associated with prosocial behavior among young peers (Attili, 1989;Hinde & Tamplin, 1983).Mothers who are constructive and agreeable tend to have children who exhibit similar forms of behavior in their own interactions with others (Putallaz, 1987).Inductive disciplinary styles, in which parents emphasize reasoning, are predictive of children's use of prosocial behavior (Becker, 1964;Zahn-Waxler et al., 1979).
Within positive parent-child relationships, "praise" is the one of the most important factors leading to the development early social skills.Previous studies on "praise" suggest that a parent's constructive verbal feedback is related to a child's achievement incentive in the early stage of development (Kamins & Dweck, 1999;Kelly et al., 2000); furthermore, a high frequency of praise is related to high self-esteem in the child (Felson & Zielinski, 1989;Kelly et al., 2000;Stipek et al., 1992).However, these reports focus on a 14-year-old child, who had been followed from 2 years of age.Few similar studies on toddlers exist in the literature.
With the increase in the antisocial behavior and aggression of school-aged children, the development of social competence in youngsters is of high interest.Health care and welfare specialists must urgently develop an evaluation method for assessing the development of social competence in children and for employing it to support childcare.However, Japan, possesses few such methods.Therefore, we developed a Japanese version of the Interaction Rating Scale (IRS), which can be easily and accurately used to assess several aspects of social competence in children.The reliability and validity of the IRS is found to be satisfactory in another study (Anme et al., 2007;Anme et al., 2010).
In a previous study, we examine the longitudinal relationship between the social competence of children at 18 months of age and the caregiver's attitude of "praise," with particular emphasis on the change in child attitudes at 4 and 9 months of age, after controlling for confounding factors.We find that the "praise" of caregivers influenced children's social competence at 18 months of age (Shinohara et al., 2009;Shinohara et al., 2010).In an effort to trace the impact of such praise over time, we go on to assess the children's social competence at 30 and 42 months and indicate the changes in their attitudes from the time when these youngsters were 4 and 9 months old.We place particular emphasis on the actual behaviors of caregivers extending praise, in a situation involving mothers and 18-monthold children.
We use a group-based semiparametric approach (Nagin, 1990: Nagin, 2005) to analyze the trajectory social competence development in children.This statistical technique enables us to evaluate subgroups of participants, multiple patterns of change in the outcome, and the shape of the change over time, etc.This model can be estimated using The SAS Trajectory Procedure "ProcTraj" described in Jones, Nagin, and Roeder (2001), and Jones and Nagin (2007).Further, Arrandale, Koehoorn, Mac-Nab, and Kennedy (2006) offer explanations of "ProcTraj".
The present study is longitudinal in nature and examines relationship between three variables: first, the trajectory of social competence in children from 18 months to 30 and 42 months; second, the actual behaviors of caregivers offering praise at 18 months; and finally, the change in the attitudes of children over time, beginning at 4 and 9 months.Demographic variables with the potential to influence this relationship are controlled for.The actual behaviors of caregivers offering praise to 18-monthold children are emphasized.

Participants and Overview
Data was obtained from 231 mother-child dyads enrolled in a longitudinal study of social development conducted as part of a project initiated by the Japan Science and Technology Agency (JST).The children and their parents were recruited from two Japanese cities (Mie and Osaka).Laboratory-based videotaped observations of all dyads were made three times, once when the children were18 months old in 2007, and the others at 30 and 42 months, in 2008 and 2009 respectively.However, the analyses are restricted to the 226 dyads for which the mothers provided demographic data and completed a questionnaire on their attitudes towards the importance of praise when their children were 4 months and 9 months old.All mothers gave informed written consent before participating in the study, the protocol of which was approved by the ethics committee of the JST.

Measures
Demographic data and attitudes regarding praise were obtained from the questionnaire.Demographic variables include the gender of the child, the presence of siblings, the family type (nuclear or extended), and the age of the mother (see Table 1).The item addressing the caregiver's attitude towards the importance of praise is termed "Praising child"; it has four response options: 1) Not at all important; 2) Of minor importance; 3) Fairly important; and 4) Very important (see Table 2).The IRS was used to assess social competence in children and child-rearing competence in mothers.This measure has 10 subscales, five addressing each of the child-related and caregiver-related factors * .Scores for all IRS subscales are derived from laboratory-based observations of mother-child interactions.The 36 items of the IRS are based on items of the Nursing Child Assessment Satellite Training (NCAST) teaching scales (Sumner & Spietz, 1994); we also refer to the Home Observation for Measurement of the Environment (HOME; Caldwell & Bradley, 1984) and the Social Skills Rating Systems (SSRS; Gresham & Elliot, 1990).

*
The child-related subscales are 1) Autonomy: Child initiates interaction with caregiver; 2) Responsiveness: Child is responsive to caregiver's behavioral cues; 3) Empathy: Child behaves in accord with caregiver's affective expression; 4) Motor regulation: Child's behavior is clearly directed toward the task and he/she is not overactive/underactive; and 5) Emotional regulation: Child adjusts his/her emotional state to a comfortable level.The caregiver-related subscales are 6) Respect for the development of autonomy Partner encourages child's autonomy; 7) Respect for the development of responsiveness: Partner encourages child's responsiveness; 8) Respect for the development of empathy: Partner encourages the child to develop empathy; 9) Respect for cognitive development: Caregiver encourages child's cognitive development; and 10) Respect for social-emotional development: Caregiver encourages child's social-emotional development.
For the actual behaviors of caregivers offering praise, we use the two following items of the subscale ( 7): (7)-1.Caregiver praises child's efforts at least once during the episode; (7)-2.Caregiver verbally praises child during the episode.

Mother-Child Interactions
Mother-child interactions took place in a playroom (4 m × 4 m) that was furnished with a small table and a child-sized chair and that contained toys (small dolls, mini cars, and plastic toys).This was a relatively natural setting, conducive both to spontaneous interaction and to a certain degree of experimental control.The play activity was standardized in being restricted to building blocks, an age-appropriate activity popular in Japan, and began when the each parent was given a box containing the blocks.Her child then took the blocks from the box, played with them, and then put them back in the box.Mothers were asked to play with their children as they would normally at home (providing instructions or help).The play activity lasted for 5 minutes.Three mother-child interactions were allowed in each task.Children who completed the task on the first attempt were not asked to perform additional tasks.Mother-child interactions were videotaped from five different angles (cameras were in each corner of the room and on the ceiling) and soundrecorded.

Coding of Mother-Child Interactions
Coders had to be sensitive to the cues of both mothers and children.In the latter, such clues were often ambiguous or confusing and quite transitory.Sensitivity was achieved via one month of training in the coding of videotaped parent-child interactions.The coders, doctoral or master's students, achieved at least 80% inter-rater agreement during training with pilot tapes.The reliability of the coding procedure was evaluated by having two coders each analyze a sample of tapes from the study (25% of the total) and was verified if the inter-rater agreement attained 87%.

Attitude Variables
Responses to the "Praising child" item were converted to a binary variable, with the response option "Very important" assigned a score of 1 (positive attitude) and all other response options assigned a score of 0 (negative attitude).From these values we construct a variable representing the transition of the attitude towards praise across the period at which the child was 4 months and 9 months old (see Table 3).This variable (Attitude variables) has four categories: 0, reflecting a negative attitude at both ages (consistently negative); 1, reflecting a change in attitude from positive at 4 months to negative at 9 months (inconsistently negative); 2, reflecting a change in attitude from negative at 4 months to positive at 9 months (inconsistently positive); and 3, reflecting a positive attitude at both ages (consistently positive).

Behavior Dummy Variables
The two items measuring praise (Behavior variables) are ( 7)-1 Caregiver praises child's efforts at least once during the episode (yes = 1, no = 0) and ( 7)-2 Caregiver verbally praises child more than once during the episode (yes = 1 no = 0).From these values we construct the dummy variables (Behavior dummy variables: X1, X2, X3), representing the occurrence of the two items (see Table 5).

Children's Social Competence
Child-related subscale items of the IRS were assigned a score of 1 if coded as "yes" and a score of 0 if coded as "no"; the summed score across all five subscales constitute the child's total score (maximum = 25).The summed score at 18, 30, and 42 months was calculated.

Statistical Approach
The Statistical Analysis System (SAS) statistical package (Ver.9.1) was used for the analysis.In performing this task, we calculated the descriptive demographic data, the attitude variables, the behavior variables, the behavior dummy variables, and the children's social competence at three points.The trajectories of children's social competence from 18 months to 30 and then to 42 months were estimated by using a group-based semiparametric approach (Nagin, 1990;Nagin, 2005).This method focuses on describing the trajectory, or pattern, of change over time in the dependent variables, and the ProcTraj estimates a regression model for each discrete group within the population (Arrandale, Koehoorn, MacNab, & Kennedy, 2006).The Bayesian information criteria (BIC) were used to decide on the optimum number of groups.In the ProcTraj procedure, the BIC values given in the output are negative; the best-fit model is the one with the smallest negative number (Jones, Nagin, & Roeder, 2001;Arrandale et al., 2006).After deciding on the optimum number of groups, we estimated the relationship between the trajectories of children's social competence, the attitude variables, and the behavior dummy variables.

Results
The demographic details, the attitude variables, the behavior variables, the behavior dummy variables, and the children's social competence at three points of the 226 mother-child dyads who participated in this study are summarized in Tables 1-6.
A group-based semiparametric approach (Nagin, 1990: Nagin, 2005) enables us to evaluate subgroups of participants, multiple patterns of change in the outcome, and the shape of the change over time.To analyze the trajectory social competence development in children and evaluate subgroups of participants, we calculated the BIC values from one to four groups.The optimum number of groups was two (the BIC = −1332.72)(see Table 7).Therefore, for the trajectory of children's social competence, two groups (Low group vs High Group) were adopted in this study (see Figure 1).There was a general trend for each child's social competence to increase with age.
In addition, to analyze the relationship between the trajectory social competence development in children from 18 to 42 months of age and the caregiver's behavior of "praise" when Note: (7)-1 Praise: caregiver praises the child's efforts at least once during the episode; (7)-2 Praise: caregiver verbally praises the child during the episode.
Table 5. Distribution of behavior dummy variables.

Behavior variables
Behavior dummy variables Note: (7)-1 Praise: caregiver praises the child's efforts at least once during the episode; (7)-2 Praise: caregiver verbally praises the child during the episode.The results show that after controlling for the effects of the demographic data, in the direction of the High group based on the Low group, the attitude variables (β = 1.203, p = 0.036) and the behavior dummy variables X3 (β = 2.619, p = 0.005) were significantly related to the trajectory of the children's social competence (Low group vs High Group).Additionally, for the demographic data, the results reveal that the gender of the child (Girls: β = 1.889, p = 0.026), siblings (Yes: β = 1.024, p = 0.045) was significantly related to the trajectory of social competence (Low Group vs High Group) (see Table 8).

Characteristics of This Research
This study is unique in the following respects.In Japan, there are few previous studies to assess the trajectory social competence development in infancy.We examined the longitudinalrelationship between the trajectory social competence development in children from 18 to 42 months of age and the parent's attitude towards the importance of praising the child when he/she was 4 or 9 months old, and the caregiver's behavior of "praise" when their children were 18 months of age.
Meanwhile, this research has limitations that hinder generalization because the participants were those who live in two cities.Moreover, the number of participants was low.However, the use of the results can be expected given that the subjects represent ordinary families in Japan.

The Trajectory of Children's Social Competence and the Caregiver's Behavior of "Praise"
The children were active and had more potential than we expected, although they could not have managed without their mothers' help and support.White (1959) explains the rich potential of competence encouragement in ensuring the healthy development of a child.Competence is defined as those "abilities [though which] the person is effectively related with the environment."In the development of a child's competence, his/her mother's response to him/her is important; that is, the However, regarding the parent-child interaction after infancy, many studies show that the more positive the parent-child interaction, the higher the child's level of social competence later in life.Mothers with high levels of social competence and their children display more positive behaviors and emotions (La-Freniere & Dumas, 1992), and the extent to which mothers give their children autonomy during play and teaching predicts positive assertiveness in preschoolers (Denham et al., 1990).This outcome is explained by the constructive response of mother to their children's emotions (Eisenberg et al., 1996); the latter then imitate these social and emotional behavioral patterns (Putallaz, 1987).
In a recent neuroscience study, 19 college students participated in functional magnetic resonance imaging (fMRI) experiments involving monetary and social rewards.Izuma et al. reported that the acquisition of a good reputation robustly activates reward-related brain areas, notably the striatum, and these overlap with the areas activated by monetary rewards (Izuma et al., 2008).
This research suggests that the attitudes of parents toward positive interactions with their child, particularly through the use of praise, has a very significant effect on the development of their children's social competence later in life.Such praise at 18 months influences the trajectory of social competence in children from 18 to 30 and 42 months.A follow-up study should be conducted to further clarify the mechanism underlying the effect of interactions involving praise, one that considers not just the praise-related behaviors of mothers but also those of fathers.It is hoped that caregivers and health care professionals will use these findings appropriately when providing parenting support to parents.

Table 2 .
Caregiver's attitude towards the importance of praise.

Table 3 .
Caregiver's change of attitude towards the importance of praise.

Table 4 .
Distribution of behavior variables.

Table 6 .
Child's social competence at three points of age.

Table 7 .
The optimal number of trajectories (Liner model).

Table 8 .
Multinominal logistic analyses for predicting child's social competence trajectories group memberships.