Parenting and Preschoolers

A large body of information on parenting practices in early childhood has been reported in the literature. Children at preschool years are becoming more independent but still need limits, supportive [falta sustantivo] and caretaking in a great amount of time (^{Chase-Lansdale, Wakschlag, & Brooks-Gunn, 1995}). In recent decades, most of the research has been focused on the mother as the primary child caretaker; however, in the last three decades, the father has been included in family research (^{Calzada, Eyberg, Rich, & Querido, 2004}; ^{Flouri, 2010}; ^{Lansford et al., 2014}; ^{Verhoeven, Junger, Van Aken, Deković, & Van Aken, 2010}). ^{Olhaberry and Santelices (2013)} have argued that the support of the father can contribute in a positive way to the family dynamic.

It has been suggested that a high quality of parenting (i.e. warmth parenting), which provides support and acceptance, develops prosocial relationships, empathy (^{Waller et al., 2014}), and effortful control in children (^{Eisenberg et al., 2005}; ^{Karreman, Van Tuijl, Van Aken, & Deković, 2006}). In contrast, negative parenting (i.e. harsh parenting), which involves inconsistent discipline, severe punishment, constant criticism, and failure to monitor, has been related to externalizing behaviors in preschoolers (^{Calzada et al., 2004}; ^{Gardner, 1989}; ^{Kroneman, Hipwell, Loeber, Kootand, & Pardini, 2011}; ^{Tung & Steve, 2013}; ^{Verhoeven et al., 2010}). Children with externalizing behaviors are at risk of peer rejection, academic failure and involvement in constant conflicts with authorities (^{Olson, Bates, Sandy, & Lanthier, 2000}).

Parenting and ODD Symptoms

Among externalizing behaviors, ODD symptoms have been found to be the most related with negative parenting (^{Deault, 2009}). This finding is based on the use of both clinical (^{Calzada et al., 2004}; ^{Kroneman et al., 2011}; ^{Tung & Steve, 2013}; ^{Burke, Pardini, & Loeber, 2008}; ^{Verhoeven et al., 2010}) and normal samples to measure ODD symptoms (^{Burnette, 2013}). On the other hand, positive parenting practices have been studied as the bidirectional relation between the parent and the child. For example, in a cross-lagged model, ^{Waller et al., (2014)}, reported a negative relation between positive parenting rearing and ODD symptoms. Based on that, it is a possibility that positive parenting could have an inverse effect, promoting a protective mechanism of ODD symptoms.

According to ^{Olson et al. (2000)}, harsh parenting on preschool children has important effects in later years and mostly predicts antisocial behaviors. They also reported a reciprocal influence between parents’ practices and children’s disruptive behavior, suggesting that negative maternal interactions with their child might have an effect at early ages. Given those approaches, it is important to explore a bidirectional relation between both parenting styles and ODD symptoms.

Parents’ EF and Parenting

Theoretical approaches assert that parent cognitions have a direct effect on parenting practices (^{Chase-Lansdale et al., 1995}; ^{Deater-Deckard, 2014}; ^{Wilson, Gardner, Burton, & Leung, 2006}). ^{Deater-Deckard (2014)} proposed a heuristic model of intergenerational transmission between parents and children. He suggested that executive functions (EF) work as a moderator of parents’ emotional and behavioral reaction to their child’s behavior. EF have been defined as a group of cognitive processes that are interrelated in order to achieve a future goal (^{Anderson, 2002}). Some of the EF are working memory, shift, planning, attention, inhibition, emotional control, among others.

^{Cuevas et al. (2014)} found that mother’s harsh parenting was negatively associated with a minor deficit of EF. Similar findings have been reported on harsh parenting among mothers with deficits in working memory (WM) (^{Deater-Deckard, Sewell, Ptrill, & Thompson, 2011}) and with low behavioral inhibition (^{Feng et al., 2007}). ^{Gonzalez, Jenkins, Steiner and Fleming, (2012)} reported that lower maternal sensitivity was associated with poorer spatial working memory and cognitive flexibility. Positive parenting has been also studied. ^{Gonzalez et al. (2012)} proposed a model examining the impact of maternal early experiences on maternal sensitivity through hypothalamic-pituitary-adernal (HPA) function and EF in 89 mothers and their infants 2-to-6-years old. They tested four mediational path analyses where they found that parental stress (HPA function) and EF are mediators linking early experiences to parenting. Furthermore, ^{Chico, González, Ali, Steiner, & Fleming (2014)} compared teenage mothers against adults, to test if poor EF development, as it is at such early ages, could be associated with EF and mothering. All mothers were approximately at 4-6 months postpartum. As expected, teenagers performed worse than adults on tasks of cognitive flexibility and were less sensitive in their infant interactions.

^{Deater-Deckard, Wang, Chen and Bell, (2012)} studied maternal EF, harsh parenting, and child conduct problems by testing 147 mothers of three-to-seven-year-old children. They found a relation between harsh parenting and child conduct problems only in mothers with a deficit of EF, highlighting the importance of the relations among those three variables.

Current Study

As summarized above, previous research has shown an important role of mother’s EF in predicting harsh parenting. An important relation between harsh parenting and ODD symptoms in clinical and normal samples has also been reported. To gain a better understanding of parenting behaviors, this study focused on a further analysis of the abovementioned relations, with the addition of effects of father’s EF and warmth parenting on ODD symptoms in a normal sample. The objective of this study was to analyze the relation between both parents’ EF, harsh and warmth parenting, and ODD symptoms. To do so, we analyzed two equation structural models. The first model was intended to confirm previous research findings on the relations between both parents’ EF and warmth and harsh parenting. We expected to find a positive relation of EF with warmth parenting and a negative relation with harsh parenting.

In the second model, we added a bidirectional relation between parenting styles and ODD symptoms. As previous research results have suggested, we expected to see a bidirectional and positive relation to harsh parenting and a negative relation to warmth parenting. Theoretical approaches about bidirectional influences between children behaviors and parenting are well documented (e.g. ^{Kiff, Lengua, & Zalewki, 2011}); specially when it comes to maladaptive child behaviors, they elicit more negative parenting practices.

Method

Measures

Analysis

Analysis on the descriptive data and correlations were conducted with SPSS version 19.0. It was decided to analyze these two models using structural equation modeling (SEM) because it allows to compare the relations between all variables at the same time. The advantage over the regression models is that SEM models are able to deal with measurement error (^{Iacobucci, 2009}).

The SEM was analyzed by using Mplus version 6.11 (^{Muthén & Muthén, 1998-2010}) with the maximum likelihood method of estimation. To determine the adjustment of the SEM models, the following adjustment coefficients were used: χ² statistics, comparative fit index (CFI), Tucker-Lewis Index (TLI), root mean square error of approximation (RMSEA), and standardized root mean square residual (SRMR). A model is considered to have good adjustment when the coefficients meet the following criteria: χ² ≤ ns, TLI and CFI of 0.9 (acceptable) and above 0.95 (excellent) (^{Hox & Bechger, 1998}), RMSEA cutoff ≤ 0.06 (^{Hu & Bentler, 1999}), and SRMR ≤ 0.08. The SEM technique have been suggested for analysis with large and small size samples (^{Hox & Bechger, 1998}; ^{Iacobucci, 2009}), the minimum size sample suggested have been over 100 participants (^{Iacobucci, 2010}).

Results

Table 1 shows descriptive variables between EF for both parents. Table 1 shows the correlations between ODD symptoms, harsh parenting, warmth parenting, and mother’s and father’s EF. ODD symptoms had a significant correlation with harsh parenting (r = 0.56, p < 0) but not with warmth parenting (r = 0.01, p < ns). ODD symptoms also had a significant correlation with father’s inhibition (r = 0.26, p < 0.05), as well as mother’s working memory (r = 0.35, p < 0) and inhibition (r = 0.3, p < 0.01). Harsh parenting was correlated with all three mother’s EF dimensions but not with father’s EF. Furthermore, warmth parenting was not significantly correlated with either parent’s EF. There was a significant correlation between all of the EF of the parents except for mother’s inhibition and cognitive shifting of the father.

Table 1 Summary of means, Standard Deviations and correlations for scores on ODD symptomologies, parenting styles and both parent’s EF 

Note. WM = Working memory; ODD = Oppositional Defiant Disorder. ***p < 0.001; **p < 0.01; *p < 0.05.

Figure 1 shows the first model, which had an excellent adjustment (χ² = 18.36, df = 16, p < 0.302, CFI = 0.98, TLI = 0.98, SRMR = 0.49, RMSEA = 0.03). Standardized betas are shown on the model. The only significant path was from mother’s EF toward harsh parenting (β = 0.87, p < 0).

Figure 1 Structural equation modeling of the relation between both parent’s EF and both warmth and harsh parenting styles. Straight lines denote significant relations. Dashed lines denote non-significant***p < .001 

Figure 2 shows the second model, which also had an excellent fit (χ² = 23.991, df = 21, p < 0.29, CFI = 0.98, TLI = 0.97, SRMR = 0.5, RMSEA = 0.03). As the previous model, standardized betas are shown on the paths. There was a significant path between mother’s EF and harsh parenting (β = 0.95, p < 0) and a trajectory from harsh parenting toward ODD symptoms (β = 0.69, p < 0). No significant difference was found in the paths from either parent’s EF toward warmth parenting or from ODD symptoms to harsh parenting.

Figure 2 Structural equation modeling of two latent and three observed variables. Rectangles denote responses of both parenting styles and ODD symptoms. Elipses denote latent variables for both parents’ EF. Straight lines indicate significant relations. Dashed lines denote non-significant. ODD = Opossitional Defiant Disorder***p < .001 

Discussion

In this study, we tested two models with the objective of finding relations between parents’ EF, parenting styles, and child ODD symptoms. Both models showed a path from both parents’ EF toward warmth and harsh parenting; the second model considered the bidirectional effects of parenting styles and ODD symptoms. We expected to find effects of both parents’ EF on their parenting styles, but it did not happen. The main finding was a significant path from mother’s EF toward harsh parenting, which affected the ODD symptomatology. Neither mother’s nor father’s EF had a significant path toward warmth parenting; father’s EF also showed no path toward harsh parenting.

The first model indicated that mother’s EF had a strong and significant path toward harsh parenting, whereas father’s EF did not predict any parenting style. We were expecting to find a relation between EF and both parenting styles (i. e. warmth and harsh parenting). However, those results were not found when it comes to father EF and a warmth parenting style.

Previous research has reported similar findings on mother’s EF and harsh parenting (^{Deater-Deckard et al., 2012}; ^{Cuevas et al., 2014}), citing that mothers were evaluated only because of their major involvement in childcare. Our results point to the importance of comparing mothers’—and not fathers’—EF. One possible explanation for the effect of mother’s EF on harsh parenting could be stress. According to ^{Lansford et al. (2014)}, having a large number of stressful events requires better executive functioning (^{Lansford et al., 2014}). ^{Calzada et al. (2004)} reported that mothers tend to experience more parental stress than fathers do. Given this result, it is possible that mother’s EF showed a strong relation with harsh parenting because of variables such as household chaos and other stressful events, which have been reported as mediators of such association (^{Deater-Deckard et al., 2012}; ^{Cuevas et al., 2014}). Thus, it is important to consider stress in future research so as to compare both parents’ measures (^{Cabrera, González, & Guevara, 2012}).

The strong effect between mother’s EF and harsh parenting opens up a discussion about how parent-cognitive-function might be involved on their own rearing practices. First, it is surprising that warmth-parenting practices did not seem to be related to parent EF as we would expected. This could mean that rejection and emotional warmth are not necessarily opposite dimensions divided into positive and negative parenting practices, but different dimensions of parenting styles.

On another point, it is possible that mothers reporting harsh parenting upbringing might be the ones that could report higher levels of executive dysfunction, but there is not an inverse effect, where the mothers reporting good EF performance are the ones reporting high levels of warmth.

After the second model was tested, we added to the first model ODD symptoms. We were expecting to find a bidirectional effect between parenting styles and ODD.

The main finding of this model was based on the significant path from mother’s EF to harsh parenting and then toward ODD symptoms. However, we did not find a bidirectional relation between parenting styles and ODD symptoms.

According to the relation between mother’s EF and harsh parenting, previous research has reported similar findings (^{Burnette, 2013}; ^{Tung & Steve, 2013}; ^{Verhoeven et al., 2010}). In particular, ^{Verhoeven et al. (2010)} found maternal behavior to be a more important predictor of child externalizing behaviors compared to paternal behavior.

On the other hand, we did not find a reciprocal relationship between parenting styles and ODD symptoms, as reported by ^{Burke et al. (2008)} and ^{Olson et al. (2000)}. Instead, we found a significant path only between harsh parenting and ODD symptoms. It is possible that we did not find such reciprocal relationship because we used a nonclinical sample, as did the other two studies. As ^{Gardner, Ward, Burton and Wilson (2003)} suggested, a mother’s negative attributions toward her child may be a cause of child behavioral problems. It could mean that strong symptoms of ODD could affect parenting practices in a major way; however, such bidirectional relation cannot be found in a normal sample.

As suggested by the model, we hoped to find a relation between warmth parenting and ODD symptoms, but that did not happen. ^{Olson et al. (2000)} and ^{Eisenberg et al. (2005)} found such relations in longitudinal studies. Specifically, ^{Olson et al. (2000)} studied children from 24 months to 17 years old and found that children at risk of externalizing behavior scored low on warmth and affective enjoyment with the mother. ^{Eisenberg et al. (2005)} worked on a longitudinal study of three waves: first, when the children were 3 years old; then, two years later; and finally, six years later. They found that warmth parenting was a predictor of child externalizing problems; this was mediated by child effortful control. This prediction was only found nine years after the first measure, not two years later when children were in preschool years. Comparing that finding with ours might enhance the supposition that positive parenting could have an effect over the long term and not in a short term, which, however, needs further research to be confirmed.

Conclusions

This family study provides information about the relations between parents’ EF, parenting styles, and ODD symptomatology. The major finding is based on the relation between mother’s EF and harsh parenting and its effect on child ODD symptomatology. ^{Lansford et al. (2014)} suggested that there is a need to eliminate parents’ violence toward children, citing the common use of harsh physical and nonphysical forms of discipline. Our findings support the importance of preventing harsh parenting, especially in early childhood, a stage in which it is a strong predictor of antisocial behavior in later years (^{Rhee et al., 2012}).

Limitations and Future Directions

This study has a number of limitations. First, it is a cross-sectional study that focuses on relations between variables. As discussed previously, a longitudinal study is necessary to determine whether ODD symptoms during preschool years can develop into conduct disorders because of harsh parenting or if children develop cognitive deficits as reported before (^{Araujo, Jané, Bonillo, & Capdevilla, 2014}). Second, according to ^{Deater-Deckard et al. (2012)}, environmental stress is an important mediator variable for mothers; controlling environmental stress at home or personal stress for both parents could be a determinant in relating the effects on EF or harsh parenting. As a future direction, it could be useful to assess prenatal and perinatal risk factors because of the major development that occurs at those stages (^{Latimer et al., 2012}). Likewise, comparing our findings to those from a clinical sample can help determine bidirectional effects between parenting styles and ODD symptoms.