The Relative Stability of Parental Overprotection and its Influence on Anxiety in Infants
Research Master Educational Sciences Thesis 2
Rosanne B. van den Born
Supervisors: dr. M. Majdandzic, & prof. dr. S. M. Bögels University of Amsterdam August 2012
Abstract
The current study aimed to provide more insight in the stability of overprotective parenting behavior of anxious and non-anxious parents and its influence on anxiety in infants. The research sample included 151 fathers and 151 mothers and their first born child, who participated in the first two measurements of an ongoing longitudinal study. At T1 children were 4 months old and at T2 children were 12 months old. Parental anxiety disorder was assessed prenatally using a semi-structured interview. At T1 and T2 fathers and mothers were each separately observed with their child during several structured tasks in the research centre and at their home, and overprotection was coded on a 5-point Lickert scale. In addition, child anxiety was based on parent reports at T2. Results demonstrate modest stability for observed parental overprotection over time, with no differences in stability between fathers and mothers. Mothers scored significantly higher on observed overprotection than fathers at both measurement occasions. Contrary to our expectations, no significant differences were found between anxious and non-anxious parents in their stability of overprotection. In addition, no significant differences were found in child anxiety between children of continuously high, low, or unstable overprotective parents. Future research should investigate the stability of overprotective parenting and its influence on child anxiety in older children, because it could help to adjust prevention and intervention programs.
Introduction
Mounting research shows that anxiety disorders are among the most common mental disorders during childhood and adolescence (Benjamin, Costello, & Warren, 1990; Boyd, Gullone, Kostanski, Ollendick, & Shek, 2000; Kashani & Orvaschel, 1990; Last, Perrin, Hersen, & Kazdin, 1996). In addition, it has been demonstrated that specific parenting behaviors, like overprotection, are associated with child anxiety (e.g. McLeod, Wood, & Weisz, 2007) and that parents’ own anxiety disorder can be associated with parenting behaviors that enhance child anxiety (Ginsburg & Schlossberg, 2002). However, little is known about the stability of these parenting behaviors and what its influence is on anxiety of the child. In order to gain more insight in these processes, the current study investigated the stability of parental overprotection and its influence on child anxiety. In addition, potential differences between fathers and mothers in overprotection over time were examined. Furthermore, it was investigated if parents own anxiety disorders are associated with continuously higher levels of overprotection over time.
The relation between parental overprotection and child anxiety
Research shows that there are two dimensions of parenting that are related to child anxiety: rejection and control. Parental rejection refers to the degree to which parents criticize and disapprove of their children against the degree to which they accept them and show them warmth and affection (Clark & Ladd, 2000; Maccoby, 1992; Rapee, 1997). Parental control refers to the degree to which parents overprotect their children and regulate their children’s behavior, thinking processes, and emotions against the degree to which they stimulate their children’s autonomy (Barber, 1996; Clark & Ladd, 2000; Pettit, Laird, Dodge, Bates, & Criss, 2001; Rapee, 1997). Various meta-analyses and reviews found associations between child anxiety symptoms and parental rejection and control (Ballash, Leyfer, Buckley, & Woodruff-Borden, 2006; McLeod et al., 2007; Van der Bruggen, Stams, & Bögels, 2008). The current
study focused on the dimension of control, because it has been demonstrated that it is more strongly associated with child anxiety than parental rejection (McLeod et al., 2007; Rapee, 1997). More specifically, the present study focused on overprotective parenting behavior, which is a part of the parental control dimension. It refers to parental behavior that is excessively protective and restricts children to engage in situations or behaviors that parents consider as potentially unsafe (Rapee, Schniering, & Hudson, 2009).
During early childhood the first relationship experiences form models for children’s patterns of self-regulation. Parenting is an essential factor in this period, since it has a strong influence on children’s development (Maccoby, 2000; Sroufe, 2000). For this reason the aim of the present study was to investigate the stability of overprotective parenting behavior of both anxious and non-anxious parents and its influence on anxiety in infants. By understanding the nature of the interaction between overprotective parenting behavior and anxious child behavior over time, an important step may be taken for risk assessments and interventions for this early developmental stage.
The stability of parental overprotection over time
Little is known about the stability of controlling or overprotective parenting behavior. However, as Holden and Miller (1999) state, longitudinal variability in parenting behaviors should be taken into consideration when analyzing child rearing. Not only to improve the understanding of the nature of parenting behaviors, but also to gain more understanding of the influence parents have on their children’s development. For example, O’leary, Slep, and Reid (1999), who investigated the stability of parenting behaviors of depressive parents, indicate that we need to know if negative parenting behaviors are a stable trait of the parent and not just a temporary approach to handle short-term developmental changes in the child. More generally, knowledge about the stability of parenting behaviors is important when analyzing changes in parenting behaviors related to non-ordinary situations (for example: life events,
illnesses, treatments etc.). Differences in parenting behaviors that are observed during these situations may partially result from the normal changeability of parenting and not because of the influence of the non-ordinary circumstances (Rimehaug, Wallander, & Berg-Nielsen, 2011).
Stability of parenting behavior can be analyzed in many forms; one can measure the stability over time, between different children in a family, and between contexts (Holden & Miller, 1999). The present study focused on the stability of overprotective parenting behavior over time. More specifically, it focused on relative stability of overprotection, which refers to the stability of a parent’s individual position within the research group on overprotection over time. It shows to what degree a parent keeps his or her position over time, relative to other parents on overprotective parenting behavior (Holden & Miller, 1999; Forehand & Jones, 2002). This type of stability takes into account children’s developmental stage when analyzing parental behavior over different measurement occasions (Holden & Miller, 1999).
In their review Holden and Miller (1999) found that parents’ rank orders on parenting behaviors are stable over time. Additionally, other research has indicated that overprotection is a stable parenting behavior over time. For example, Rimehaug et al. (2011) found moderate stability on overprotection in parents of 8 to 15 year old children, over a time period of nine months. In addition, Kennedy, Rubin, Hastings, and Maisel (2004) demonstrated that mother’s over-controlling behavior, when children were 2 years old, significantly predicted their over-controlling behavior when their children were 4 years old. They also found moderate stability. However, it should be noted that their studies included older children than those who were investigated in the current study and that their studies were based on overprotection measured with a questionnaire, whereas the current study used observational data to measure the degree of overprotection.
It has been suggested that fathers and mothers have different roles in childrearing. Where maternal parenting behavior is characterized by calming and comforting behavior, paternal parenting behavior is more aimed at exiting, surprising, and destabilizing their children (Paquette, 2004). For this reason it could be theorized that mothers’ parenting behavior could easier result in overprotective parenting behavior than that of fathers (Majdandžić, Möller, Bögels, & Van den Boom, 2011).
Few studies have investigated the difference in stability between paternal and maternal parenting behavior, since most of the studies on parenting stability only included mothers in their samples (Holden & Miller, 1999). However, it has been demonstrated that fathers and mothers show similarity in parenting stability across 1 year in toddlers (Verhoeven, Junger, Van Aken, Decović, & Van Aken, 2007). In addition, Rimehaug et al. (2011) showed that overprotective parenting behavior of parents with 8 to 15 year old children did not differ between fathers and mothers across 9 months.
The influence of parents own anxiety on their overprotective parenting behavior
Parental control and overprotection have been associated with parents’ own anxiety (e.g. Adam, Gunnar, & Tanaka, 2004; Bögels & Brechman-Toussaint, 2006; Edwards, Rapee, & Kennedy, 2010; Ginsburg, Grover, & Ialongo, 2004; Whaley, Pinto, & Sigman, 1999). Since anxious parents are likely to foresee more potential danger, they may demonstrate more controlling behavior in order to avoid unsafe situations (Rapee et al., 2009). Because of the restrictions they place on their children in a wide range of behaviors and situations, parents not only communicate to their children that the world is an unsafe place, but they also prevent their children from having mastery experiences and enhance their dependency on caregivers (Rapee, 1997; Wood, 2006). However, the relationship between parental control and parental anxiety still remains questionable. Van der Bruggen et al. (2008) only found a small and non-siginificant effect between the two variables in their meta-analysis.
The current study
The present study examined four research questions. First, the stability of parental overprotection over time was analyzed for fathers and mothers of 4 to 12 months old children. Overprotective behavior was measured at both occasions with several structured tasks in the research centre and during home visits. Based on previous research it was expected that there would be considerable stability in parental overprotection and that fathers and mothers would not differ in their stability on overprotection over time. Second, it was examined whether parental overprotection differs between fathers and mothers at both measurement occasions. Although it could be hypothesized that maternal parenting behavior could easier result in overprotective parenting behavior, the outcomes over time are still unclear. The third aim of the current study was to examine if there are differences between anxious and non-anxious parents in their stability of parental overprotection over time. Do anxious parents score continuously higher on overprotection in comparison to non-anxious parents? Parents’ anxiety status was measured with a semi-structured interview when they were still expecting their first child. Finally, it was investigated whether child anxiety depends on the stability of parental overprotection. Child anxiety was measured with a questionnaire when children were 12 months old. It was expected that children who were exposed to continuously high levels of overprotective behavior of their parents would be more anxious than children of parents who scored continuously low on overprotection. In addition, it was expected that unstable overprotective parenting behavior would result in more child anxiety than that observed for children of parents who show continuously low levels of overprotection.
Method
Participants
The current study was part of a larger ongoing longitudinal study on social anxiety in young children with five measurements in total. Participants were couples and their first born child.
They were identified through leaflets provided at midwives’ practices and baby shops and through advertisements on internet forums and in magazines and newspapers, when the mothers were still pregnant. Of the 210 couples that were identified, 151 were included in the study and participated in the pre-measurement. Participants were included if they mastered the Dutch or English language and if the child had a birth weight of at least 2500 gram, no neurological disorders, and an Apgarscore of at least 8. The current study used data from the pre-measurement (prenatal) and the first two measurements occasions (T1 and T2) of the longitudinal study, when children were respectively 4 months (T1) and 12 months (T2) old. At T1 128 fathers and 125 mothers participated in the measurement. At that time 69 girls and 59 boys were included in the sample. At T2 124 fathers and 123 mothers participated in the measurement and the sample existed of 69 girls and 55 boys. During the pre-measurement mothers were on average 30.7 years old (SD = 4.17) and fathers on average 33.5 years (SD = 5.25). The average level of education was fairly high; for mothers M = 7.11 (SD = 1.14) and for fathers M = 6.57 (SD = 1.59) on a scale from 1 (= primary education) to 8 (= university). Most of the participants had a Dutch background (89.39 % of the mothers and 95.42 % of the mothers).
Procedure
During pregnancy, couples visited the university research centre together and individually took part in a semi-structured interview to establish their diagnostic anxiety status. After the child was born, fathers and mothers separately visited the research centre with their child on both measurement occasions. Subsequent to each measurement in the research centre a home visit took place to enhance the ecological validity. Parents and their children performed several tasks to measure parenting behavior and child anxiety. In addition, parents filled in a questionnaire about anxiety of their child. The current study only used data from the tasks that measured parental overprotection at T1 and T2 and of parent-reported child anxiety at T2.
Measures before birth of the child Parental anxiety
Parental anxiety was measured with the Anxiety Disorder Interview Schedule (ADIS; Di Nardo, Brown, & Barlow, 1994). This semi-structured interview measures current and lifetime anxiety disorders. The assessments were carried out by two experienced and trained interviewers with a master’s degree in Psychology. They could use their clinical impression to fine-tune the answers of parents in order to specify his/her diagnostic status. If there was uncertainty about the diagnostic status, an experienced psychotherapist was consulted to resolve that case. Parents were categorized in a group of non-anxious parents and in a group of parents with a current or lifetime anxiety disorder.
Measures at 4 months, Time 1 Parental overprotection
Overprotective parenting refers to parental behaviors that are too cautious and/or restrict children to take part in certain behaviors or situations (Rapee, et al., 2009). In the present study overprotective parenting was scored on a dimension which was (to some degree) based on the involvement dimension of Hudson and Rapee (2001). Examples of overprotective behaviors are: supporting the child when it is not needed or holding the child too tightly. In addition, verbal expressions like: ‘Careful’ and ‘Watch out’ are also indicators of overprotection. At T1 overprotection was measured with nine tasks in total.
In the research centre two free play tasks, a dance task, and three movement tasks were conducted. For the first play task a play mat was put on the floor along with a rattle, a teething ring, a squeaky duck, and a soft block. Parents were asked to play with their child for a while, in a way as they would do when they were at home. During the second play task parents were asked again to play with their child for a while, but this time without toys. Both tasks lasted
2.5 minutes. For the dance task parents were asked to dance with their child to a song that was played in the room for about three minutes. During the first two movement tasks the child lay on a changing mat on a table and the parent sat in front of the child on a chair. First, the parent was instructed to bring the feet of the child to the hands of the child as if he/she could grab them and then the parent was asked to suddenly let the feet go. Secondly, the parent was asked to move the child into sit position, without further instructions on how to do so. The parent had to repeat this once more. For the third movement task, the parent was asked to hold the child around the chest and back and lift the child above his/her head as if he/she was going to play a flying game with the child. The parent had to repeat this task once more.
During the house visit overprotection was measured with a 5 minute free play task and two movement tasks. For the play task the father and mother were asked to separately play with their child for a while in way that they could decide for themselves, as long as they did not use toys and tried to hold the child’s attention. The movement tasks at home were conducted with both parents together. For the first task, parents were asked to give the child to each other in a difficult way. They had to place the child with his/her stomach on the arm of the other parent and to hold the child in that position for a while. For the second task parents were asked to swing the child back and forth in a cloth for a while. During both movement tasks parents were allowed to talk to each other and the child.
Measures at 12 months, Time 2 Child anxiety
Child anxiety was measured with the fear dimension of the Infant Behavior Questionnaire- Revised (IBQ-R). The IBQ-R is a standardized instrument developed to measure infant temperament. The fear subscale measures the degree to which children demonstrate distress to unexpected changes in stimulation, novel situations or people. Example items were: ‘How often during the last week did your child startle to a sudden or loud noise?’ and ‘When an
unfamiliar person came to your home or apartment, how often did your baby cry when the visitor attempted to pick her/him up?’ (Gartstein & Rothbart, 2003). Children were rated by their father and mother on a 7-point Lickert scale, ranging from 1 (never) to 7 (always). The internal consistency of the fear scale was excellent in this study, with Cronbach’s alpha = .88 for mothers and Cronbach’s alpha for fathers = .89. Gartstein and Rothbart (2003) have indicated that the IBQ-R is a valid instrument to measure children’s temperament. Although temperament is often thought of as an innate characteristic of the child, there is evidence that genes only account for 40-50% of the variability in temperament. Furthermore, twin and adoption studies have found evidence for genetic as well as environmental influences on individual differences (Rothbart, 2011). Therefore, the fear subscale of the IBQ-R was used in the current study, to examine if overprotective parenting behaviour influences child anxiety.
Parental overprotection
At T2 overprotection was measured with ten tasks in total. In the research centre two 5 minutes free play tasks and a dance task were conducted. For the first play task age appropriate toys were used, including a glove with finger puppets, a book which could make different animal noises, a soft ball, magnetic building blocks, and several pillows to provoke potential rough and tumble play. During the second play task the test leader put away the different toys, but left the pillows in the room. For the dance task parents were asked to dance with their child to a song that would be played in the room for about three minutes.
The house visit at T2 included a free play task, three movement tasks in which both parents had to cooperate, and three individual movement tasks that were conducted with the father and mother separately. For the first movement task parents had to give the child to each other in a difficult way, by placing the child on each other’s neck and hold the child in that position for a while. The second task was a horse riding task, where one parent was the horse on his/her hands and knees with the child on his/her back, while the other hold the child in
position. After a while parents switched roles. The third task was again a swing task, with the child in a cloth.
For the first individual movement task, the father and mother were separately asked to swing the child for a while, by grabbing both hands of the child in one hand and his/her feet in the other hand and lift the child up to move back and forth. For the second task the father and mother were individually asked to test their child’s parachute reaction, by letting the child fall forward and catch him/her in time again. The final movement task was a flying task, where the father and mother were individually asked if they ever played a game with their child where it could experience a flying sensation, and if they would demonstrate that for the test leader.
Coding of the overprotection tasks at both measurements
The nine tasks of T1 and the ten tasks of T2 were videotaped and coded by a group of trained students. The training included an explanation of the coding protocol, coding of example cases, and discussing the results of these trials. During the coding process there were several meetings where the results of one coded participant were discussed to avoid observer drift. Observer drift is the tendency of observers to change the way in which they use definitions of behavior over time (Kazdin, 1977). All tasks were divided in intervals and coded on a 5-point Lickert scale of overprotection, with 1 indicating no overprotective behavior and 5 indicating a high frequency or intensity of overprotective behavior. The way of coding low and high frequency or intensity of observed beahvior (in the present study overprotective behavior) was to some degree based on the Meso-analytic Behavioral Rating System for Family interactions or MeBRF (Mahoney, Coffield, Lewis, & Lashley, 2001). The play and dance tasks were coded in one minute intervals and the movement tasks were divided in intervals based on the different operations in each task.
At both measurements a random sample of the parents was coded by multiple observers, in order to examine the inter-coder reliability. The inter-coder reliability was computed with the intra-class correlations (ICC) for each measurement, which are demonstrated in Table 1. As can be seen, the inter-coder reliability of the overprotection scale was good at T1 and satisfactory at T2.
Table 1.
Inter-coder Reliability of the Overprotection Scales Overprotection M SD Min Max
4 months .85 .03 .81 .88 12 months .73 .16 .58 .89
After coding, the scores of the nine tasks at T1 were aggregated into one scale of overprotection for the first measurement occasion and the scores of the ten tasks at T2 were aggregated in one scale of overprotection for the second measurement occasion. The internal consistency (Cronbach’s alpha) of those scales is presented in Table 2 for fathers and mothers separately. The internal consistency was satisfactory for fathers and mothers at both measurement occasions.
Table 2.
Internal Consistency of the Overprotection Scales Overprotection Mother Father
4 months .68 .62
Results
Overview of analyses
First, the stability of overprotective parenting behavior from 4 (T1) to 12 months (T2) age was examined with Pearson product moment correlations. In order to test if mothers and fathers differed in their stability on overprotection the correlation coefficients were compared using Fisher’s Z-transformations. After that, paired t-tests were performed to examine the difference between fathers and mothers on overprotection at T1 and T2. Next, chi-square tests were performed (separately for fathers and mothers), in order to examine if there was a significant difference between anxious and non-anxious parents in continuously high, low, or unstable levels of parental overprotection over time. Finally, one-way ANOVAs were performed (separately for fathers and mothers) to determine whether children demonstrate differences in anxiety, when having a parent with continuously high, low, or unstable overprotection scores over time.
Inspection of the data revealed that values at both measurements were missing completely at random (for T1: χ² (148) = 151.24, p = .411 and for T2: χ² (190) = 155.00, p = .970). Some tasks had more than 10% missing values; at T1 the task where parents had to swing the child in a cloth (10.7%) and at T2 the task where parents had to swing their child back and forth by holding his/her hands and feet (19.0%), the task where parents had to let the child fall forward and catch him/her in time again (14.2%), and the task where parents had to play horse with the child on their back (16.6%). Since these tasks were part of the overprotection scales, which were aggregated over nine tasks at T1 and ten tasks at T2, cases were retained for analysis. Univariate outliers (z < -3.29 or z > 3.29) of the overprotection dimension were replaced by the value of the first non-outlier (Tabachnick & Fidell, 2007).
The relative stability of parental overprotection from 4 months to 12 months
The correlations from 4 months to 12 months of parental overprotection were calculated for fathers and mothers separately. For both fathers and mothers a positive relation was found between parental overprotection on T1 and T2. The correlation of maternal overprotection was r (123) = .29, p < .01 and of paternal overprotection r (120) = .19, p < .05. This indicates that parents’ rank order on overprotection shows modest stability over time. As indicated by the non-significant Fisher’s Z-test, fathers and mothers do not differ in their stability on overprotection: z = .82, p > .05.
Differences in overprotection between fathers and mothers over time
Table 3 demonstrates the results of the paired t-tests that were performed to examine the difference between fathers and mothers on overprotection at T1 and T2. It can be seen that mothers score significantly higher than fathers on overprotection at both measurement occasions.
Table 3.
Differences in parental overprotection between mothers and fathers at both measurement occasions
Overprotection Mother Father
M SD M SD t df p
4 months 1.87 0.35 1.78 0.31 2.73 124 .007
12 months 1.70 0.36 1.56 0.31 4.23 122 .000
Differences in stability of overprotection between anxious and non-anxious parents
In order to test the differences in stability of overprotection between anxious and non-anxious parents, parents were assigned to a group of high and low overprotective parents at each measurement occasion by means of a median split. Subsequently, four stability groups were
formed: one of parents who scored low at both measurements (LL), a second group of parents with low scores at T1 and high scores at T2 (LH), a third group of parents with high scores at T1 and low scores at T2 (HL), and a fourth group of parents who scored continuously high at both measurement occasions (HH).
Table 4 and Table 5 display the chi-square tests that were conducted for mothers and fathers respectively. No significant differences between anxious and non-anxious parents in stability of overprotection were found (for mothers: χ² (3) = .92, p = .821 and for fathers: χ² (3) =3.62, p = .305). This indicates that whether parents have an anxiety disorder or not does not predict if they score continuously high, low, or unstable on overprotection.
Table 4.
Differences in stability of overprotection between anxious and non-anxious mothers
Table 5.
Differences in stability of overprotection between anxious and non-anxious fathers Stability of overprotection Non-anxious Anxious
N % N % χ² p
LL 22 30.1 9 19.1 3.62a .305
LH 14 19.2 15 31.9
HL 19 26.0 10 21.3
HH 18 24.7 13 27.7
Stability of overprotection Non-anxious Anxious
N % N % χ² p
LL 13 29.5 19 26.4 .92a .821
LH 10 22.7 16 22.2
HL 11 25.0 15 20.8
Differences in anxiety of children with continuously high, low or, unstable overprotective parents
Two one-way ANOVA’s were performed to determine whether children demonstrate differences in anxiety, when they had a parent scoring continuously high, low or unstable on overprotection. Table 6 demonstrates the analysis for mothers and Table 7 the analysis for fathers. Both tables show that there are no significant differences in anxiety between children of continuously high, low, or unstable overprotective scoring parents, which indicates that child anxiety does not depend on the stability status of overprotection of their parents.
Table 6.
Differences in anxiety of children with continuously high, low or, unstable overprotective mothers LL (N=31) LH (N = 24) HL (N= 24) HH (N=32) M SD M SD M SD M SD F df p Child Anxiety 2.57 .78 2.33 .81 2.52 .75 2.25 .81 1.11 3,110 .350 Table 7.
Differences in anxiety of children with continuously high, low or, unstable overprotective fathers LL (N=29) LH (N = 28) HL (N= 24) HH (N=30) M SD M SD M SD M SD F df p Child Anxiety 2.52 .84 2.43 .70 2.20 .59 2.48 .83 .94 3,110 .425
Discussion
This study investigated the stability of overprotective parenting behavior of anxious and non-anxious parents and its influence on anxiety in infants. The first goal of the study was to investigate if overprotective parenting in early infancy was stable over time. The results show modest stability for parental overprotection. Based on previous research (Kennedy et al., 2004; Rimehaug et al., 2011) it was expected that there would be considerable stability over time. A possible explanation for the modest stability could be that the current study used an observational method instead of questionnaires. As underlined by Holden and Miller (1999) observational methods have a tendency to find less parenting stability. In order to avoid measurement error, the parental overprotection scores in the present study were based on the aggregated scores of different tasks so that they would represent the parent in multiple contexts (Epstein, 1979). In addition, house visits took place to enhance the ecological validity. Still, the variance of behavior across tasks, the different tasks used at T2 compared to T1 that introduced new variance, and the interrater variance, as different raters were used within and across times, may have all contributed to the modest stability over overprotective behavior. The advantage of using observational data is however, that it allows us drawing conclusions about the stability of actual parenting behavior.
Another explanation for the modest stability on overprotection over time could be the young age of the children. Holden and Miller (1999) hypothesized that the behavior of parents and their interactions with the child will be more established when their children are older, than during infancy, where a lot of developmental changes occur. They indeed found the highest stability in parenting behavior for mothers of school-aged children and the lowest stability for mothers of infants. This indicates that it is important to investigate the stability of parenting behavior over more than just two measurement occasions (as suggested by Dallaire
& Weinraub, 2005; Forehand & Jones, 2002). Through multiple assessments it could be investigated if the stability of overprotective parenting behavior increases over time.
Secondly, it was investigated whether fathers and mothers differ in their stability on overprotection and whether the mean level of parental overprotection differs between fathers and mothers at both measurement occasions. As expected, no difference was found between the stability on overprotection of fathers and mothers. However, mothers scored higher on overprotection at both measurement occasions. This is in line with the hypothesis that the caretaking and supporting role of mothers could easier turn into overprotective behavior than the role of fathers, which is more aimed at challenging and exciting behavior towards the child (Majdandžić et al., 2011).
The third aim of the study was to examine difference in stability of overprotection between anxious and non-anxious parents. Although it was expected that anxious parents would score continuously higher on overprotection than non-anxious parents, the stability of parental overprotection did not depend on whether or not parents had an anxiety diagnosis in the current study.An explanation for this result could be that anxious and non-anxious parents do not differ on overprotection when their child is still an infant. It can be theorized that parents are overall more overprotective when their child is this young. In that case one would expect differences between anxious and non-anxious parents to appear when children grow older and become more independent; where non-anxious parents become less overprotective, anxious parents might hold on to their controlling behavior.
Another explanation for the fact that the stability of parental overprotection did not depend on the anxiety diagnosis may be found in the kind of tasks parents had to perform in the present study. As suggested by Van der Bruggen et al. (2008) it is reasonable that anxious parents will only demonstrate overprotective behavior in situations that trigger their own
anxiety. It might be possible that parental anxiety was not triggered during the research centre and home visits of the current study.
Finally, it was investigated if child anxiety depends on the stability of parental overprotection. It was expected that parents who score continuously high on overprotection would have more anxious children than parents who show unstable or continuously low levels of overprotection. In addition, it was expected that parents who show unstable levels of overprotection would have more anxious children than parents with continuously low scores on overprotection. However, the results were all non-significant, indicating that child anxiety at this age does not depend on continuously high overprotective parenting behavior.
It could be that the effect of stable overprotective parenting behavior on child anxiety is larger when children grow older and are exposed to the overprotective behavior of their parent for a longer period of time. As suggested by Van der Bruggen et al. (2008) it is possible that parents and children mutually influence each other over time, with overprotective parents enhancing child anxiety and anxious children eliciting controlling parenting behavior when children grow older. However, in contrast to this mutually reinforcement theory, there is a theory which expects larger effects of parenting behavior when children are younger. Since there are less competing influences of others, like schoolteachers or peers, it is thought that parents have the greatest influence during early childhood. In addition, children at this age go through an enormous learning period, during which they may be more susceptible for the behavior of their parents than when they are older (Connel & Goodman, 2002; Rothbaum & Weisz, 1994). Future research should therefore investigate what the influence of continuously high overprotective parenting behavior is on anxiety in older children.
The present study had some limitations. First, the population was quit homogeneous. Around 90% of the participants had a Dutch background and their mean education level was
fairly high. Future research should investigate if the findings of the current study can be replicated in a group of participants that represents the general population better.
A second limitation is the structured nature of the observations which might have influenced parent’s behavior and interaction with their child, as some tasks might have provoked overprotective behavior of the parent in an unknown environment, while this was not representative behavior of the parent in daily life. However, the advantage of using observational data is that it allows us to draw conclusions about the stability of actual parenting behavior, instead of parental intentions or beliefs about their behavior. In addition, overprotective parenting behavior was measured over multiple tasks (nine and ten tasks at T1 and T2, respectively), of which some were less structured (free play tasks), and a number of tasks were conducted in the own environment of the participants during home visits. Also, Majdandžić, de Vente, and Bögels (manuscript submitted for publication) demonstrated preliminary evidence for validity of the observational instrument measuring overprotection. The observations of parental overprotection show modest significant positive correlations with questionnaires about overprotective parenting behavior at T1, suggesting measurement convergence. Furthermore, negative medium to strong correlations with observed challenging behavior of parents at both measurement occasions provide support for divergent validity (Majdandžić, de Vente, & Bögels, manuscript submitted for publication).
Third, another methodological factor could (partly) be responsible for the non-significant results that were obtained for the third and fourth question. In order to create stability groups, a median split was used. As a consequence, parents who scored around the cutoff score were assigned to different groups, while they in fact might not have differed that much from each other (Field, 2009).
A fourth limitation concerns the instrument that was used to measure child anxiety; the IBQ-R. Research has demonstrated that parents with a high level of negative affect reported
higher scores for their children on the fear subscale than was expected based on laboratory observations (Gartstein & Marmion, 2008). The reports on child anxiety from anxious parents of the present study could therefore have been influenced by their own anxiety.
Finally, the current study did not investigate what influence fathers and mothers have on each other’s behavior towards the child. As Holden and Miller (1999) indicate, dyadic relationships between parent and child do not take place in a vacuum. Based on the second-order effect theory of Bronfenbrenner (1979) and triadic family interactions, they indicate that the behaviors of parents and children can be influenced by others. It is plausible that parents reinforced or decreased each other’s behavior in the present study, especially during tasks where fathers and mothers had to work together. It is important to take this into account in future research, particularly since evidence suggests that more problematic coparenting is related to child anxiety and that parent’s own anxiety might negatively influence supportive coparenting behavior (Majdandžić, de Vente, Feinberg, Aktar, & Bögels, 2012).
Conclusion
The current study has taken a first step in investigating the stability of overprotective parenting behavior by observations of actual parenting behavior. Results indicate that observed overprotection is modestly stable over time, with no difference in stability between fathers and mothers. Contrary to our expectations, the stability of parental overprotection does not significantly depend on parent’s anxiety diagnosis. In addition, child anxiety does not significantly depend on the stability status of parental overprotection.
Future research should investigate the stability of overprotective parenting and child anxiety in older children. Through multiple measurement occasions it is possible to examine whether the stability of overprotective parenting behavior increases with children’s age. In addition, inclusion of questionnaires on overprotective parenting behavior, next to
observations, could provide a more complete representation of the amount of parenting stability. These investigations are necessary for fine-tuning risk assessments and intervention programs. The results could for example indicate when overprotective parenting behavior becomes more stable and habituated and therefore less susceptible for change. In addition, the results indicate that special attention is needed for the role of the mothers when it comes to overprotective parenting behavior, since they scored higher than fathers, on both measurement occasions.
References
Adam, E. K., Gunnar, M. R., & Tanaka, A. (2004). Adult attachment emotion, and observed parenting behavior: Mediator and moderator models. Child Development, 75, 110-122. Ballash, N., Leyfer, O., Buckley, A. F., & Woodruff-Borden, J. (2006). Parental control in the
etiology of anxiety. Clinical Child and Family Psychology Review, 9, 113-133. Barber, B. K. (1996). Parental psychological control: Revisiting a neglected construct. Child
Development, 67, 3296-3319.
Benjamin, R. S., Costello, E. J., & Warren, M. (1990). Anxiety disorders in a pediatric sample. Journal of Anxiety Disorders, 4, 293-316.
Bögels, S. M., & Brechman-Toussaint, M. L. (2006). Family issues in child anxiety: Attachment, family functioning, parental rearing and beliefs. Clinical Psychology Review, 26, 834-856.
Boyd, C. P., Gullone, E., Kostanski, M., Ollendick, T. H., & Shek, D. T. L. (2000). Prevalence of anxiety and depression in Australian adolescents: Comparisons with worldwide data. Journal of Genetic Psychology, 161, 479-492.
Bronfenbrenner, U. (1979). The ecology of human development: Experiments by nature and design. Cambridge, MA: Harvard University Press.
Clark, K. E., & Ladd, G. W. (2000). Connectedness and autonomy support in parent-child relationships: Links to children’s socioemotional orientation and peer relationships. Developmental Psychology, 36, 485-498.
Connell, A. M., & Goodman, S. H. (2002). The association between psychopathology in fathers versus mothers and children’s internalizing and externalizing behavior problems: A meta- analysis. Psychological Bulletin, 128, 746–773.
Dallaire, D. H., & Weinraub, M. (2005). The stability of parenting behaviors over the first 6 years of life. Early Childhood Research Quarterly, 20, 201–219.
Di Nardo, P. A., Brown, T. A., & Barlow, D. H. (1994). Anxiety Disorders Interview Schedule for DSM-IV: Lifetime version (ADIS-1V-L). San Antonio, TX: Psychological
Corporation.
Edwards, S. L., Rapee, R. M., & Kennedy, S. (2010). Prediction of anxiety symptoms in preschool-aged children: examination of maternal and paternal perspectives. Journal of Child Psychology and Psychiatry 51, 313–321.
Epstein, S. The stability of behavior: І. On predicting most of the people much of the time. Journal of Personality and Social Psychology, 37, 1097-1126.
Field, A. (2009). Discovering statistics using SPSS. London: SAGE Publications Ltd. Forehand, R., & Jones, D. J. (2002). The stability of parenting: A longitudinal analysis of
inner-city African-American mothers. Journal of Child and Family Studies, 11, 455-467.
Gartstein, M. A., & Marmion, J. (2008). Fear and positive affectivity in infancy:
Convergence/discrepancy between parent-report and laboratory-based indicators. Infant Behavior & Development, 31, 227–238.
Gartstein, M. A., & Rothbart, M. K. (2003). Studying infant temperament via the Revised Infant Behavior Questionnaire. Infant Behavior & Development, 26, 64-86.
Ginsburg, G. S., Grover, R. L., & Ialongo, N. (2004). Parenting behavors among anxious and non-anxious mothers: Relation with concurent and long-term child outcomes. Child & Family Behavior Therapy, 26, 23-41.
Ginsburg, G. S., & Schlossberg, M. C. (2002). Family-based treatment of childhood anxiety disorders. International Review of Psychiatry, 14, 143-154.
Holden, G. W., & Miller, P. C. (1999). Enduring and different: A meta-analysis of the similarity in parents’ child rearing. Psychological Bulletin, 125, 223-254.
Hudson, J. L., & Rapee, R. M. (2001). Parent-child interactions and anxiety disorders: An observational study. Behavior Research and Therapy, 39, 1411-1427.
Kashani, J. H., & Orvaschel, H. (1990). A community study of anxiety in children and adolescents. American Journal of Psychiatry, 147, 313-318.
Kazdin, A. E. (1977). Artifacts, bias, and complexity if assessment: The ABCs of reliability. Journal of Applied Behavior Analysis, 10, 141-150.
Kennedy, A. E., Rubin, K. H., Hastings, P. D., Maisel, B. (2004). Longitudinal relations between child vagal tone and parenting behavior: 2 to 4 years. Developmental Psychobiology 45, 10-21.
Last, C. G., Perrin, S., Hersen, M., & Kazdin, A. E. (1996). A prospective study of childhood anxiety disorders. Journal of the American Academy of Child & Adolescent
Psychiatry, 35, 1502-1510.
Maccoby, E. E. (1992). The role of parents in the socialization of children: An historical overview. Developmental Psychology, 28, 1006-1017.
Maccoby, E. E. (2000). Parenting and its effects on children: On reading and misreading behavior genetics. Annual Review of Psychology, 51, 1-27.
Mahoney, A., Coffield, A., Lewis, T., & Lashley, S. L. (2001). Meso-analytic behavioral rating system for family interactions: Observing play and forced-compliance tasks with young children. In P. K. Kerig, & K. M. Lindahl (Eds.), Family observational coding systems: Resources for systemic research. Mahwah, NJ: Lawrence Erlbaum Associates.
Majdandžić, M., Möller, E., Bögels, S., & Van den Boom, D. (2011). Verschillen tussen vaders en moeders in de relatie tussen opvoedingsgedrag en sociale angst van hun kinderen. Pedagogiek, 31, 11-28.
Majdandžić, M., de Vente, W., Feinberg, M. E., Aktar, E., & Bögels, S. M. (2012).
Bidirectional associations between coparenting relations and family member anxiety: A review and conceptual model. Clinical Child and Family Psychology Review, 15, 28-42.
Majdandžić, M., de Vente, W., & Bögels, S. M. (2012). Challenging parenting behavior in infancy: Etiology, measurement and differences between fathers and mothers. Manuscript submitted for publication.
McLeod, B. D., Wood, J. J., & Weisz, J. R. (2007). Examining the association between parenting and childhood anxiety: A meta-analysis. Clinical Psychological Review, 27, 155-172.
Paquette, D. (2004). Theorizing the father-child relationship: Mechanisms and developmental outcomes. Human Development, 47, 193-219.
Pettit, G. S., Laird, R. D., Dodge, K. A., Bates, J. E., and Criss, M. M. (2001). Antecedents and behavior-problem outcomes of parental monitoring and psychological control in early adolescence. Child Development, 72, 583-598.
O’Leary, S. G., Smith Slep, A. M., & Reid, M. J. (1999). A longitudinal study of mothers’ overreactive discipline and toddlers’ externalizing behavior. Journal of Abnormal Child Psychology, 27, 331–341.
Rapee, R. M. (1997). Potential role of childrearing practices in the development of anxiety and depression. Clinical Psychology Review, 17, 47-67.
Rapee, R. M., Schniering, C. A., & Hudson, J. L. (2009). Anxiety disorders during childhood and adolescence: Origins and treatment. Annual Review of Clinical Psychology, 5, 311-341.
Rimehaug T., Wallander J., Berg-Nielsen T. S. (2011). Group and individual stability of three parenting dimensions. Child and Adolescent Psychiatry and Mental Health, 5, 1-12.
Rothbart, M. K. (2011). Becoming who we are: Temperament and personality in development. New York, NY: The Guilford Press.
Rothbaum, F., & Weisz, J. R. (1994). Parental caregiving and child externalizing behavior in nonclinical samples: A meta-analysis. Psychological Bulletin, 116, 55-74.
Sroufe, L. A. (2000). Early relationships and the development of children. Infant Mental Health Journal, 21, 67–74.
Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics. Boston, MA: Pearson Education.
Van der Bruggen, C. O., Stams, G. J. J. M., & Bögels, S. M. (2008). Research review: The relation between child and parent anxiety and parental control: a meta-analytic review. Journal of Child Psychology and Psychiatry, 49, 1257–1269.
Verhoeven, M., Junger, M., Van Aken, C., Deković, M., Van Aken, M. (2007). A short-term longitudinal study of the development of self-reported parenting during toddlerhood. Parenting: Science and Practice, 7, 367-394.
Whaley, S. E., Pinto, A., & Sigman, M. (1999). Characterizing interactions between anxious mothers and their children. Journal of Consulting and Clinical Psychology, 67, 826-836.
Wood, J. J. (2006). Parental intrusiveness and children's separation anxiety in a clinical sample. Child Psychiatry & Human Development, 37, 73−87.
