The study was approved by the Research Ethics Committee at the authors’ institution. Written consent was obtained from the participants before the study. All participants were randomly selected among the parents of preschool children, aged 2.5–6 years, in Henan Province, China. For the interviews, we randomly selected 24 parents (16 mothers and eight fathers) and six grandparents who had lived with preschool children and their parents for a long time. We interviewed them face-to-face about their opinions of what constitutes parental support and involvement. In the following study, the scales was distributed and collected through the Internet. Web scales were selected because of their ability to reach a large number of parents from different areas. In Study 2, a total of 483 scales were distributed, and 447 were collected, with a collection rate of 92.5% (Table 1). In Study 3, 502 scales were distributed randomly, and 471 were collected, with a collection rate of 93.8% (Table 2). Study 4 had two different samples. Among Sample 1, 480 scales were distributed, and 442 valid scales were obtained, with an effective rate of 92.1% (Table 3). Among Sample 2, 350 scales were distributed, and 318 valid scales were obtained, with an effective rate of 90.9% (Table 4).

Chinese versions were available for all questionnaire tools that were used. Each measure had previously been shown to have satisfactory reliability and validity.

The Parental Burnout Assessment (PBA) was designed by Roskam et al. (2018), with 22 items that measured four dimensions: exhaustion in one’s parental role, contrast with previous parental self, feelings of weariness with one’s parental role, and emotional distancing from one’s children (e.g., “I can’t stand being a parent anymore”; “Apart from the usual care activities [driving, sleeping, and eating], I can’t pay more for my children”). Cheng et al. (2020) revised the Chinese version with a total of 23 items, using the 7-point scale, in which a higher score indicated a higher number of burnout symptoms. In the present study, Cronbach’s α was 0.926.

The Chinese version of the Perceived Social Support Scale (PSSS) was revised by Jiang (2001) based on the original scale of Zimet et al. (1988; e.g., “I can freely express my opinion to my partner or group of friends”; “If I have problems, my friend/partner would help me”). The scale emphasizes an individual’s perception of social support, including the degree of support from family, friends, and others (Cohen and Matthews, 1987). The PSSS has 12 items and a Likert-type 5-point scale, from 1 (complete agreement) to 5 (complete disagreement). Higher total scores indicate higher perceived support. In the present study, Cronbach’s α was 0.925.

The Chinese version of the University of California, Los Angeles, Loneliness Scale (Wang, 1995) was revised by Wang based of the original scale of Russell et al. (1980; e.g., “How often do you feel as if nobody really understands you?”). It was based on the results from 320 college students, and an 18-item version was formed. Ten items among them were in reverse wording. Cronbach’s α was 0.895. The correlation of even-number items and odd-number items was 0.869, with an adjusted split-half reliability score of 0.925.

The Marriage Satisfaction Scale was a subscale from the Marital Gratification Scale, which measures marriage quality (Schaefer and Olson, 1981). This scale contained 10 items, including five reverse-scoring items (e.g., “I don’t like my spouse’s personality and personal habits”; “I am very satisfied with the responsibilities of both spouses in the marriage”). These items were rated from 1 (indeed) to 5 (indeed not). Higher scores indicate higher marital satisfaction of parents. In the present study, Cronbach’s α was 0.869.

The Children’s Self-Efficacy Assessment Questionnaire (CSEAQ; Huang and Yang, 2015) was revised based on the General Self-Efficacy Scale (GSES; Schwarzer et al., 1997, 1999). The GSES originally consisted of 20 items (e.g., “I am confident that I can deal with any unexpected things effectively”) and was reduced to 10 items. The Chinese version with six items, scored from 1 (extremely inconsistent) to 5 (fully compliant), was used. Higher scores indicated higher self-efficacy of children. In the present study, Cronbach’s α was 0.816.

Demographic items were collected, including the children’s gender and age and parents’ socioeconomic status and education level.

During the first stage of developing the present scale, an extensive literature review and interviews were conducted to generate the scale’s items. For the literature review, the keywords “parental involvement and support,” “parental involvement and support of preschool children,” and “family support” were searched in both Chinese and English literature databases. Parents and grandparents were asked five questions about parental support and involvement, which were compiled by the authors and experts in preschool education. For example, “What kind of parental involvement and support do you provide to your child?” “Which aspect of parental involvement and support do you want to improve the most?” “Is there any pressure on your family to support your child? What are the pressures?” By collating the literature and interview data, we found that parental involvement and support for preschool children were primarily reflected in daily life, material conditions, emotions, education, quality time, mental health, play, and entertainment.

In Study 2 and Study 3, two rounds of item analysis and Exploratory Factor Analysis were conducted to examine the appropriateness of the items and ensure validity of the scale. Through correlation analysis and difference testing between high and low groups, inappropriate items were eliminated, and Principal Components Analysis and maximum variance were used to analyze the data. Common factors were extracted to explore dimensions and structural validity of the scale, and dimensions that were formed by common factors were named.

In Study 4, Confirmatory Factor Analysis and correlation analysis were conducted for the data from Sample 1 to test constructive validity and internal validity of the CPISSPC. To test concurrent validity of the CPISSPC, the data from Sample 2 were used to analyze correlations between parental involvement and support, parental burnout, perceived social support, loneliness, marriage satisfaction, and children’s self-efficacy.

These Exploratory Factor Analyses and correlation analyses were performed using SPSS 22.0 software. AMOS 24.0 software was used for the Confirmatory Factor Analysis.

The first author of the present study created an initial pool of 58 items based on the literature search and interview results, which were reviewed by other authors. Several of the items were revised. Three experts in preschool education and five parents of preschool children were then invited to evaluate content validity of the scale and revise the items according to the following criteria: (1) items with unclear and inaccurate language expressions, (2) repetitions of other items, (3) items that were difficult for parents of preschool children to understand, and (4) items that did not reflect well the content of family support for preschool children. Finally, experts and parents of preschoolers were asked to review the revised items again. Through the literature review and interviews, we preliminarily determined the content of parental involvement and support for preschool children and compiled 58 items. We then invited experts to modify the content and language of the items. We then preliminarily determined the initial scale.

The item analysis included calculating the critical ratio (CR) and performing the item homogeneity test. The CR was used to calculate the difference of each item for the 27% highest scores group versus the 27% lowest scores group. Items were deleted if the CR was less than 3. Item-total statistics were used to measure content homogeneity between items, with a value greater than 0.3 indicating that the item correlated well with the total scale (Table 5). A total of 47 items survived the two criteria of the item analysis.

Exploratory Factor Analysis of the 47 items was then conducted using Principal Components Analysis and maximum variance. The results showed that the Kaiser–Meyer–Olkin (KMO) value was 0.923, which is greater than 0.6, and the Bartlett test value was 8242.141 (df = 1081, p < 0.001), indicating that there were common factors in the correlation coefficient matrix and that the scale items possessed high commonality, which was suitable for factor analysis.

Six common factors with eigenvalues greater than 1 were extracted, and inappropriate items were deleted according to the following two conditions: loading value less than 0.4 and cross-loading value on several common factors. As a consequence, 17 items were deleted. The cumulative explanatory variance of the six common factors was 55.6% (Table 6). Next, the six dimensions were named educational method (Factor 1), psychological concern (Factor 2), life care (Factor 3), understanding of play (Factor 4), educational behavior (Factor 5), and educational concept (Factor 6).

In this round of item analysis, all items and total scores were significantly correlated at a level of 0.01 (Table 7). Specifically, 26 items had a correlation coefficient greater than 0.5, accounting for 74.3% of the total number of items. The correlation coefficients of the six items were greater than 0.4, accounting for 17.1% of the total items, and the correlation coefficients of the three items were greater than 0.3, accounting for 8.6% of the total items.

To form a more stable dimension structure of the scale, Principal Components Analysis was performed on 35 items. The KMO value of the Exploratory Factor Analysis in the second round of analysis was 0.895 (i.e., greater than 0.6). The Bartlett test value was 3029.383 (df = 153, p < 0.001), thus demonstrating its suitability for factor analysis. According to the standard of deleting items that were used in the first round of testing, 18 items were retained.

Finally, four stable dimensions were formed, with four to six items in each dimension (Table 8). The four factors were named psychological support (Factor 1), play support (Factor 2), educational support (Factor 3), and life support (Factor 4). The items and dimensions of the scale were then determined, and exploratory analysis was completed.

Construct validity was investigated in two aspects. The first was to explore factors that affected parental involvement and support from aspects of the parents. The second was to explore the impact of parental involvement support on children from aspects of the children. The results showed that a high level of marital gratification and perceived social support positively correlated with parental involvement support. In contrast, parents’ high levels of loneliness and burnout could hinder parental involvement and support. These results indicated that if we want to encourage parents to have more involvement and provide more support for their children, then we need to pay attention to reducing their burnout and loneliness, helping them improve their marital satisfaction, and providing more social support. We also found that when parents devoted themselves more to raising their children, their children’s self-efficacy improved. This result provided further evidence of the positive impact of parental involvement and support on children’s development.

Life support or material support is the basis for the development of preschool children. Surprisingly, however, no scale has been previously developed for preschool children that measures life support that is provided by children’s families. In the CPISSPC, the life support factor consists of four items: balanced nutrition, providing children with their own bed or room, buying clothes regularly, and prioritizing the quality of clothes. A previous study explored the relationship between dietary patterns and cognitive development and found higher full-scale IQ at 4 years of age in children who consumed more fruit, vegetables, and food overall (Gale et al., 2009). For preschool children, the rate of physical development is rapid (Tran et al., 2019), so clothing should be selected with “short-term” use in mind. Therefore, we determined that life support or material support should be included in the present parental involvement and support scale.

Psychological support was an important factor in the CPISSPC. Previous scales about parenting also included some aspects of psychological support. However, compared with the previous scales, the CPISSPC has some advantages. For example, the CECPAQ has a support dimension (13 items) that refers to the extent to which parents are attuned, supportive, and acquiescent to their child’s needs and demands (Dong et al., 2021). In the CPISSPC, psychological support includes six items: treat the child in a friendly manner, communicate with the child about daily life, trust the child, praise the child, express love toward the child, and accompany the child while playing. In our opinion, the CECPAQ mainly focuses on children’s emotional needs, whereas the CPISSPC covers a wider range.

The educational support factor in the CPISSPC consists of encouraging the child to play with peers, independence, politeness, and conflict resolution. Playing with peers, independence, and politeness have been included in previous parenting scales. However, these previous parenting scales failed to include the conflict resolution item. For example, the “play with other children” item in the present scale is very similar to the “I regularly let my child play with other children” item in the CECPAQ, which belongs to the “exposure” dimension (Dong et al., 2021). The CRPR also has items on encouraging independence (“If my child gets into trouble, I expect him or her to handle the problem mostly by himself or herself”; Chen et al., 1998). In the CPPM (Wu et al., 2002), the “encouragement of modesty” dimension (Wu et al., 2002) is similar to the politeness item in the CPISSPC. The conflict resolution item is not seen in other scales. Considering that children often have conflicts over toys, parents should teach their children to resolve conflicts with peers.

Play is key to preschool children’s learning, development, confidence, and wellbeing (Yogman et al., 2018). Playing with toys is an important form of play. Parents generally view toys as important for children’s development. Nonetheless, the majority of preschool children scales do not measure the content of toys. Only the CECPAQ has a four-item subscale that measures the extent to which parents engage with their child while playing with toys. These four items only involve four kinds of toys. In the CPISSPC, the play dimension also includes four items: two items that assess parents’ attitudes about toys, one item that evaluates whether parents provide space for children to play with toys, and one item that asks parents if they exchange ideas and experiences about toys after their children play with toys. Unlike the toys subscale in the CECPAQ, which only focuses on toy types, the play dimension of the CPISSPC covers a wider range.