Erasmus MC heeft ervoor gezorgd dat je Mijn BSL eenvoudig en snel kunt raadplegen. Je kunt je links eenvoudig registreren. Met deze gegevens kun je thuis, of waar ook ter wereld toegang krijgen tot Mijn BSL.
Om ook buiten de locaties van Erasmus MC, thuis bijvoorbeeld, van Mijn BSL gebruik te kunnen maken, moet je jezelf eenmalig registreren. Dit kan alleen vanaf een computer op een van de locaties van Erasmus MC.
Eenmaal geregistreerd kun je thuis of waar ook ter wereld onbeperkt toegang krijgen tot Mijn BSL.
Login
Als u al geregistreerd bent, hoeft u alleen maar in te loggen om onbeperkt toegang te krijgen tot Mijn BSL.
This study aimed to validate the Chinese version of the Infant/Toddler HOME and Early Childhood HOME inventories. Participants included 131 Chinese families with children aged 3 years 4 months to 6 years, and 129 families with children under 3 years 4 months. The HOME inventory was completed by trained social workers during home visits. Parents completed measures on their parental stress, their children’s behaviour, and provided demographic information. Children’s development was assessed by individual assessment (early childhood group) or observation (infant/toddler group). HOME scores correlated negatively with child behaviour problems and parental stress, and positively with child development. Families with higher income and higher parental education levels reported higher HOME scores. Rasch analysis results indicated that the HOME inventories largely fulfilled the unidimensionality requirement though the items loaded on the lower end. The Chinese version of the HOME inventory is considered a promising tool for assessing family environment in Chinese families.
Opmerkingen
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Introduction
According to the ecological theory, a child’s development is shaped by the child’s own characteristics, the child’s family and school, the wider community and the cultural context (Bronfenbrenner, 1979). In understanding child development and family dynamics, it is important to understand the child’s family and the kind of environment available to the child. A reliable, valid and culturally appropriate instrument with good measurement properties is needed to measure the family environment. This allows professionals and researchers to assess, understand and support the development of children in their natural environment, to promote the quality of life of families and nurture the success of the children. By doing this, professionals and researchers could obtain a more complete view of the child’s needs and strengths, build trust and rapport with families, address environmental factors contributing to optimal child development, promote parent-child interaction, enhance quality of life in the family, and facilitate individualized support.
The Home Observation for Measurement of the Environment (HOME) inventory is a widely used instrument for measuring home environment (Bradley, 2015). There are four standard versions of the HOME inventory. The Infant/Toddler HOME (IT-HOME) is designed for families with children aged 0–3 years. The Early Childhood HOME (EC-HOME) is designed for families with children aged 3–6 years. The Middle Childhood HOME (MC-HOME) is designed for families with children aged 6–10 years. The Early Adolescent HOME (EA-HOME) is designed for families with adolescents aged 10–16 years. (Goemans et al., 2016; Bradley, 2015).
HOME is designed and theorized to be social determinants predicting cognitive development (Bradley & Corwyn, 2003), social development (Wu et al., 2012), maladaptive behaviors (Lee et al., 2013), and health (Bradley, 2015). HOME scores have been found to be positively associated with family income and wealth, education level of parents, maternal intelligence, access to social support, family status (two-parent families), and negatively associated with household crowding, maternal depression and parental stress (Bradley, 2015; Goemans et al., 2016). HOME scores are also positively associated with child outcomes such as language, cognitive development and achievement, but negatively associated with child behaviour problems (Bradley, 2015; Goemans et al., 2016). As there is no locally validated measure on home environment in Hong Kong, the association between home environment and child outcomes in Hong Kong can be inferred from studies which demonstrated that authoritative parenting style and healthy family feeding practice were negatively associated with child behaviour problems in preschool children (Leung et al., 2018). Parenting stress was positively associated with child behaviour problems (Leung & Tsang, 2010). In a longitudinal study on Hong Kong preschool children, it was found that children’s learning was positively related to authoritative parenting style, parental control and parents eating meals together with children (Leung et al., 2017). Parental involvement was a mediator between socioeconomic status and child school readiness (Wong et al., 2018). Placement of electronic devices in bedroom (Fu et al., 2017) was associated with lower school readiness in children. Parenting style is related to Acceptance and Responsiveness in the HOME inventories, whereas mealtime practices and use of electronic device are related to Modelling. Parents having meals with children is related to Variety in the HOME inventories.
HOME has been used as a standard instrument in diverse cultures to assess the quality of the stimulating home environment and its significance in contributing to outcomes in child development, parent-child relationships, and school readiness. Underweight was significantly correlated with lower overall IT-HOME scores as well as lower scores in Organization and Learning materials subscales in Avan et al.’s (2014) cross-sectional study in 26 communities in Pakistan. The low IT-HOME scores may indicate families lacking access to external resources in Nepal (Parajuli et al., 2014). Insecure-resistant attachment was associated with lower scores on the Organization and Learning materials subscales in IT-HOME in Indonesia (Zevalkink et al., 2008). The continuous impact of the home environment is evident in longitudinal studies (Korucu & Schmitt, 2020; Son & Morrison, 2010). High scores on the EC-HOME with Learning materials, Language stimulation, Academic stimulation and Variety subscales at age 3 predicted better performance in letter-word recognition and problem solving at age 5 in a national child study (Son & Morrison, 2010). Similar findings were found in Korucu & Schmitt’s (2020) national family and child study across the United States indicating that a high-quality home environment at age 3 predicted gains in attention regulation and language skills at age 5.
There have been many translations of the HOME inventory (e.g., Findik & Aral, 2023; Glad et al., 2014; Goemans et al., 2016; Holding et al., 2011), but there is no available validated Chinese version yet. Practitioners and researchers in Chinese settings do not have a comprehensive family assessment framework or checklist for measuring the home environment of families with preschool children. At present, there is only one evaluation framework available for home visits involving children aged 0–6 years in Hong Kong, The Parenting Capacity Assessment Framework (Education Bureau, Department of Health, Hospital Authority, & Social Welfare Department, 2023). This framework primarily focuses on assessing parental physical care of the children and does not offer a comprehensive evaluation of the home environment, educational materials, language stimulation, and parenting practices. This study aimed to fill this research gap.
The aim of the present study was to validate the Chinese versions of the IT-HOME and EC-HOME among a Chinese sample in Hong Kong. In addition to an examination of the measurement properties of the HOME inventory using Rasch analysis, the validity of the HOME instrument was examined in terms of its associations with household income, parent education level, parental stress, and child outcomes including behaviour and cognitive outcomes. Interrater reliability was examined as the instrument was based on scoring by an observer. Test-retest reliability was also examined.
There were three hypotheses in this study. The first hypothesis was that the IT-HOME and EC-HOME scores would correlate negatively with child behaviour problems and parental stress. The second hypothesis was that the IT-HOME and EC-HOME scores would correlate positively with child development. The third hypothesis was that families with higher income and parental education would achieve higher scores on IT-HOME and EC-HOME.
Method
Participants
The participants included 131 Chinese families with children aged 3 years 4 months to 6 years, and 129 families with children younger than 3 years 4 months. The participants in the older group were recruited by preschool-based social workers of Social Work Service for Pre-primary Institutions from 39 kindergartens located in different districts of Hong Kong. Some participants were service recipients of preschool rehabilitation services of a nongovernmental organization (NGO). The participants in the toddler group were mainly recruited through service recipients from an early childhood intervention program by the same NGO and three families from the social contacts of program staff. The interviewers contacted the participants’ families by phone to obtain preliminary consent for the validation research of the HOME inventory. Subsequently, during home visits, they collected written consent using paper forms. The informants were provided with copies of the consent documents for their records. According to Linacre (1994), a sample size of 100 is sufficient for 95% confidence that no item calibration is more than 0.5 logit away from its stable value. According to Cohen (1992), a sample size of 64 per group is sufficient for medium effect size in comparing two groups (e.g., high versus low-income group) with power =0.80 and α = 0.05.
Measures
Home environment
The Home Observation for Measurement of the Environment (HOME; Caldwell & Bradley, 2016) scales were used to measure home environment. The Infant/Toddler HOME (IT-HOME) is designed for families with children aged 0–3 years and consists of 45 items, with six subscales: Responsivity, Acceptance, Learning materials, Variety; Organization, and Involvement. The Early Childhood HOME (EC-HOME) is designed for families with children aged 3–6 years, and consists of 55 items, with eight subscales: Learning materials, Language stimulation, Physical environment, Responsivity, Academic stimulation, Modelling, Variety, and Acceptance. All items are scored as + (1) or – (0).
Child behaviour
Child behaviour was measured using the Strength and Difficulties Questionnaire (SDQ; Goodman & Scott, 1999). The SDQ consists of five subscales (5 items for each subscale), Emotional symptoms, Conduct problems, Hyperactivity/inattention, Peer relationship problems, and Prosocial behaviour. Parents rate each item on a 3-point rating scale (0: not true to 2: certainly true). Higher scores indicate higher endorsement of the behaviour domain. A Total problem behaviour score could be computed by summing the raw scores from the first four subscales. The Chinese version of the scale was validated by Lai et al. (2010) with reliability estimates above 0.70 for the Total problem behaviour scale and all sub-scales except Peer relationship problems (alpha = 0.45). For those aged under 3 years 4 months, the 2-4-yeaer-old version was used. In the present study, the reliability for Total problem behaviour was 0.70 and that for Prosocial behaviour was 0.86. For those aged 3 years 4 months and above, the 4-17-year-old version was used. In the present study, the reliability for Total problem behaviour was 0.77 and that for Prosocial behaviour was 0.75.
Parental stress
This Parental Stress Scale (PSS; Cheung, 2000) was used to measure parenting stress. There are 17 items on a 6-point rating scale to be completed by parents. The Chinese version was validated by Leung & Tsang (2010), with a reliability above 0.80. Higher scores indicate higher parental stress. In the present study, the reliability for the IT sample was 0.87 and that for the EC sample was 0.88.
Child development (preschool children)
The Cognition Scale of The Hong Kong Comprehensive Assessment Scale for Preschool Children (HKCAS-P; Leung et al., 2013) was used as a measure of child development. This scale is designed for Hong Kong Chinese children aged 3 years 4 months to 6 years 3 months. There are 40 items measuring basic preschool concepts (Leung et al., 2013). The scale is individually administered to children by qualified users who have completed the required training. Age standardized scores are available (mean = 10, sd = 3). In the validation study, the reliability was 0.93.
Child development (infant/toddlers)
The Ages and Stages Questionnaire Third Edition (ASQ-3; Squires & Bricker, 2009) was used to measure the development of infants/toddlers. This questionnaire is a general developmental screening tool to be completed by parents. It consists of five domains (each with 30 items), Personal social, Gross motor, Fine motor, Problem solving and Communication, for children from 1 month to 5.5 years (66 months). The Chinese version of the scale was validated by Bian et al. (2010), with a reliability of 0.77. Higher scores indicate better performance in the respective developmental domain. In another project with Chinese parents, parents reported great difficulties in completing the ASQ. In the present study, the ASQ was completed by two social workers who completed the form independently and then they revisited the interview again to check the performance of the child to check for discrepancies between the two raters. Discrepancies were resolved based on a discussion of what the workers saw and how they rated and the rating criteria.
Demographic information
Parents were requested to provide information on their background such as education level, marital status, employment and family income.
Procedures
The HOME inventory was translated into Chinese and back translated by social workers in the abovementioned NGO. The back-translated version was checked against the original English version by an independent third social worker. These three social workers hold a master’s degree in social work, have achieved a score of 7.5 on the IELTS, are native Cantonese speakers, and possess a minimum of five years of experience in early childhood intervention.
The experienced social workers and research assistants, acting as home visit interviewers, participated in a comprehensive 6-hour training session prior to data collection. During this training conducted by the training coordinator (the third author who is an experienced social worker with over 20 years’ experience working with children and families), the content of the HOME Inventory and interview skills as outlined in the manual were thoroughly reviewed. A demonstration home visit adhering to the standards of HOME Inventory interview skills and scoring was made available online for the home visit interviewers to review at their convenience, further enhancing their training and consistency in scoring.
Experienced social workers and trained research assistants administered the respective versions of the HOME inventories during home visits. Families with children under 3 years 4 months were administered the IT-HOME whereas those with children aged 3 years 4 months or above were administered the EC-HOME. The parents were also required to complete a questionnaire on their parental stress and their children’s behaviour, and to provide some information on their demographic characteristics during home visits. Children aged 3 years 4 months or above were administered the Cognition Scale of the Hong HKCAS-P by an educational psychologist qualified to administer this test. For younger children, they were observed and rated by experienced social workers on the ASQ (Chinese version). The child assessment was conducted in the kindergartens attended by the children, at home, or in NGO centres.
For test-retest reliability, the two versions of the HOME inventory were administered to a randomly selected sample of 40 families four weeks after the first administration, with 20 families from each age group. Another randomly selected 40 families (20 families from each age group) were assessed on the respective versions of the HOME by two social workers to test for interrater reliability. These 40 home visits were video recorded and eight of which have been reviewed by the training coordinator, with 90% agreement between the training coordinator and each of the scorer.
This study was approved by the Human Subjects Ethics Committee of the City University of Hong Kong on July 6, 2021. The Application Number is H002799 and Reference Number is 3-2021-38-F.
Data Analysis
As family structures and dynamics differ across cultures, cross-cultural equivalence for the HOME inventory should be examined in different cultures. Choi et al. (2006) argued for the use of Rasch analysis for this purpose. Rasch analysis, a form of item response theory, is built upon a dichotomous logistic response model that is suitable for the analysis of yes/no items, as in the case of the HOME inventory. It assumes that each item response is a linear probabilistic interaction of the question’s difficulty and the person’s ability (Prieto et al., 2003). The principle of invariant comparison is a defining property of Rasch analysis. In this case, the location of a person can be measured independent of the location of the item distribution and vice versa (Glad et al., 2014). Rasch analysis produces measures that are sample independent, test free, and linear (Choi et al., 2006). In this way, the generality of the findings can be assured, notwithstanding sample size or characteristics. Rasch analysis is useful in the development and evaluation of new (or existing) questionnaires that are to be summed to form a total score. Additivity is related to the issue of unidimensionality, which assumes that all items in a scale measure the same underlying construct. If the scale items do not measure the same dimension, they should not be added up (Bond & Fox, 2007). Rasch analysis was conducted using Winsteps 5.6.3. Unidimensionality was examined using infit and outfit statistics, point measure correlation and principal component analysis of the residuals (PCA). The targeting of the HOME inventories was examined using the Wright map.
The validity of the HOME inventories was investigated through its correlation with measures of parental stress, child behaviour and child development. Socioeconomic differences were investigated by comparing the scores of families from different income groups (above and below median household income) and from parents with different levels of education. Reliability (internal consistency) was assessed using KR-20. Test-retest reliability was assessed using the intraclass correlation coefficient (ICC). Interrater reliability was assessed using the kappa statistics. Data analysis was conducted using SPSS version 29.
Results
The demographic characteristics of the IT-HOME and EC-HOME samples are shown in Table 1.
Table 1
Demographic characteristics of participating families
Variables
IT-HOME
(N = 129)
EC-HOME (N = 131)
Number
Percentage
Number
Percentage
Boys
64
49.6
73
55.7
Girls
65
50.4
58
44.3
Developmental problems
37
28.7
42
32.1
Attending preschool
36
27.9
128
97.7
Not attending preschool
92
71.3
2
1.5
Father
6
4.7
14
10.7
Mother
123
95.3
115
87.8
Others
0
0
1
0.76
Father with tertiary education
34
26.4
33
25.2
Mother with tertiary education
39
30.2
26
19.8
Father with post-secondary education
46
35.7
39
29.8
Mother with post-secondary education
57
44.2
41
31.3
Father employed
95
73.6
100
76.3
Mother employed
52
40.3
45
34.4
Married/de facto
94
72.9
114
87.0
Income ≥$26501
59
45.7
64
48.9
On welfare benefits
30
23.3
17
13.0
Mean
SD
Mean
SD
Child age (months)
19.95
10.31
56.05
10.04
Parent age (years)
31.99
5.93
37.16
6.39
IT-HOME
In terms of unidimensionality, all infit statistics were within 0.6 to 1.4 (Bond & Fox, 2007), except items 2 and 7 of the Acceptance subscale. The outfit statistics of items 3, 6, 7, 10 of the Responsivity sub-scale, items 2, 5, 6, 7 of the Acceptance subscale, items 2 and 4 of the Organization subscale, items 5 and 6 of the Involvement subscale, and items 3 and 5 of the Variety subscale, were outside 0.6 and 1.4. Bond & Fox (2007) argued that more attention should be placed on infit statistics as outfit statistics are sensitive to outlying scores. Another measure of unidimensionality is the point measure correlation and all correlations were positive. Finally, for PCA, the criteria for assessing unidimensionality are (i) at least 40% of the variance should be explained by measures; (ii) less than 15% of the variance should be explained by the first principal component of the residuals; and (iii) a ratio of at least 3:1 is required for the variance explained by measures relative to the variance explained by the first principal component of the residuals (McCreary et al., 2013). In the present case, the variance explained by measures was 32.1% and that explained by the first principal component of the residuals was 6.3%, with a ratio of 5.10:1, fulfilling most of the criteria set out by McCreary et al. (2013). The infit and outfit statistics and point measure correlations are in Table 2.
Table 2
Infit and outfit statistics and point measure correlations of IT-HOME
Items
Infit mean squares
Outfit mean squares
Point measure correlations
Responsivity 1
1.22
1.36
0.33
Responsivity 2
0.89
0.60
0.36
Responsivity 3
0.82
0.55
0.44
Responsivity 4
1.06
1.12
0.42
Responsivity 5
1.02
1.35
0.10
Responsivity 6
0.79
0.35
0.44
Responsivity 7
0.90
0.57
0.34
Responsivity 8
1.01
0.99
0.50
Responsivity 9
0.97
0.87
0.35
Responsivity 10
1.30
1.54
0.23
Responsivity 11
1.07
1.40
0.32
Acceptance 1
1.37
1.36
0.25
Acceptance 2
1.44
2.85
0.07
Acceptance 3
1.22
1.25
0.24
Acceptance 4
1.06
0.91
0.29
Acceptance 5
1.05
0.59
0.18
Acceptance 6
1.06
1.63
0.28
Acceptance 7
1.43
1.43
0.16
Acceptance 8
0.66
0.61
0.71
Organization 1
1.11
1.03
0.40
Organization 2
1.34
1.58
0.25
Organization 3
1.25
1.24
0.35
Organization 4
1.16
3.00
0.09
Organization 5
1.01
0.67
0.41
Organization 6
0.96
0.92
0.41
Learning materials 1
0.78
0.65
0.63
Learning materials 2
0.75
0.65
0.66
Learning materials 3
1.01
0.98
0.50
Learning materials 4
0.80
0.83
0.57
Learning materials 5
0.84
0.70
0.57
Learning materials 6
0.98
0.79
0.43
Learning materials 7
1.08
1.07
0.46
Learning materials 8
0.88
0.91
0.53
Learning materials 9
0.86
0.71
0.54
Involvement 1
1.12
1.28
0.34
Involvement 2
0.82
0.67
0.56
Involvement 3
0.79
0.69
0.64
Involvement 4
0.79
0.62
0.60
Involvement 5
0.68
0.58
0.70
Involvement 6
0.96
0.45
0.31
Variety 1
1.06
1.14
0.46
Variety 2
0.79
0.69
0.65
Variety 3
1.25
1.45
0.32
Variety 4
0.89
0.79
0.53
Variety 5
0.64
0.54
0.72
For targeting, the Wright map showed that the IT-HOME was relatively “easy” for Hong Kong Chinese families (Fig. 1). There were few items targeting the “upper end” families.
Fig. 1
IT-HOME Wright Map
×
The mean infit mean square was 0.98 (sd = 0) and the mean outfit mean square was 1.02 (sd = 0.16). The person reliability and separation scores were 0.86 and 2.43 respectively. Item reliability and separation scores were 0.94 and 4.06 respectively.
For validity, independent t test results indicated that there were significant differences in IT-HOME scores by family income (above and below median household income) and parent education levels (postsecondary or above versus secondary or below). In both cases, families with higher income and parent education levels attained higher IT-HOME scores. The details are shown in Table 3. There were significant negative correlations between IT-HOME scores and parental stress, and positive correlations between IT-HOME scores and child prosocial behaviour, and child development (ASQ-3). The correlations are shown in Table 4.
Table 3
Differences in HOME scores by family income and parent education levels
Variables
Family Income
p level
Father education
p level
Mother education
p level
HK$26,500 or below
HK$26,501 or above
Secondary or below
Post-secondary or above
Secondary or below
Post-secondary or above
IT-HOME
Responsivity
8.39 (2.23)
9.98 (1.38)
<0.001
9.23 (1.97)
9.83 (1.42)
0.097
8.55 (2.33)
9.91 (1.26)
<0.001
Acceptance
5.56 (1.48)
6.25 (1.36)
0.007
5.69 (1.67)
6.41 (1.11)
0.015
5.45 (1.63)
6.40 (1.03)
<0.001
Learning materials
5.07 (3.07)
7.58 (1.46)
<0.001
6.40 (2.91)
7.67 (1.35)
0.008
5.42 (3.13)
7.39 (1.58)
<0.001
Organization
4.21 (1.35)
5.08 (0.95)
<0.001
4.46 (1.34)
5.17 (0.88)
0.003
4.38 (1.35)
5.07 (0.98)
0.001
Involvement
3.76 (1.84)
5.20 (1.20)
<0.001
4.35 (1.58)
5.43 (0.96)
<0.001
3.72 (1.88)
5.28 (1.06)
<0.001
Variety
2.17 (1.63)
4.07 (1.08)
<0.001
2.98 (1.58)
4.20 (1.02)
<0.001
2.31 (1.70)
3.95 (1.25)
<0.001
Total
29.16 (8.33)
38.17 (4.55)
<0.001
33.10 (7.57)
38.72 (4.00)
<0.001
29.83 (8.65)
38.00 (4.52)
<0.001
EC-HOME
Learning materials
6.10 (2.72)
8.13 (2.10)
<0.001
6.52 (2.51)
8.77 (1.60)
<0.001
6.36 (2.74)
8.49 (1.83)
<0.001
Language stimulation
5.61 (1.22)
6.27 (0.93)
<0.001
5.90 (1.13)
6.15 (0.96)
0.234
5.79 (1.19)
6.24 (0.97)
0.040
Physical environment
5.25 (1.81)
6.19 (1.11)
<0.001
5.45 (1.77)
6.31 (1.00)
0.002
5.40 (1.76)
6.22 (1.06)
0.002
Responsivity
5.28 (1.44)
5.52 (1.45)
0.360
5.33 (1.33)
5.49 (1.54)
0.577
5.36 (1.39)
5.46 (1.55)
0.708
Academic stimulation
3.66 (1.65)
4.28 (1.13)
0.012
3.78 (1.51)
4.38 (1.07)
0.017
3.65 (1.64)
4.56 (0.95)
<0.001
Modelling
3.58 (1.14)
3.97 (1.10)
0.051
3.70 (1.17)
4.10 (0.94)
0.071
3.67 (1.16)
4.20 (0.90)
0.012
Variety
5.63 (1.54)
6.69 (1.50)
<0.001
5.82 (1.46)
6.79 (1.49)
0.001
5.88 (1.56)
6.83 (1.34)
0.001
Acceptance
3.64 (0.69)
3.59 (0.73)
0.699
3.69 (0.58)
3.67 (0.62)
0.869
3.62 (0.69)
3.76 (0.54)
0.257
Total
38.76 (7.91)
44.63 (5.50)
<0.001
40.18 (6.89)
45.67 (5.28)
<0.001
39.73 (7.75)
45.76 (5.28)
<0.001
Table 4
Correlations between IT-HOME scores and parental stress, child behaviour and development
Variables
Parental stress
Problem behaviour
Prosocial behaviour
ASQ communication
ASQ gross motor
ASQ fine motor
ASQ problem solving
ASQ personal social
Responsivity
−0.18*
−0.08
0.18
0.27**
0.11
0.15
0.04
0.22*
Acceptance
−0.12
−0.09
0.17
0.31***
0.07
0.22*
−0.00
0.35***
Learning materials
−0.27**
0.12
0.47***
0.03
0.22*
0.18*
−0.10
0.09
Organization
−0.26**
−0.04
0.37***
0.07
0.12
0.17
−0.04
0.14
Involvement
−0.20*
−0.02
0.28**
0.23**
0.23*
0.32***
0.09
0.24**
Variety
−0.28***
0.01
0.25**
0.24**
0.21*
0.26**
−0.03
0.25**
Total
−0.30***
−0.00
0.42***
0.24**
0.22*
0.29***
−0.02
0.27**
* ≤ 0.05
** ≤ 0.01
*** ≤ 0.001
With regard to reliability, the internal consistency (KR-20) was 0.91 and the test-retest reliability (4-week interval with 20 pairs) was 0.85. For interrater reliability, the average kappa was 0.98 (range: 0.90–1.00). The reliability estimates of the subscales are shown in Table 7.
EC-HOME
In terms of unidimensionality, all infit statistics were within 0.6–1.4 (Bond & Fox, 2007). The outfit statistics of item 4 of the Language stimulation subscale, items 1 and 2 of the Responsivity subscale, item 5 of the Modelling subscale, and item 6 of the Variety subscale, were outside 0.6 and 1.4. All point measure correlations were positive. For PCA results, the variance explained by measures was 27.1% and the variance explained by the first principal component of the residuals was 5.4%, with a ratio of 5.02:1, fulfilling two of the criteria set out by McCreary et al. (2013). The infit and outfit statistics and point measure correlations are shown in Table 5.
Table 5
Infit, outfit statistics and point measure correlations of EC-HOME
Items
Infit mean squares
Outfit mean squares
Point measure correlations
Learning materials 1
0.91
0.75
0.45
Learning materials 2
0.95
0.93
0.44
Learning materials 3
1.08
1.07
0.33
Learning materials 4
0.98
0.60
0.31
Learning materials 5
0.87
0.84
0.43
Learning materials 6
0.91
0.80
0.46
Learning materials 7
0.81
0.73
0.53
Learning materials 8
0.87
0.80
0.52
Learning materials 9
1.08
1.05
0.32
Learning materials 10
0.89
0.71
0.42
Learning materials 11
0.84
0.67
0.50
Language stimulation 1
0.92
0.79
0.46
Language stimulation 2
0.91
0.71
0.37
Language stimulation 3
1.01
0.90
0.21
Language stimulation 4
0.87
0.50
0.39
Language stimulation 5
1.10
1.07
0.28
Language stimulation 6
1.10
1.28
0.23
Language stimulation 7
0.99
0.89
0.29
Physical environment 1
0.96
0.84
0.38
Physical environment 2
1.13
1.16
0.14
Physical environment 3
0.85
0.63
0.47
Physical environment 4
1.16
1.14
0.20
Physical environment 5
1.10
1.04
0.31
Physical environment 6
1.13
1.00
0.26
Physical environment 7
1.00
0.70
0.26
Responsivity 1
1.06
1.68
0.26
Responsivity 2
0.92
0.31
0.27
Responsivity 3
1.00
1.32
0.15
Responsivity 4
0.92
0.66
0.37
Responsivity 5
1.02
1.02
0.38
Responsivity 6
1.28
1.30
0.13
Responsivity 7
1.10
1.24
0.28
Academic stimulation 1
0.96
0.76
0.33
Academic stimulation 2
0.85
0.79
0.51
Academic stimulation 3
0.86
0.81
0.50
Academic stimulation 4
0.85
0.69
0.46
Academic stimulation 5
086
0.67
0.49
Modelling 1
0.99
0.98
0.38
Modelling 2
1.11
1.13
0.28
Modelling 3
1.02
1.09
0.33
Modelling 4
1.09
1.15
0.25
Modelling 5
1.11
2.47
0.15
Variety 1
0.84
0.78
0.55
Variety 2
1.06
0.87
0.22
Variety 3
1.12
1.15
0.28
Variety 4
0.95
0.96
0.44
Variety 5
1.10
1.21
0.20
Variety 6
1.24
2.12
0.02
Variety 7
1.07
1.09
0.21
Variety 8
1.06
1.09
0.29
Variety 9
1.02
1.04
0.38
Acceptance 1
1.07
1.17
0.24
Acceptance 2
0.99
1.00
0.31
Acceptance 3
0.99
0.99
0.15
Acceptance 4
1.01
0.80
0.16
For targeting, the Wright map again showed that the EC-HOME was relatively “easy” for Hong Kong Chinese families (Fig. 2). There were fewer items targeting the “upper end” families.
Fig. 2
EC-HOME Wright Map
×
The mean infit mean square was 1.00 (sd = 0.07) and the mean outfit mean square was 0.98 (sd = 0.01). The person reliability and separation scores were 0.82 and 2.10 respectively. Item reliability and separation scores were 0.94 and 4.14 respectively.
For validity, independent t test results indicated that there was a significant difference in EC-HOME scores by family income (above and below median household income) and parent education levels (postsecondary or above versus secondary or below). In both cases, families with higher income and parent education levels attained higher EC-HOME scores. The details are shown in Table 3. There were negative correlations between EC-HOME scores and parental stress and child problem behaviour, and positive correlations between EC-HOME scores and child prosocial behaviour, and child development (HKCAS-P Cognition Scale scores). The details are shown in Table 6.
Table 6
Correlations between EC-HOME scores and parental stress, child behaviour and development
Parental stress
Problem behaviour
Prosocial behaviour
HKCAS-P Cognition
Learning materials
−0.13
−0.20*
0.03
0.14
Language stimulation
−0.22*
−0.20*
0.15
0.14
Physical environment
−0.03
−0.04
−0.08
0.04
Responsivity
−0.27**
−0.27**
0.23**
0.08
Academic stimulation
−0.15
−0.22*
0.09
0.09
Modelling
−0.33***
−0.14
0.16
0.19*
Variety
−0.26**
−0.23**
0.15
0.09
Acceptance
−0.09
−0.06
0.12
−0.01
Total
−0.28***
−0.28***
0.15
0.16
* ≤ 0.05
** ≤ 0.01
*** ≤ 0.001
With regard to reliability, the internal consistency (KR-20) was 0.86 and the test-retest reliability (4-week interval with 20 pairs) was 0.97. The test-retest reliability is comparably higher than 0.75 as reported in Findik & Aral’s (2023) study. For interrater reliability, the average kappa was 0.93 (range: 0.64–1.00). The reliability estimates of the subscales are shown in Table 7.
Table 7
Reliability estimates of IT-HOME and EC-HOME
Internal consistency (KR-20)
Test-retest reliability
IT-HOME
Responsivity
0.73
0.93
Acceptance
0.54
0.93
Learning materials
0.85
0.67
Organization
0.45
0.93
Involvement
0.77
0.76
Variety
0.74
0.96
EC-HOME
Learning materials
0.79
0.95
Language stimulation
0.37
0.90
Physical environment
0.70
0.98
Responsivity
0.57
0.86
Academic stimulation
0.77
0.92
Modelling
0.36
0.83
Variety
0.42
0.90
Acceptance
0.45
0.93
Discussion
In studying the psychometric properties of the Chinese version of EC-HOME and IT-HOME, we have examined its reliability in terms of internal consistency, test-retest reliability and interrater reliability. We have also examined the validity of the EC-HOME and IT-HOME in terms of their correlation with various constructs guided by the literature, including child behaviour problems, child development, parental stress and family income and father and mother education. We have also examined their measurement properties using Rasch analysis.
In terms of the validity of the EC-HOME and IT-HOME, Hypothesis 1 on the correlation between the HOME inventory scores and child behaviour problems and parental stress was supported. There were negative correlations between parental stress and both IT-HOME and EC-HOME scores. There were also positive correlations between child prosocial behaviour and both IT-HOME and EC-HOME scores. Furthermore, there was a negative correlation between EC-HOME scores and child problem behaviour. The findings are consistent with Bradley (2015) and Goemans et al. (2016). Similar findings are found in Korucu & Schmitt’s (2020) longitudinal study that home quality at age 3 predicted problem behaviour at age 5 as well as in Son & Morrison’s (2010) longitudinal study.
Hypothesis 2 on the correlation between the HOME inventory scores and child development was supported. Both IT-HOME and EC-HOME scores correlated positively with child development (individual child assessment in the case of EC-HOME and social worker observation in the case of IT-HOME). Recent studies have continuously shown evidence that a stimulating home environment has a significant positive and consistent association with positive child outcomes (Avan et al., 2014), and school readiness (Korucu & Schmitt, 2020; Son & Morrison, 2010), and is particularly important for young children in less advantaged families (Frongillo et al., 2022; Harden & Whittaker, 2011; Wu et al., 2012).
For Hypothesis 3 on differences by family income and parent education, the results indicated that families with higher income and parent education levels achieved higher scores on both IT-HOME and EC-HOME. The results are consistent with Bradley (2015), Holding et al. (2011) and Goemans et al. (2016). Both low family income and a low-quality home environment are risk factors for intergenerational cycles of poverty to be closely monitored across countries (Frongillo et al., 2022), child maltreatment (Lee et al., 2013), and poor sleep hygiene (Doane et al., 2019).
With regard to the measurement properties of IT-HOME and EC-HOME, the Rasch analysis results largely supported the unidimensionality of the scales. In terms of targeting, both the IT-HOME and EC-HOME items were on the “easy” end for Hong Kong families, with fewer items on the “difficult” end. In the context of early intervention where families at risk are to be identified for support, the IT-HOME and EC-HOME in their present forms are not inappropriate. Families which have difficulties achieving the “easy” items (where most Hong Kong families can achieve) are families who are mostly likely to need support. With the IT-HOME and EC-HOME targeting “easy” items, service providers can identify at risk families with difficulties and provide support to them. For reliability, the internal consistency, test-retest reliability and interrater reliability of both the IT-HOME and EC-HOME were above 0.80.
The Chinese version of IT-HOME and EC-HOME used in the current study also provides comparable and consistent evidence found in other Asian countries. Responsive parent-child interaction and enriching early home environment are connected with various dimensions of child development, including physical development in Pakistan (Avan et al., 2014), social development in Indonesia (Zevalkink et al., 2008), learning motivation in Korea (Choi & Cho, 2020), and overall developmental outcomes in Turkey (Findik & Aral, 2023). Xu et al. (2024) used IT-HOME to examine criteria-related validity of the Parent-Child Interaction Teaching Scale. HOME offers a capacious view of the complex relations among home environment and child outcomes and serves both as predictors of physical and social affordances and outcomes of parenting enhancement in Chinese and Asian contexts.
In summary, there is promising evidence on the the reliability, validity and measurement properties of the Chinese versions of the IT-HOME and EC-HOME. It could be considered a tool with some potential for researchers and service providers who are working with Chinese population, though further studies with a larger sample is needed.
There are some limitations in the present study. First, although the sample size was adequate for Rasch analysis, correlation and tests of group differences, the sample size was inadequate for confirmatory factor analysis. We could not test whether the original model could be reproduced or not. Second, we did not investigate the predictive validity of IT-HOME and EC-HOME in terms of the children’s behaviour and cognitive/academic achievement in primary or secondary schools. Third, although the reliability estimates (internal consistency) of the overall scales were satisfactory, the reliability estimates of some subscales were less satisfactory. Interpretation of test results would need to take this into consideration. Fourth, the items of IT-HOME and EC-HOME were found to be on the “easy” side for Chinese population. While this might be acceptable in terms of the identification of at risk families for early intervention service, it may be less adequate in research on family environment and child outcomes or predictors of family environment as there are few items measuring high functioning families. This may produce a ceiling effect. Fifth, the IT sample and some of the EC sample were service users of a social service agency. Generalization of the results to the general population who are not using social services would need to take this into consideration.
In the context of clinical practice and practical applications, the HOME inventory ensures thoroughness where practitioners can assess all the necessary areas while assessing the home environment of the child (Findik & Aral, 2023; Goemans et al., 2016). It promotes consistency in practice by providing a clear set of guidelines and procedures for assessing the home environment. This can be particularly important where multiple helping professionals may be working with the same client. As a standard tool, multidisciplinary professionals share similar concepts, knowledge and practice of how a high-quality home environment can be constructed (Hughes-Belding et al., 2022; Saitadze, 2021).
Moreover, the use of the HOME inventory can enhance accountability in clinical practice by providing a record of what was covered during family assessment or home visitation (Goemans et al., 2016). By having a clear record, practitioners can demonstrate that they have fulfilled their professional responsibilities and provided appropriate levels of service. The HOME inventory provides a reachable standard for the family to establish a quality home environment for their child (Avan et al., 2014). During a family assessment/visitation, client-worker communication can be improved by providing a clear outline of the assessment/visitation process by using HOME inventory. This can help families understand what to expect during the assessment/visitation, and can also help them feel more comfortable and at ease (Findik & Aral, 2023).
With increasing emphasis on practice research or program evaluation in clinical practice, the HOME inventory can facilitate evaluation by providing clear information on family situations before and after intervention. Practitioners can design an intervention plan by examining the results of the HOME inventory, such as providing tailor-made interventions in specific domains. In addition, the differences in data results before and after intervention can demonstrate the quality/effectiveness of intervention of practitioners. Given the standard assessment of HOME, timely intervention to provide a stimulating environment for young children in socioeconomically diverse families not only on a microlevel (Doane et al, 2019; Wu et al., 2012), but also on a macrolevel could provide systematic strategy for family support policy and services (Frongillo et al., 2022; Saitadze, 2021).
The HOME inventory provides a set of major domains or areas on family environment for frontline workers such as psychologists and social workers to assess the needs of families and to develop programs to meet the needs of the families. It is also a set of useful guidelines for schools and educators so they can work together with parents to promote a nurturing environment conducive to the optimal development of children. For policy makers, the inventory can also help them understand the policy areas to focus on as proposed in Bradley (2015). If population surveys using the HOME inventory can be conducted, the results could inform policy makers on areas where more support for families should be provided as evident in Hughes-Belding et al. (2022).
The validated HOME Inventory has been widely adopted within early childhood intervention services of the nongovernmental organization mentioned above, including a project which supports families with children under the age of three, as well as preschool-based support services for children with developmental difficulties. The organization has made online training available on their website, enabling professionals to learn how to effectively implement the HOME Inventory in their respective settings. Users have reported that the HOME Inventory offers clear guidelines and intervention directions during home visits. Notably, children’s developmental progress, as measured by the ASQ, showed improvement after the implementation of the recommendations outlined in the HOME Inventory.
Overall, the HOME inventory is a potential tool for practitioners and researchers in assessing the family environment and conducting home visitations, as well as research on family environment, among Chinese famillies. The HOME inventory can help professionals provide high-quality services to their clients by promoting thoroughness, consistency, accountability, communication, and evaluation. It is recommended that there should be further studies with a larger and representative sample with different age groups and socioeconomic status to understand the factor structure of the HOME inventory in Chinese societies, to establish a norm for Chinese families, and to test the validity of the inventory with families in other Chinese-speaking societies. Longitudinal studies should be conducted to examine the predictive validity of the HOME inventory.
Compliance with ethical standards
Conflict of interest
The authors declare no competing interests.
Ethical approval
Ethics approval was obtained from City University of Hong Kong (application number: H002799). All parents provided written consent for participation.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
share
DELEN
Deel dit onderdeel of sectie (kopieer de link)
Optie A:
Klik op de rechtermuisknop op de link en selecteer de optie “linkadres kopiëren”
Met BSL Psychologie Totaal blijf je als professional steeds op de hoogte van de nieuwste ontwikkelingen binnen jouw vak. Met het online abonnement heb je toegang tot een groot aantal boeken, protocollen, vaktijdschriften en e-learnings op het gebied van psychologie en psychiatrie. Zo kun je op je gemak en wanneer het jou het beste uitkomt verdiepen in jouw vakgebied.