Children’s involvement in home-based childcare: are boys more susceptible to caregiver sensitivity than girls?

ABSTRACT

Children who are in high-quality childcare are encouraged to explore the material and social environment. Boys and girls might differ in their susceptibility to this quality of childcare. In the current study, we tested whether childcare quality is related to children’s involvement in home-based childcare, and whether child gender acts as a moderator in this association. Childcare quality (caregiver sensitivity and global quality) and children’s involvement were observed at 103 home-based childcare settings, including 103 female caregivers and 162 children (86 boys). Results showed that boys showed lower involvement (than girls) when they were in the care of a less sensitive caregiver and higher involvement (than girls) when they were in the care of a more sensitive caregiver. In addition, boys and girls in smaller groups showed higher involvement than children in larger groups. Global childcare quality was not related to children’s involvement. This study shows that it is important, especially for boys, to be taken care of by a sensitive caregiver, who supports and encourages children to be involved in their play, and to explore the world around them.

KEYWORDS:

• Gender
• childcare
• involvement
• Quality of care

Introduction

Young children are curious about the world around them. In the professional childcare setting, this world revolves around caregivers, space, and peers. Children who are in a high-quality childcare setting, characterized by, for example, sensitive caregivers, age-appropriate toys, a safe and healthy environment, and an adequate caregiver-child ratio, are encouraged to explore the material and social environment. These children might show higher levels of involvement in all types of activities, and as a consequence develop better educational skills. Several studies have suggested that boys and girls differ in their susceptibility to childcare quality: boys might be more susceptible to their caregiving environment than girls (e.g. Gunnar et al. 2010). In the current study, we focus on home-based childcare in the Netherlands, provided from a caregiver’s personal home, which makes the daily environment more similar to a child’s home than center-based childcare. In Dutch childcare homes, each professional (i.e. formal, not-related) caregiver takes care of a small group of children in her own house. This implies that children in home-based childcare are taken care of by one caregiver in the same environment the entire day. We tested whether childcare quality is related to children’s involvement, and whether child gender acts as a moderator in this association.

Childcare quality and children’s involvement

Childcare quality is a multidimensional concept and can be defined in terms of structural quality and process quality (Slot, 2018). Indicators of structural quality entail distal and regulable factors and include caregiver education, caregiver-child ratios, and group size. This structural quality is partly determined by legislation, policy, and funding (Slot, 2018). In the Netherlands, for example, caregivers in home-based childcare are required to have an adequate childcare diploma and are restricted to a maximum of six children per day (including their own children, with age restrictions for children, e.g. maximum of two infants at the same time).

Process quality refers to more proximal processes of children’s actual experiences during childcare and involves the social, emotional, physical, and instructional aspects of caregiver–child interactions and peer interactions (Slot, 2018). An important aspect of process quality is caregiver sensitivity, that is, the ability of caregivers to accurately perceive the children’s signals and to respond promptly and adequately to these signals (Ainsworth et al. 1978). Sensitive caregiving facilitates children to build a secure relationship with their caregiver. According to attachment theory, children use their caregivers as a haven of safety, from which they can explore the environment (Bowlby 1969). Although attachment theory originated from family research, a small but growing body of research shows that caregiver sensitivity in professional childcare is a fundamental aspect of process quality and predicts children’s social competence (Ahnert, Pinquart, and Lamb 2006, Arnett, 1989; Clarke-Stewart et al. 2002; de Schipper, Tavecchio, and van IJzendoorn 2008; Eckhardt and Egert 2020; Elicker 2005; Groeneveld et al. 2010). Sensitive caregivers might not only stimulate social development of children, but they might also enable children’s cognitive development by anticipating young children’s natural curiosity, and encouraging children to be actively involved in all types of activities during childcare. The concept of involvement refers to a quality of human activity in which there is intense mental activity, characterized by deep concentration and persistence (Laevers 1993). Children scoring high on involvement are concentrating, absorbed and fascinated by their activities, and operate at the limits of their capacities (Laevers and Declercq 2018). Csikszentmihalyi (1979) used the concept of ‘flow’ to refer to a state of mind that results from being involved in an activity that is chosen for its own sake and that promotes personal growth by challenging existing abilities. Involvement can be observed in a variety of situations and is not related to a specific age group. Infants playing with their toes may experience a state of involvement just as well as mathematicians working out complex equations. Children’s involvement within childcare, which is characterized by intrinsically motivated play, fascination, and exploration, has an effect on children’s learning (Laevers 1997; Pascal, Bertram, and Ramsden 1997).

Authors often use the terms ‘involvement’ and ‘engagement’ interchangeably. However, empirical studies on these constructs differ in focus. Research on engagement tends to focus more on cognitive skills and strategies, whereas involvement is more related to intrinsic motivation and interest (Turner et al. 1998). Ridley, McWilliam, and Oates (2000) define engagement as the amount of time children spend interacting with adults, other children, or materials in an appropriate manner. Types of engagement range from sophisticated behaviours like problem solving and pretend play to non-engaged behaviours like wandering aimlessly, fighting or crying. According to Turner et al. (1998), engagement is often a prerequisite to involvement but is not sufficient to cause it.

Laevers (1993, 1994, 1997) further elaborated on the concept of involvement for young children. According to Laevers (1993) involvement is determined by the drive to explore and the child’s individual developmental needs. One of the most predominant characteristics is intrinsic motivation. An involved child is fascinated by and totally absorbed in the activity. Involvement only occurs when the activity matches the capabilities of the child, so when it is not too easy or too demanding. Young children experience a state of involvement mostly in play situations (Laevers 1994).

High levels of involvement facilitate learning. Children who are highly involved are concentrating, creative, energetic, and persistent during their activities, as opposed to children who are less involved and show more simple, repetitive, and passive play. Uninvolved children may stare into space or wander around the classroom (Laevers 1994). Children who are more involved during childcare show higher levels of achievement. Although the longitudinal effects of involvement have not been studied in childcare, Ladd et al. (1999) found that children who showed more active participation in kindergarten (more cooperative and independent participation) scored higher on a school readiness test 12 weeks later. Children’s involvement in childcare might be a precursor for later educational success since the key elements of involvement (intrinsic motivation, fascination, and exploration) are all important aspects of educational success (Augustyniak et al. 2016; Krapp 1999; Prevatt et al. 2011)

Professional caregivers who provide high-quality care can promote children’s involvement in different ways, first by offering stimulating activities and an environment that facilitates these activities (high global quality of childcare) and second, by using sensitive caregiving strategies. A previous study from Belgium showed significant associations between emotional and educational support offered by caregivers in childcare centers and children’s involvement (Laevers and Declercq 2018). In their Ten Action points, an inventory of ten types of initiatives that favor wellbeing and involvement, Laevers and Moons (1997) suggest a wide range of interventions, including the organization of the space and provision of interesting materials and activities (global quality), but also stimulating caregivers to observe carefully how children interact with their environment in order to identify interests and support ongoing activities, which requires a caregiving style including stimulation, sensitivity and giving autonomy (Laevers 2000; Laevers, Bogaerts, and Moons 1997). In the current study, we focus on the moderating role of gender in the association between childcare quality (global quality and caregiver sensitivity) and children’s involvement.

Differential susceptibility to the childcare environment

Boys and girls may vary in their susceptibility to caregiving. According to the differential susceptibility theory (J. Belsky 1997; J. Belsky, Bakermans-Kranenburg, and van IJzendoorn 2007), effects of caregiver behavior may be moderated by child characteristics, for better and for worse.

Consistent with this differential susceptibility theory, Belsky and Pluess (2009) found that children with a difficult temperament as infants showed less social competence when they experienced low childcare quality at 54 months, but that children with a difficult temperament experiencing high childcare quality showed fewer behavior problems, compared with children with an easy temperament. A central role is the idea of ‘susceptibility’ instead of ‘vulnerability’. Children with a difficult temperament may not only suffer more from low quality but also benefit more from high-quality care, compared with their peers with an easy temperament. When focusing on gender, maternal unresponsiveness to one year-old children has been found to be related to child disruptive behavior one year later, but only in boys (Shaw et al. 1998). In a sample of three-year-olds, more parental warmth was related to less disruptive behavior problems, but again only for boys (Kerr et al. 2004). This suggests that child gender may act as a differential susceptibility factor, with boys being more susceptible to the care they receive, thus showing less favorable outcomes (than girls) in lower-quality care and more favorable outcomes in high-quality care (Burchinal et al. 2000).

In childcare, several studies have found no evidence for a moderation effect of gender on the association between childcare quality and children’s socio-emotional development (Burchinal et al. 2000; Keys et al. 2013). Other studies have found that high-quality childcare was related to fewer behavior problems in boys (Broekhuizen et al., 2015; Howes and Olenick 1986; Votruba-Drzal et al. 2010; Votruba-Drzal, Levine Coley, and Chase-Lansdale 2004), while these effects were less strong or absent for girls. In addition, Gunnar et al. (2010) have shown that boys showed more angry, aggressive behavior than girls with a more intrusive and controlling caregiver. These studies suggest that child gender may act as a differential susceptibility factor, with boys showing more behavior problems in childcare of low quality (compared to girls), and fewer behavior problems in high-quality childcare. In the current study, the focus is not on behavior problems, but child involvement. To our knowledge, no studies thus far have focused on child gender as a moderator of the association between childcare quality and child involvement in home-based childcare.

Aim of the study

In the current study, we tested whether childcare quality is related to children’s involvement in home-based childcare, and whether child gender acts as a moderator in this association, resulting in two research questions: (1) Is childcare quality (e.g. global childcare quality and caregiver sensitivity) related to children’s involvement during childcare? and (2) Is the association between childcare quality and child involvement moderated by child gender? Based on the childcare and parenting literature, we expected a significant association between childcare quality (global childcare quality and sensitivity) and child involvement, with boys being more susceptible to childcare quality than girls: with higher levels of quality of care we expect boys to show more involvement than girls, and with lower levels of quality of care, we expect boys to show lower involvement than girls.

Method

Sampling

The current sample draws on two closely related studies in home-based (non-relative) childcare that were performed within a two-year time-frame by the same research group. The first study was aimed at examining children’s cortisol levels and wellbeing (Groeneveld et al. 2010). The second study was aimed at assessing the effect of a short-term intervention program (the Video-feedback Intervention to promote Positive Parenting – Childcare; Groeneveld et al. 2011). All of the data analyzed in the current paper were collected during the first study and the pre-intervention phase of the second study. The combined sample consists of 103 caregivers: 55 from the first study and 48 from the second. Across these 103 caregivers, data were collected on 162 children for whom parental permission was obtained. In both studies caregivers were contacted through childcare agencies by invitation letters. If caregivers were willing to participate, they approached the parents of the children in their care for permission to proceed with the study. For more information see Groeneveld et al (2010) and Groeneveld et al (2011). In both studies, the same measures and similar procedures were used on samples from the same population, which justifies combining the samples theoretically. Independent t-tests showed no differences (p’s > 0.05) between the two studies on caregiver characteristics (age, experience, education, sensitivity), child characteristics (age, gender, involvement, parent’s education) or outcomes (sensitivity, global childcare quality, and involvement).

Sample characteristics

Demographics of children and caregivers are shown in Table 1. Caregivers were all female, between 28 and 65 years old (M = 43.16, SD = 9.17), and had on average 6.07 years of experience in childcare (SD = 4.92, range 5–25 years). They had on average 12.32 years of education after primary school entry at age 6 (SD = 2.00, range 10–16) and 65% of them completed a vocational education in the field of childcare. The number of work hours in childcare per week ranged from 7.5 to 55 (M = 33.49, SD = 11.60). The number of children per group varied from 1 to 9 (M = 3.11, SD = 1.44). The sample of target children consisted of 86 boys and 76 girls. The mean age of these children was 27.28 months (SD = 9.55, range 2–42 months), and children visited the childcare setting between 7 and 44 hours per week (M = 20.18, SD = 7.61). Mothers of the children completed on average 13.71 (SD = 2.12) and fathers 13.31 (SD = 2.42) years of education from age 6.

Table 1. Descriptives of caregiver and child background variables and Pearson correlations with caregiver sensitivity, global quality, and child involvement.

	Descriptives		Pearson Correlations
	M	(SD)	Caregiver Sensitivity	Global quality	Child Involvement
Caregivers (n = 103)
Age^a	43.16	(9.17)	.12	−.01	.20*
Education^a	12.32	(2.00)	.15	.19*	.02
Experience in childcare^a	6.07	(4.92)	.08	.01	−.03
Working hours^b	33.49	(11.60)	−.14	.20*	.20*
Caregiver sensitivity	4.80	(0.85)		.28**	.16*
Global quality	0.86	(0.07)	.28**		.05
Children (n = 163)
Age^c	27.28	(9.55)	.03	.06	−.30*
Time spent at childcare^c	20.18	(7.61)	−.09	−.01	−.13
Child involvement	2.79	(0.54)	.16*	.05
Group
Number of children	3.11	(1.44)	−.02	.12	−.28**
Percentage of boys	.53	(.33)	−.13	−.05	.01

^ain years ^bin hours per week ^cin months.

*p < .05.

Procedure

Each setting was visited by an observer who spent a morning in the childcare homes. This observer administered the Infant Toddler Child Care Home Observation for Measurement of the Environment inventory (IT-CC-HOME; Bradley, Caldwell, and Corwyn 2003) to measure global childcare quality from the perspective of the target child. Furthermore, at pre-set time points, the observer video-taped each target child for three 8-minute sessions and the caregiver for three 10-minutes episodes. Video material was rated afterwards on child involvement and caregiver sensitivity by observers who were trained in reliability. To obtain independent ratings, researchers who collected material from one particular childcare setting never coded video material from that setting, and child involvement and caregiver sensitivity were coded by different researchers.

In both studies, all procedures were carried out with adequate understanding and written consent of caregivers and parents. Ethical approval was provided by the Leiden Institute of Education and Child Studies.

Measures

Caregiver sensitivity

Coding of group-focused caregiver sensitivity was based on three 10-minutes episodes of video-material of the caregiver during regular daily activities at childcare across one morning. We used the Sensitive Responsiveness scale, one of the Caregiver Interaction Profile (CIP) scales, developed and validated by the Dutch Consortium for Childcare Research (NCKO; Helmerhorst et al. 2014). This scale is based on rating scales for sensitivity with parent–child dyads (Ainsworth, Bell, and Stayton 1974; Erickson, Sroufe, and Egeland 1985), but specifically focuses on sensitivity towards a group of children. Sensitivity ratings range from (1) very low sensitivity to (7) very high sensitivity. A high score indicates that the caregiver is able to give all children in the setting adequate emotional support when they need it, during stressful as well as non-stressful situations. A caregiver scoring high on sensitivity is a safe haven for the children. The caregivers recognize the signals of the child promptly and adequately, show interest in children’s activities and show awareness of needs, moods, and possibilities of children. A low score on the Sensitivity scale indicates that the caregiver is not able to provide children with emotional support when they need it. The caregiver does not recognize the signals of the children, reacts very slowly, does not react, or shows an inadequate response: The response does not fit with the situation, emotions, or developmental level of the child. In addition, caregivers scoring low on sensitivity might show indifferent or detached behavior. Coders scored every 10-minute episode. Final scores were averaged across the three observation sessions. In the current study, Cronbach’s alpha for the scale was .72. Mean intra-class correlation (two-way mixed, absolute agreement with trainer) of the eight coders was .75 (ranging from .69 to .80).

Global quality of childcare

The IT-CC-HOME (Bradley, Caldwell, and Corwyn 2003) is designed to measure the quality and quantity of stimulation and support available to a child in the childcare home environment, and covers various domains of childcare: responsivity, acceptation, organization, learning materials, involvement, and variation. A positive (1) or a negative (0) score is achieved for each of the 43 items. Two items were deleted from the scale: item 21 ‘Child gets out of house at least four times a week’ and item 42 ‘Caregiver and child visit or receive visits from neighbor or friends once a month or so’. These items were not applicable to the Dutch situation because in the Netherlands children attend home-based childcare on average two or three days a week, in contrast to other countries where children are more likely to attend full-time. Internal consistency (Cronbach’s alpha) of this scale was .60. Observers were trained prior to the study. After a general introduction, observers visited at least four caregivers in pairs to complete the IT-CC-HOME. Each observation was followed by an item-by-item debriefing with the trainer. Prior to this debriefing, interrater reliability was established to a criterion of 80% agreement.

Child involvement

Coding of child involvement was based on three 8-minute episodes of video-material recording regular daily activities at childcare across one morning. Each episode was divided into 2-minute fragments. Each fragment was assigned a score according to the Involvement scale (Laevers 1994), which was validated by the Dutch Consortium for Childcare Research (NCKO; de Kruif et al., 2007). Within this scale, involvement is operationalized by concentration and persistence; children scoring high on involvement are motivated, open to stimuli within this activity, satisfied, and energetic. Scores on the Involvement scale may range from (1) to (5), where a (5) indicates that signs of involvement, such as concentration, persistence, are clearly present, a (3) indicates that the child shows mainly continuous activity, the child is busy with a routine activity but does not show energy or concentration, and a (1) indicates that the child is very passive, or doing an activity that is simple and repetitive. Final involvement scores were averaged across observation sessions to obtain a stable and representative score for involvement across the morning. Internal consistency (Cronbach’s alpha) across four 2-minute fragments was .83. Mean intra-class correlation (two-way mixed, absolute agreement with trainer) was .76 (ranging from .70 to .81).

Data analyses

To test whether childcare quality was associated with children’s involvement correlation analyses were performed. Next, multiple regression analyses were performed to test whether child gender acted as a moderator in the association between childcare quality and involvement. As some of the observed children came from the same child care setting, we had to deal with a hierarchical data structure. However, the number of children per childcare setting was small (never more than 3). In addition, because in 50 out of 103 settings we observed only one child, the within- and between-setting variance partly overlapped. Therefore, we performed simple multiple regression analyses as a starting point, and corroborated the results using a multilevel (random intercept) model. As expected, multilevel outcomes were similar to the outcomes of the multiple regression analyses and led to the same conclusions on the interaction terms. Therefore, we only report the simple regression results.

Results

Childcare quality

Mean global childcare quality was 0.86 (SD = 0.07) and ranged between 0.61 (61% of the items were scored positively) and 1.00 (all items were scored positively). Global childcare quality was not related to group size or the percentage of boys present. Global childcare quality was positively related to caregiver education, working hours, and sensitivity. Caregivers with a higher education, caregivers who worked more hours per week in childcare, and caregivers who were more sensitive tended to score higher on global childcare quality (Table 1). Caregiver sensitivity ranged between 2.33 and 6.67 with a mean of 4.80 (SD = 0.85) and was not related to group size, the percentage of boys present or caregiver characteristics.

Child involvement

Child involvement ranged between 1.42 and 4.67 with a mean score of 2.79 (SD = 0.54). Involvement was positively related to caregiver age and working hours. Children who were cared for by older caregivers or by caregivers who worked more hours per week showed more involvement during childcare. Considering child characteristics, children’s involvement was negatively related to their age: younger children showed higher levels of involvement during childcare. There was no gender difference: boys and girls showed the same level of involvement during childcare: t (160) = 0.93, p = .36, d = 0.15.

Associations between quality of care and involvement

Child involvement was negatively related to group size: children showed more involvement in smaller groups of children. While children’s involvement was not related to global childcare quality, a significant association was present between children’s involvement and caregiver sensitivity: children with a more sensitive caregiver showed higher involvement.

Gender as a moderator

A hierarchical linear regression analyses was performed to test whether child gender moderated the association between childcare quality and child involvement. The results are shown in Table 2. In the first step (R² = 0.04), child gender (β = −0.11, p = .16) and global childcare quality (β = 0.01, p = .96) did not predict child involvement, while caregiver sensitivity did (β = 0.18, p < .05). In the second step, the interaction terms are added. The interaction between global childcare quality and child gender was not significant and was deleted from the model (β = 0.06, p = .47, ∆R² = .01, p = .47). The interaction between child gender and sensitivity was significant: the association between caregiver sensitivity and child involvement was significantly moderated by child gender (β = −0.17, p < .05, ∆R² = .03, p < .05). This interaction effect is shown in Figure 1. For boys, sensitive care is related to involvement (β = 0.13, p < .01), whereas this association is not present for girls (β = −0.03, p = .78).

Figure 1. The association between caregiver sensitivity and child involvement for boys and girls.

Table 2. Regression analyses: predicting children’s involvement.

	Step 1 (R^2 = 0.04)				Step 2 (R^2 = 0.07)
	B	SEB	β	p	B	SEB	β	p
Constant	2.79	0.04		.00	2.28	0.04		.00
Child Gender	−0.12	0.09	−0.11	.16	−0.12	0.08	−0.11	.16
Caregiver Sensitivity	0.12	0.05	0.18	.03	0.11	0.05	0.17	.04
Global Quality	0.04	0.62	0.01	.96	0.04	0.62	0.01	0.95
Gender * Sensitivity					−0.23	0.10	−0.17	.03

Discussion

In the current study, we found an association between caregiver sensitivity and children’s involvement, but only for boys: boys were more susceptible to caregiver sensitivity than girls. Boys’ involvement was higher when receiving more sensitive care and lower when receiving less sensitive care. For girls, this association between caregiver sensitivity and involvement was not present. In addition, we found no association between global childcare quality and boys’ or girls’ involvement.

Sensitive care is a central indicator of process quality, which refers to interactions between caregivers and children. Sensitive caregivers have the ability to accurately perceive the children’s signals and to respond promptly and adequately to these signals (Ainsworth et al. 1978), but also act as a haven of safety, from which children can explore their environment (Bowlby 1969). In this study, we showed that especially boys are susceptible to (in)sensitive caregiving in childcare, confirming previous studies that showed that childcare quality was more strongly associated with social and behavioral outcomes in boys than in girls (Broekhuizen et al. 2015; Howes and Olenick 1986; Votruba-Drzal et al. 2010).

On the positive side, boys may profit more than girls from encouragement to be actively involved in all types of activities during childcare. On the downside, boys’ involvement may be disproportionally negatively affected by insensitive caregiving, such as intrusiveness or negative control.

Girls’ involvement was not associated with caregiver sensitivity. The question arises whether caregivers differentiate in sensitive caregiving between boys and girls. Several studies in families have shown gender-differentiated parenting: mothers and fathers tend to be more sensitive towards girls than boys (Barnett et al. 2008; Bornstein et al. 2008; Lovas 2005), but there are also studies that did not find these differences (Hallers-Haalboom et al. 2014). A meta-analysis including 120 studies has shown that in the last two decades, mothers and fathers use slightly more positive control strategies (e.g. induction, support, praise) with girls and more negative control strategies (e.g. negative comments, physical punishment, power assertion) with boys, but these differences are very small (Endendijk et al. 2016). Only a few studies have investigated gender-differentiated caregiving in professional childcare. In childcare centers as well as in childcare homes, boys seem to be at a disadvantage compared to girls. Some studies showed that boys received lower global childcare quality (ECCRN, NICHD 1997; Winer and Philips 2012), less positive caregiving behaviour (e.g. sensitivity, positive regard for the child, warmth, stimulation; ECCRN, NICHD 1997; Gunnar et al. 2010) and more negative caregiving behaviour (e.g. negative comments; Winer and Philips 2012). However, other studies found no significant differences between boys and girls in positive caregiving (Winer and Philips 2012) and negative caregiving (Gunnar et al. 2010) or sensitivity (Ereky-Stevens et al. 2018). In the current study, the percentage of boys and girls in the group was not related to the sensitivity of the caregiver or to global childcare quality. It should be noted that in the current study, the instrument to measure caregiver sensitivity is a group measure. This indicates that a caregiver receives a score for sensitivity for the core of a group of children (consisting of boys and girls). For this reason, we cannot draw conclusions regarding (dyad) sensitive caregiving towards children in the group and possible associations for girls (and boys) between dyad sensitive caregiving and their involvement.

Surprisingly, global quality was not related to children’s involvement in the current study. Based on previous studies, it was hypothesized that global childcare quality would also be important for children’s involvement (Laevers 1993, 1997). We assumed that a safe environment with adequate toys, materials, and activities would result in higher involvement. Not finding such an association might be explained by the relative high levels of global childcare quality in the current study. It might be expected that children need a minimal global childcare quality, to have the opportunity to be involved in play, a ‘good enough environment’. In addition, this study suggests that caregiver sensitivity seems more important than global childcare quality when it comes to involvement. The caregiver plays a crucial role in encouraging children to play with the (adequate) materials, to explore the (safe) environment and to stimulate children in (age appropriate) activities.

In addition to sensitivity, the group size was also related to the involvement of the children during home-based childcare. Children were more involved in smaller groups. This means that children in smaller groups were more concentrated, creative, energetic, and persistent during their activities as opposed to children who were in larger groups, who showed more simple, repetitive, and passive play. This is consistent with previous findings (Powell et al. 2008). The group size of the current study varied from one child to nine children present during childcare, although the statutory maximum is six in the Netherlands. After excluding the children in the groups with more than six children present, the association between group size and involvement was still present.

Previous studies have found that group size or caregiver–child ratio (an aspect of structural quality) is related to childcare quality (Phillips et al. 2000; Palmérus 1996; Schipper, Riksen-Walraven, and Geurts 2006; Slot, 2018). In our study, group size was not related to caregiver sensitivity or global childcare quality. This might be due to relatively small group sizes, or to other structural childcare components that might be more important than group size. A study from South Africa, for example, showed that the management quality of the childcare setting was a predictor of process quality, above and beyond staff characteristics, including caregiver education and child-staff ratios (Biersteker et al. 2016).

Group size was associated with child involvement. This finding suggests that it is not the caregiver who is affected by the large number of children, but the child itself, resulting in lower involvement. This might be explained by the higher noise levels that are produced by the children in the room. Previous studies in childcare have shown that noise levels are related to the wellbeing of children. Linting et al. (2013) found that mean noise levels and noise variability were negatively related to children’s wellbeing in home-based childcare when scores rose above a critical threshold. In addition, other studies (in the school setting) have shown that noise has a negative effect on cognition, for example, on performance on verbal tasks and understanding (Evans 2006; Klatte, Lachmann, and Meis 2010) and reading skills (Maxwell and Evans 2000). In addition, the groups are relatively small in Dutch home-based childcare. There might be an association present between group size and childcare quality in childcare centers, with larger groups of children and more variation in group size.

The current study also has some limitations. First, the level of caregiver sensitivity is relatively high compared to the sensitivity of caregivers in childcare centers (Groeneveld et al. 2010), which may have caused a restricted range. Another limitation is that we used a measure with relatively low internal consistency: Generally, the internal consistency reported for the IT-CC-HOME is comparable to what we found (Cronbach’s alpha 0.60), except for the NICHD study (Vandell 1996) in which an internal consistency of 0.81 was reported. The authors of the CC-HOME inventories stated, ‘We no longer report internal consistencies estimates for the HOME Inventories. The CC-HOMEs are composed of cause rather than effect indicators and reliability estimates such as the alpha coefficient assume effect indicators’ (Bradley, Caldwell, and Corwyn 2003, 308). In addition, the dichotomous nature of the items might have led to an underestimated correlation. Finally, our data are derived from a cross-sectional design, making causal inferences impossible. The direction of effects may not be from caregiver sensitivity to involvement. That is, caring for less involved children may be more challenging than caring for children who are more involved. Although it seems more natural that children’s involvement is affected by their caregivers’ behavior than vice versa, these issues require further examination with experimental data or in longitudinal studies with larger sample sizes.

To conclude, we found that boys were more susceptible to caregiver sensitivity than girls: boys’ involvement scores were higher (than girls) when receiving more sensitive care and lower when receiving less sensitive care. For girls, there was no association between caregiver sensitivity and involvement. This study shows that it is important, especially for boys, to be taken care of by a sensitive caregiver during the day at childcare: a caregiver who accurately perceives their signals and who responds promptly and adequately to these signals. These caregivers support and encourage the children to be involved in their play and to explore the world around them.

Disclosure statement

No potential conflict of interest was reported by the author(s).