An initial validation of a home–school partnership questionnaire related to teachers’ attitudes and actions: Predicting teachers’ burnout

ABSTRACT

Home-school partnership is key to students’ success at schools. Japanese schools have attempted to invite families to participate in school educational activities. However, these efforts are embedded into Japanese school cultures and were not studied empirically. This study aimed to explore teachers’ attitudes and actions toward home–school partnership and examine their effects on teachers’ burnout. Based on a preliminary study, we developed a questionnaire with 30 items for the partnership attitude (PAT) scale, 37 items for the partnership actions (PAC) scale and three other scales. We examined the concurrent validity of the scales with a sample of 254 teachers. Exploratory factor analysis yielded four subscales for the PAT (Appreciation, Positive Attitude, Traditional Attitude, Difficulty) and five subscales for the PAC (Respect Parents, Individual Contact, Classroom Information Sharing, Being Proactive, Facilitating Conversation). Reliability was moderately supported, while the criterion validity coefficients were moderately supported using correlation analysis. The results of the multiple regressions show that some PAT factors had negative effects, while some PAC factors had positive effects on teachers’ burnout. Thus, while it is important for teachers to build a good home–school partnership through partnership actions, a systematic support system is necessary for teachers experiencing difficulties in collaborating with parents.

KEYWORDS:

• Home-school partnership
• partnership attitudes
• partnership actions
• teachers
• burnout

Introduction

Children spend most of their time at home and school (Bronfenbrenner, 1979). According to the ecological model (Bronfenbrenner, 1979), home and school are regarded as micro-systems, in which children directly gain various experiences. In the same model, the relationships between the home and school are considered mezzo systems (interaction among microsystems), and they have a significant influence on children’s lives; thus, if home and school have inconsistent attitudes toward children, they may become confused and subsequently develop internal conflicts owing to not knowing which taught attitude they should follow. Therefore, homes and schools have to collaborate with each other to support their children.

Home-school partnership

The topic of partnership establishment between teachers and parents has been studied in various countries (Chen, 2019; Christenson & Sheridan, 2001; Sheridan et al., 2010). This includes home-school partnership, home-school collaboration (Christenson & Sheridan, 2001; Sheridan & Gutkin, 2000), and home-school contact (Chen, 2019). In school psychology, many studies focus on the home–school partnership construct (Buerkle et al., 2009; Sheridan et al., 2010). Sheridan et al. (2010) provided a framework to develop a connection between school and parents called the five A’s: “approach,” “attitude,” “atmosphere,” “action,” and “achievement.”

“Approach” refers to the philosophy or orientation implemented by schools at a system level; the approach of the school is used to establish a tone, method, and manner by which it can enable the construction of its relationship with family members and encourage family participation in the school setting (Christenson & Sheridan, 2001). “Attitude” refers to the feelings, emotions, and positions that family members and school personnel hold about each other, their relationships, and each other’s roles and responsibilities (Christenson & Sheridan, 2001). This construct allows for the examination of whether the school considers the children’s – and their families’ – values, lifestyles, and culture as factors that are promoting or hindering children’s growth and development. “Atmosphere” refers to the physical and affective climate of the school that helps parents understand that their participation in school is welcome. This construct forms the foundation for a home–school relationship and predicts parents’ willingness to support, together with educators, their children in school. “Action” refers to the strategies/practices used by the school to allow for an effective home-school co-work environment; that is, the actions taken by the school to coordinate all systems responsible for supporting children in their education.

Sheridan et al. (2010) stated that appropriate approach, attitude, and atmosphere of a school are preconditions for the promotion of effective actions toward the establishment of a healthy home–school relationship. Notwithstanding, meeting the standards of such appropriateness in these three prerequisites necessitates taking effective actions to promote a partnership between home and school. Moreover, when these four preconditions are followed effectively, “achievement” will be improved (Sheridan et al., 2010, pp. 441–443).

Characteristics of Japanese home-school partnership and recent trends

Japanese schools have a long tradition of valuing the home–school partnership, although the effect has not been empirically examined. They try to invite families to participate in educational activities in the school environment. For example, Japanese schools provide many events in which students’ families and community members can participate, such as entrance ceremonies, graduation ceremonies, sports festivals, cultural festivals, and choral festivals. These events are mainly prepared for 1 month in advance on average and conducted, by teachers and students. Teachers use these events to give students the opportunity to learn social-emotional skills such as working collaboratively with their peers while sustaining their emotional stability (Taruki & Ishikuma, 2006). These events are often open to the community, with the intent to bring families and the community to school to share and celebrate students’ performances (Japanese Ministry of Education, Culture, Sports, Science, and Technology [MEXT], 1996).

Besides these school-wide events, homeroom teachers conduct class-wide parent-teacher meetings each semester,¹ and individual parent-teacher meetings at least once a year. Teachers also visit each students’ homes at the beginning of the school year at many schools, usually in May, to build up relationships with parents and collect information on students’ home environments (Ishikuma, 1999). Thus, it is clear that Japanese schools have traditionally emphasized collaboration between the families and schools, which helps with the prevention of problematic behavior for many children.

However, the picture of home-school collaboration has been changing in recent years in Japan. Many educational reforms have begun in the last 20 years in order to meet students’ individual needs in response to the increase in absenteeism (futoko, chronic absenteeism) and bullying (Yoshikawa et al., 2019). The system of placing school counselors began in 1997 and that of school social workers in 2008. Additionally, special support education began in 2007. It requires schools to create individual support plans and provide special support for children with special needs (Yoshikawa et al., 2019). Nevertheless, the number of students in classrooms remained unchanged during this period.

Moreover, the School Education Law in Japan allows a classroom to have up to 40 students (MEXT, 2010). This means that homeroom teachers typically support 30 to 40 students, academically and behaviorally. Even though the system of placing mental health professions at schools has started, school counselors provide services for only six to eight hours per week: school social workers visit schools only upon request (MEXT, 2016). Consequently, teachers do not have enough support to deal with students that have significant behavioral problems and/or difficult parents. Thus, it can be assumed that all responsibilities are placed mainly on teachers (especially home-room teachers) who have their own duties teaching subjects, managing their classroom, and managing school events.

Mental health issues among Japanese teachers

Under these circumstances, the mental health issues among Japanese teachers are drawing attention. According to the MEXT (2017), among teachers from the 34 countries and regions that participated in the Organization for Economic Co-operation and Development (OECD) Teaching and Learning International Survey conducted in 2013, Japanese teachers are the ones with the lengthiest working schedule and spend the longest time with other-than-teaching activities. Owing to these problems in Japanese schools, many teachers experience significant burnout. In addition, an increasing number of school teachers have been on a long-term sick leave from school because of both physical and emotional illnesses. According to the MEXT (2018), 5,077 teachers (0.55% of all teachers) were absent from work in 2017 because of illnesses, which was an increase of 186 from the previous year. Apart from the lengthiest working hours, one of the stressors faced by Japanese homeroom teachers are handling complaints from parents (Kasai, 2018; Onoda, 2008, 2015). With the rise of individual focus in Japanese educational settings, some parents claim their excessive rights at schools causing great difficulties for teachers in maintaining good relationships with such parents. Onoda (2008, 2015) used the term “monster parents” to refer to such parents who make excessive demands of schools, although his position is not to consider parents as monsters because most parents are not.

In this context, MEXT (2013) reported that, according to a survey conducted in 2012, among 3,911 teachers who responded to a consignment survey, the percentage of those who felt acute or chronic stress caused by their dealing with students’ behavioral problems was 68%, with 64% caused by dealing with paperwork, 62% by teaching, 60% by the quality of their work, and 57% by their dealing with difficult parents or by their communication with them. Kasai (2018) also conducted interviews with 21 teachers and found that one of their sources of psychological stress is the difficulty in interacting with parents. Nowadays, the focus on individual rights in education is increasing, and teachers, who are the center of school management, are being overloaded. Under such circumstances, it is important to examine what kind of teachers’ attitude and actions are effective in building home–school partnerships and what kind of attitude and actions are leading to their burn-out.

Although Japanese schools have historically made various practical efforts to collaborate with parents, research on home–school partnership is scarce. Some research has been published recently that emphasize the importance of collaboration between families and schools; however, those studies focus on students with needs such as those who are often absent (Mori, 2011; Tamura & Ishikuma, 2003) and children with special educational needs (Tamura & Ishikuma, 2007). These research studies focused on tertiary-level interventions in a Response to Intervention (RtI) model,² and home-school collaborations that focus on primary-level interventions have not been thoroughly examined. One such study was done by Kamimura and Ishikuma (2007) and examined the characteristics of parents’ meetings conducted by homeroom teachers. Their results showed that (1) parents’ needs are not always considered the most important, (2) parents tend not to have choices about who to talk to and when to stop coming to talk with teachers, and (3) what parents want to talk about is not always discussed during the meetings. However, this study only dealt with parents’ meetings conducted by home-room teachers, and other activities performed by teachers have yet to be researched in the Japanese setting.

Current study

This study defined home–school partnership as that upon which parents and teachers collaborate to support their children’s education to promote a positive school life. With this in mind, this study had two purposes: first, to develop scales that allow for the measuring of teachers’ individual differences regarding their attitudes toward home–school partnerships and the actions they take to build them; and, second, to examine the effect of teachers’ attitudes and actions toward these partnerships on their mental health status as measured by a burnout scale. Although this study was conducted solely in Japan, our findings have the potential to be shared with these countries that have similar educational settings (and potentially similar home–school partnership issues). Iida (2020) reviewed the journal articles published on the International Journal of School & Educational Psychology (IJSEP) from 2017 to 2019 (up until volume 3), and found out that some countries, especially Asian countries have some similarities with Japanese educational system in terms of limited availability of mental health professionals (e.g., China in Salinger, 2019; South Korea in Phosaly et al., 2019), large classroom size (India in Patwa et al., 2019), having difficulty collaborating with parents (Malaysia in Kok & Low, 2017).

Methods

Participants

Participants were recruited through reaching out to educational boards and school principals that one of the authors has connections within three different Japanese prefectures: one metropolitan, one suburban, and one rural. Thus, the method of sampling is non-probability sampling. We did not have any specific selection criteria for participants. Two of the prefectures were located in the suburban area of Kanto region (i.e., Central Japan), and the other was located in rural areas of Western Japan.

In total, 254 teachers [126 males (50%), 122 females (48%), and six did not specify (2.3%)] completed the questionnaires (Table 1). Regarding their ages, 38 participants (15%) were in their 20s, 65 in their 30s (26%), 58 in their 40s (23%), 83 in their 50s (33%), four in their 60s(1.5%), and six participants did not specify(2.3%). Regarding school type, 141 teachers worked at primary schools (56%), 94 at middle schools (37%), one at a high school (0.3%), nine at special education schools (3.5%), two at other types of school (0.8%), and seven did not specify (2.8%).

Table 1. Study sample

	N	%
N	254
Gender
Males	126	50
Females	122	48
Not Specified	6	2.3
Ages
20s	38	15
30s	65	26
40s	58	23
50s	83	33
60s	4	1.5
Not Specified	6	2.3
School Type
Primary School	141	56
Middle School	94	37
High School	1	0.3
Special Education School	9	3.5
Other types of School	2	0.8
Not Specified	7	2.8

Measures

Partnership Attitude Scale (PAT)

This scale was developed through a pilot study (Iida et al., 2014). The pilot study was completed by 92 teachers [45 males (49%), 46 females (50%), and one did not specify (1%)] who worked for several elementary schools, middle schools, high schools, special education schools, and private schools, teaching from 7 to 12 grades in Japan.

To create this scale, the researchers utilized one open-ended question: “What do you think about when you hear home–school partnership?” Three professors in the field of school psychology and two school counselors analyzed the collected data. After all answers were entered in a computer by a graduate student, a coding procedure was employed and 131 segments for the construct of home-school partnership were obtained. Following the procedures of the constant-comparison method (Glaser & Strauss, 1967; Merriam, 1998), segments with similar meanings were grouped. Moreover, the present study applied the KJ method (Kawakita, 1970) during the process of constant-comparison, which is a method commonly used in Japan owing to its usefulness. Through this hands-on process, the researchers achieved a clearer understanding of the outcomes of the constant-comparison process. Each segment was put into a paper slip, and paper slips with similar meanings were placed spatially close. After 131 paper slips were created for the 131 segments, the researchers found 30 main themes/categories that had similarities.

Based on these categories, 30 items were created on the basis of one for each theme/category and used to measure teachers’ attitude levels toward home–school partnership. A four-point, Likert-type scale was used, requiring respondents to choose either “Strongly disagree” (scored 1 point), “Disagree” (2 points), “Agree” (3 points), or “Strongly agree” (4 points).

Partnership Action Scale (PAC)

In the same aforementioned pilot study (Iida et al., 2014), to create the scale, three open-ended questions about teachers’ actions toward building, maintaining, and improving the home–school relationship were used: “what do you do to build a good relationship with parents?,” “what do you do to maintain a good relationship with parents?,” and “what do you do to improve your relationship with parents?”

The collected data were analyzed through the same procedures as those used in the PAT. After 169, 123, and 133 segments were delineated for building, maintaining, and improving the home–school relationship, respectively, the researchers found 12 main themes for teachers’ partnership actions: contacting, sending information, conference, strength focus, basic social skills, two-way communication, school events, child-focused, respect and showing empathy, class-wide parents meeting, advanced social skills, and networking.

In total, 37 items were created, with two to four items for each of the 12 themes, and they were used to measure teachers’ partnership action levels toward home–school partnership. The same 4-point, Likert-type scale used in the PAT was applied in the PAC.

The Beliefs in Cooperation Scale

The Beliefs in Cooperation scale (BCS) (Nagahama et al., 2009; Yasunaga, 2006) was used to examine the validity of the PAC. BCS was chosen because it measures the attitude toward collaboration in general. The scores of the 3 subscales of BCS (usefulness of cooperation, individual orientation, and inequity) showed correlation with discussion skills, discussion image, and university adaptation, supporting concurrent validity and the α for the 3 factors were .64–83, thereby supporting reliability on Nagahama et al. (2009). The BCS is a 5-point, Likert-type scale (1 = Strongly disagree, 2 = Disagree, 3 = Neither agree nor disagree, 4 = Agree, and 5 = Strongly agree) composed of 18 items, with three factors: usefulness of cooperation (e.g., “I feel I can do a lot of work when I work with people;” nine items), individual orientation (e.g., “When working with other people, I cannot do things my own way;” six items), and inequity (e.g., “We work together to help others who cannot perform the job by themselves;” three items).

Items in the “usefulness of cooperation” and “individual orientation” factors were expected to relate to home-school partnership, so they were used in this study. The other items were not included in our study to minimize participants’ burden.

Social Skills Self-Rating Scale for Adults

The Social Skills Self-Rating Scale for Adults (SSA) (Aikawa & Fujita, 2005) was administered to examine the validity of the PAT. SSA was chosen because the framework of social skills (building, maintaining, and improving relationships) was used in the pilot study, and we expected that the PAT would closely relate to certain aspects of this social skills scale. The total SSA score and the 6 subscale (relationship initiation, decoding, self-assertiveness, emotional control, relationship maintenance, and encoding) scores correlated with loneliness, anxiety, and depression, supporting concurrent validity and α for the six factors were .68–89, thereby, supporting reliability on Aikawa and Fujita (2005). SSA is a 4-point, Likert-type scale composed of 35 items, with the aforementioned six factors. The items in the “relationship initiation” (e.g., “I can easily befriend others;” eight items), “self-assertiveness” (e.g., “I share my opinions often;” seven items), and “relationship maintenance” (“I have respect for others;” four items) factors were employed in the present study. The “self-assertiveness” factor was chosen as related to “improving relationships” since self-assertiveness is a way of communicating one’s own opinions while respecting others. Thus, it is considered important for improving relationships.

Japanese version of the Maslach Burnout Inventory for teachers

The Japanese version of the Maslach Burnout Inventory (JMBI, Kubo & Tao, 1994) for teachers (Tamura & Ishikuma, 2001) was used to measure teachers’ burnout levels. JMBI was chosen because it has been repeatedly used to measure teachers’ burnout levels as an index of mental health status in Japan (Kubo, 2014; Tamura & Ishikuma, 2001) and considered to be an appropriate index to examine the effects of certain attitudes and actions on teachers’ mental health statuses. The factor analysis revealed same factor structures, supporting factorial validity and α for the three factors (Decreased Sense of Personal Accomplishment (PA), Depersonalization (DP), and Emotional Exhaustion (EE)) were .73–86, thereby, supporting reliability as in Tamura and Ishikuma (2001). It is a 5-point, Likert-type scale that was originally developed by Maslach and Jackson (1981). The JMBI is composed of 17 items, with three factors: PA (e.g., “I am filled up with my job” [a reversed item]; six items), DP (e.g., “I sometimes do not feel like talking to my colleagues as well as my students;” eight items), and EE (e.g., “I sometimes feel exhausted, both physically and emotionally;” three items).

Procedures

The University Institutional Review Board (IRB) approval was obtained prior to data collection. Permissions for data collection were also obtained from the municipal board of education and school principals. Cover letters provided to all participants explained the purpose of the study, the confidential safeguards that would be taken, and the voluntary nature of participation. To ensure that the study followed ethical guidelines, all questionnaires were answered anonymously and were given out to participants in a sealed envelope through either the municipal board of education or the school counselors working in the school where the questionnaires were administered and collected. In addition, those who consented to participate in the research completed the questionnaires in the following order: demographic questionnaire, PAC, PAT, BCS, SSA, and JMBI. All questionnaires were completed individually.

Data analyses

The IBM Statistical Package for Social Sciences (SPSS) Version 25 and IBM Amos Version 26 were used for data analysis. First, exploratory factor analyses (EFA) were conducted on the PAC and the PAT because they were under development during this study. After conducting EFA, confirmatory factor analyses (CFA) were conducted to examine the model fit of the structures. Internal consistency coefficients were obtained for each scale as well as each of their factors. In terms of the convergent validity, bivariate correlational analyses were conducted between the factor scores of the PAC and the BCS, and between the factor scores of the PAT and the SSA. Lastly, simultaneous multiple regression analyses were conducted with the PAC and the PAT as independent variables and each of the aspects of the teachers’ burnout (PA, DP, and EE) as dependent variables.

Results

Factor analysis of the PAT

An EFA using the principal component method with promax rotation was performed on the scores of the 30 items of the PAT. An eigenvalue greater than 1.0 and the Scree plots indicated a four-factor solution and it was considered the most appropriate one. Results showed that eight items were excluded because they did not have a salient factor loading (factor loading greater than .30) or they loaded to more than one factor almost equally (some of the excluded items were “if parents and schools use the same language and teach children in the same way, it will result in positive outcomes;” and “it is important that schools initiate the communication with parents”).

Another EFA using the principal component method with promax rotation was performed on the remaining 22 items of the PAT, and the four-factor model (the scree plot results were 6.23, 1.98, 1.62, 1.35, .99, .91…) explained 40.72% of the variance. The results of the EFA appear in Table 2. The first factor included items such as “collaboration with parents is necessary to educate children” and “it is important to sincerely try to hear parents’ thoughts,” so it was named “Appreciation” (α = .83). The second factor included items such as “class-wide parent-teacher meetings are meaningful” and “I learn various ways of thinking through communicating with parents,” so it was named “Positive Attitude” (α = .76). The third factor included items such as “parents need to understand the school educational activities to fit their children into the school goal” and “it is essential to gain cooperation from parents to fit the children into the school goal,” so it was named “Traditional Attitude” (α = .68). The fourth factor included items such as “I feel difficulty when dealing with some parents” and “it is difficult to accept various parents’ thoughts and wishes,” so it was named “Difficulty” (α = .69).

Table 2. Factor analysis of the partnership attitude scale (PAT) (promax rotation) (n= 242)

Items		I	II	III	IV	I-T correlation	M	SD	h^2
Factor 1: Appreciation (α = .83)
3	Collaboration with parents is essential to educate their children.	.75	−.11	.00	−.16	.43	3.88	.34	.47
12	It is important to empathetically listen to the parents.	.69	.11	−.10	.04	.57	3.73	.45	.50
7	It is important to communicate with parents and build a trustful relationship with them.	.68	−.20	.04	.05	.41	3.84	.39	.39
9	It is important that parents and schools understand each other and support children’s growth on their own way.	.55	−.02	.21	−.03	.53	3.66	.53	.45
11	All parents have at least some issues or concerns about their children’s education.	.54	.05	−.23	.10	.34	3.67	.51	.26
15	Even when parents have different opinions and standpoints, it is necessary to collaborate and cooperate with them for the children’s future.	.53	.11	.03	.03	.53	3.74	.46	.39
17	It is important to communicate with parents to have a better underanding of their children.	.53	.11	.18	.04	.63	3.68	.51	.52
2	It is important to build a relationship with parents to support children’s growth.	.52	−.05	.09	−.10	.40	3.87	.34	.29
8	I want to share both positive and negative things about the children with their parents.	.45	.02	.22	.03	.55	3.54	.58	.38
Factor 2: Positive Attitude (α =.76)
25	Class-wide parent-teacher meetings are meaningful.	−.07	.67	.12	−.11	.48	2.87	.68	.46
20	I learn various ways of thinking through communicating with parents.	.27	.58	−.10	−.09	.56	3.29	.63	.50
26	I feel that most parents want to communicate with teachers.	−.05	.57	−.19	.13	.32	2.76	.65	.29
24	As I got older, I became more comfortable with talking to parents and started enjoying it.	.04	.57	−.14	−.05	.33	2.78	.77	.30
21	I believe teachers and parents are in an equal partnership.	−.20	.48	.23	−.11	.37	3.06	.78	.25
28	Since few parents actually get involved in school activities, it is important to invite parents to more opportunities.	−.17	.48	.27	−.03	.41	2.45	.68	.29
14	When children often talk about their school to their parents, I feel it is easier to talk with their parents because there are common topics between us.	.11	.42	.08	.19	.54	3.33	.66	.34
10	I like to talk with parents to have an understanding of their children’s home environment.	.29	.38	−.01	.04	.51	3.02	.70	.35
Factor 3: Traditional Attitude (α = .68)
5	Parents need to understand school educational activities to fit children into the school goal.	.03	−.11	.75	.06	.46	3.59	.56	.56
4	It is essential to gain cooperation from parents to fit children into the school goal.	.31	−.01	.49	−.01	.58	3.75	.44	.49
16	I want parents to participate in school events.	.05	.13	.48	.08	.48	3.45	.58	.35
Factor 4: Difficulty (α =.69)
13	I experience difficulty when dealing with some parents.	−.07	.01	.05	.89	.22	3.50	.65	.79
6	It is difficult to accept various parents’ thoughts and wishes.	.02	−.09	.07	.57	.19	3.43	.61	.34
	Cumulative Proportion (%)	25.67	32.02	37.27	40.72
	Factor Correlations (r)		II	III	IV
		I	.54	.55	.19
		II		.36	.12
		III			.17

Factor loadings >.40 are in bold face.

Using the same sample, CFA was conducted to examine the model fit of the PAT structure. The results showed that the items’ loading was .42 to .73 for Appreciation, .42 to .69 for Positive Attitude, .53 to .76 for Traditional Attitude, and .72 to .74 for Difficulty. The model fit index was X² (203) = 408.438 (p< .000), CFI = .866, SRMR = .066, RMSEA = .063.

Exploratory factor analysis of the PAC

An EFA using principal component method with promax rotation was performed on the scores of the 37 items of the PAC. An eigenvalue greater than 1.0 and the interpretability suggested that the five-factor solution was the most appropriate one. Results showed that 12 items were excluded because they did not have a salient factor loading or they loaded to more than one factor almost equally (some of the excluded items were “I inform parents when children are late to school, absent from school, or leave early” and “I send parents useful information electronically, such as E-mails, in a timely manner”).

Another EFA using principal component method with promax rotation was performed on the remaining 25 items of the PAC, and the five-factor solution (the scree plot scores were 8.05, 2.05, 1.74, 1.33, 1.19, .96, .91…) explained 47.23% of variance. The results of the EFA appear in Table 3. The first factor included items such as “I respect parents’ wishes that their children will be successful and well-received at school” and “I invite parents to talk to teachers if something comes up,” so it was named “Respect Parents” (α = .86). The second factor included items such as “I try to share even minor things about the children with their parents” and “I often communicate with parents about their children with a correspondence notebook or by phone,” so it was named “Individual Contact” (α = .83). The third factor included items such as “I issue classroom newsletters about school events and children’s everyday lives” and “I share information through classroom newsletters or websites,” so it was named “Classroom Information Sharing” (α = .80). The fourth factor included items such as “I consider parent-involved school events very important;” and “when a student gets in trouble, I visit his/her home and talk to their parents in person,” so it was named “Being Proactive” (α = .59). The fifth factor included items such as “when I talk to parents, I try to have some chats and make jokes so that we can enjoy the conversation;” and “I let parents talk as much as possible during a parent meeting,” so it was named “Facilitating Conversation” (α = .67).

Table 3. Factor analysis of the partnership action scale (PAC)

Items		I	II	III	IV	V	I-T correlation	M	SD	h^2
Factor 1: Respect Parents (α = .86)
13	I respect parents’ wishes that their children will be successful and well-received at school.	.76	.01	.02	.12	−.10	.66	3.54	.57	.65
14	I invite parents to talk to the teachers if something comes up.	.71	.05	−.10	.09	−.03	.60	3.57	.62	.56
5	I listen to the parents.	.67	.13	.05	−.19	.00	.57	3.47	.64	.47
4	When I meet parents at school, I exchange greetings with them.	.67	−.10	.09	−.23	.09	.45	3.79	.42	.35
29	I try to think from the parents point of view.	.62	.04	−.12	.16	.03	.59	3.48	.60	.52
10	I value relationships with children because parents feel at ease when their children are lively.	.62	−.22	.03	.08	.07	.48	3.66	.51	.37
30	I inform parents that collaboration between the school and the parents is important.	.62	−.16	.03	.08	.13	.54	3.53	.65	.43
11	When talking to parents, I try to create a friendly atmosphere.	.61	.10	−.13	−.05	.05	.52	3.46	.56	.39
34	I try to obtain trust from the parents through children’s everyday behaviors and positive changes.	.34	−.04	.06	.19	.22	.56	3.49	.57	.38
Factor 2: Individual Contact (α = .83)
19	I try to share even minor things about children with their parents.	−.15	.82	−.11	.03	.04	.49	2.59	.78	.53
1	I often communicate with parents about their children through a correpondence notebook or by phone.	−.01	.76	.05	−.25	.08	.50	2.98	.82	.55
36	When the children perform good behaviors, I contact their parents and praise them in a concrete way.	−.19	.65	−.05	.11	.21	.51	2.98	.81	.46
17	I share information with parents about my students’ behaviors at home and at school on a regular basis.	.01	.56	.23	−.03	−.01	.54	2.75	.83	.45
15	I emphasize taking time to communicate with parents by having frequent meetings and taking extra time to perform such activities.	.15	.54	−.02	.13	−.31	.48	2.92	.80	.40
18	I contact parents by phone, through e-mail, or make a home visit according to the topics in need of talking.	.11	.44	−.02	.22	.00	.60	3.09	.77	.41
3	I inform parents about the positive aspects of children as well as about their efforts.	.25	.41	.19	−.14	.14	.63	3.28	.68	.51
Factor 3: Classroom Information Sharing (α = .80)
2	I issue classroom newsletters about school events and children’s everyday lives.	−.02	.01	.91	−.03	.04	.47	2.60	1.09	.84
16	I share information through classroom newsletters or websites.	−.09	−.02	.86	.11	−.02	.44	2.86	1.00	.73
35	I inform parents on the classroom policies and goals through the classroom newsletter.	.12	.06	.49	.19	−.13	.51	3.10	.82	.41
Factor 4: Being Proactive (α = .59)
24	I emphasize parent-involved school events.	−.02	−.11	.11	.61	.17	.48	3.27	.65	.44
23	When a student gets in trouble, I visit his/her home and talk to his/her parents in person.	−.05	−.07	.05	.57	.09	.38	2.98	.90	.31
8	I participate in events held by parental organizations and other PTA (parent-teacher association) activities.	.04	.13	.03	.51	−.16	.41	2.91	.78	.33
Factor 5: Facilitating Conversation (α = .67)
27	When I talk to parents, I try to have some chats and make jokes so that we can enjoy the conversation.	.12	.01	−.02	.01	.50	.37	3.04	.83	.32
25	I let parents talk as much as possible during parent meetings.	.04	.25	−.05	.25	.45	.61	3.27	.65	.53
37	I try to begin with positive comments about the children when communicating with their parents.	.28	.16	−.03	−.04	.43	.56	3.43	.66	.46
	Cumulative Proportion (%)	30.15	36.59	41.82	44.75	47.23
	Factor Correlations (r)		II	III	IV	V
		I	.59	.38	.60	.43
		II		.43	.41	.39
		III			.33	.34
		IV				.26

Factor loadings >.40 are in bold face.

Using the same sample, CFA was conducted to examine the model fit of the PAC structure. The result showed that the items’ loading was .53 to .79 for Respect Parents, .50 to .71 for Individual Contact, .57 to .92 for Classroom Information Sharing, .50 to .76 for Being Proactive, and .50 to .72 for Facilitating Conversation. The model fit index was X² (265) = 501.820 (p< .000), CFI = .881, SRMR = .067, RMSEA = .065.

Convergent validity of the PAT and the PAC

Before examining the relationship between concurrent scales and the PAT and the PAC, we examined the internal consistency for the existing scales: the BCS, the SSA, and the JMBI. For the BCS, the level of internal consistency was α = .84 for both Individual Orientation and Usefulness of Cooperation. For the SSA, the level of internal consistency was α = .78 for Relationship Initiation, α = .76 for Self-Assertiveness, and α = .67 for Relationship Maintenance. The results of bivariate correlation analyses between the PAT and the BCS subscale scores and between the PAC and the SSA appear in Table 4.

Table 4. Correlations between the partnership scales and the other scales

		Attitude Toward Cooperation
		Individual Orientation	Usefulness of Cooperation
Partnership Attitude Scale (PAT)
	Appreciation	−.22***	.41***
	Positive Attitude	−.13*	.20**
	Traditional Attitude	−.12+	.30***
	Difficulty	.20***	.08
		Social Skills
		Relationship Initiation Relationship	Assertiveness	Relationship Maintenance
Partnership Action Scale (PAC)
	Respect Parents	.24***	.08	.39***
	Individual Contact	.20**	.15*	.19**
	Classroom Information Sharing	.18**	.14*	.22***
	Being Proactive	.12+	.14*	.14*
	Facilitating Conversation	.37***	.06	.24***

^†p< .10, * p< .05, ** p< .01, *** p< .001.

As shown in a reference (Hattori & Kaiho, 1996), the level of correlation was seen as low if r = .20–.40, moderate if r = .40–.70, and high if r > .70. Regarding the PAT and the BCS correlation, Individual Orientation was found to have a low negative correlation with Appreciation (r = −.22, p < .001) and a low positive correlation with Difficulty (r= .20, p < .001); Usefulness of Cooperation was found to have a moderate positive correlation with Appreciation (r = .41, p< .001) and a low positive correlation with Positive Attitude (r = .20, p < .01) and Traditional Attitude (r = .30, p < .001).

Regarding the PAC and the SSA correlation, Relationship Initiation was found to have low positive correlation with Respect Parents (r = .24, p < .001), Individual Contact (r = .20, p< .01), and Facilitating Conversation (r = .37, p < .001). Relationship Maintenance was found to have a low positive correlation with Respect Parents (r = .39, p < .001), Classroom Information Sharing (r= .22, p < .001), and Facilitating Conversation (r = .24, p < .001).

The effects of the PAT and the PAC on teachers’ burnout

An EFA using principal component method with promax rotation was performed on the scores of the JMBI 17 items.³ Two items were omitted because one item showed a low factor loading (below .40) and the other one showed high factor loadings for more than two factors (above .40). The final result of the EFA showed a three-factor structure, showing the same structure as indicated in Kubo (2014): Personal Accomplishment (PA, five items), Depersonalization (DP, seven items), and Emotional Exhaustion (EE, three items). The Cronbach’s alphas were .81 for PA, .80 for DP, and .73 for EE, thereby showing enough internal consistency.

Results of the multiple regressions appear in Table 5. The regression analysis revealed significance for three factors of burnout (R² = .14 ~ .25, p< .001). For PA, the Traditional Attitude (β = .25, p< .01) and Respect Parents (β = −.21, p< .10) factors showed significant effects. The items in PA were all reversed items; therefore, high scores of this subscale indicates the opposite condition indicated by the items, such as “I am so occupied with the work that I forget myself,” “I enjoy my work, and sometimes I run out of time without knowing it” so these results indicate that Traditional Attitude may lower teachers’ PA, whereas Respect Parents may increase their sense of PA, thereby functioning as a protective factor against PA.

Table 5. Correlation and regression analysis for teachers’ burnout

		Burnout
		Personal Accomplishment (PA)		Depersonalization (DP)		Emotional Exhaustion (EE)
		r	B	r	B	r	B
Partnership Attitude (PAT)
	Appreciation	−.12+	−.02	−.12†	−.06	.04	.06
	Positive Attitude	−.19**	−.10	−.17*	−.04	−.11†	−.11
	Traditional Attitude	.02	.25**	−.02	.06	.16*	.16*
	Difficulty	.04	.05	.28***	.32***	.36***	.41***
Partnership Action (PAC)
	Respect Parents	−.27***	−.21+	−.20**	−.04	−.07	.01
	Individual Contact	−.17**	.03	−.08	.02	−.07	−.01
	Classroom Information Sharing	−.19**	−.11	−.10	.03	−.12†	−.03
	Being Proactive	−.20**	−.08	−.24***	−.21**	−.14*	−.16*
	Facilitating Conversation	−.24***	−.08	−.17**	−.14†	−.14*	−.17*
	Total variance explained (R2)		.14***		.17***		.25***

^†p < .10, * p< .05, ** p < .01, *** p<.001.

For DP, Difficulty (β = .32, p< .001), Being Proactive (β = −.21, p< .01), Facilitating Conversation (β = −.14, p< .10) showed significant effects. This result indicates that having difficulty collaborating with parents may be a risk factor for teachers’ DP in one hand, but building a good relationship with parents through actions that relate to Being Proactive and Facilitating Conversation factors may work as protective factors against DP on the other hand.

For EE, Traditional Attitude (β = .16, p< .05), Difficulty (β = .41, p< .001), Being Proactive (β = −.16, p< .05), and Facilitating Conversation (β = −.17, p< .05) showed significant effects. The results indicated that, while Traditional Attitude and Difficulty may increase teachers’ EE, building a good relationship with parents through actions that relate to Being Proactive and Facilitating Conversation factors may work as protective factors against EE.

Discussion

To fulfill the first aim of this study, the PAT and the PAC scales were developed through a pilot study and the present study. The PAT scale consisted of four factors: Appreciation, Positive Attitude, Traditional Attitude, and Difficulty. Appreciation and Positive Attitudes factors seemed to be similar to each other; however, no item showed a high load to both factors, indicating that teachers’ appreciation of their collaboration with parents not always overlap with their positive attitude toward that collaboration. Moreover, Appreciation seems to be more of a cognitive aspect of teachers’ partnership attitude, meaning that, even if teachers believe that collaboration is important, it does not automatically create a positive attitude toward it.

The Traditional Attitude factor in the present study was similar to the concept of traditional partnership in a previous study (Christenson & Sheridan, 2001). In traditional partnership, schools ask parents to help their child to achieve a school goal, and teachers are responsible for teaching and providing guidance, with clear-cut differentiation between the roles of teachers and parents. Lastly, the Difficulty factor was also extracted, which relates to teachers’ difficulty to collaborate with parents. Based on prior research, this was expected since many current teachers are having difficulty dealing with parents (Kasai, 2018; Onoda, 2008, 2015). The four factors were extracted from items developed through an open-ended questionnaire that was applied to Japanese teachers, so they are considered to have content validity themselves.

Reliability coefficients for all four factors were above .60, thereby showing moderate levels of internal consistency. Each of the factors of the PAT correlated with at least one aspect of the BCS, as expected (and explored in the Methods section), therefore the concurrent validity of the PAT was supported. In sum, the PAT can be used as a tool to understand teachers’ individual differences on their attitudes toward home-school collaboration.

The PAC consisted of five factors: Respect Parents, Individual Contact, Classroom Information Sharing, Being Proactive, and Facilitating Conversation. The Respect Parents factor included items that related to teachers’ actions that showed they empathized with the parents. The Individual Contact factor included items that related to building parents’ trust through the creation of opportunities to meet in person, either through school events and/or parent-teacher meetings. The Classroom Information Sharing factor related to teachers’ actions regarding classroom management, mainly because Japanese teachers put much effort into this specific task, so they often issue classroom newsletters to students and, sometimes, to parents.

The Being Proactive factor related to teachers’ actions aimed at displaying honesty and earnestness toward parents. In past literature on parental trust in schools (Forsyth et al., 2002), benevolence and reliability were indicated as two factors that lead to parental trust of teachers. Thus, our PAC scale having a Being Proactive factor was consistent with the findings of previous studies. The Facilitating Conversation factor included items related to teachers’ actions toward promoting an effective communication with parents and the items indicated that teachers go through a lot of work to use various ideas from their past experiences that may help build their relationship with parents since home-school meetings are rarely conducted upon parents’ requests (Kamimura, 2014).

The model fit of the factor structures of PAT and PAC indicated moderate fit since the SRMR and RMSEA met the criteria with respect to Hu and Bentler’s guidelines (CFI ≧ .90, SRMR ≦ .08, and RMSEA ≦ .08). The reliability coefficients for these factors showed an acceptable level for Cronbach’s alphas, except for the Being Proactive factor (α = .59). This happened because the three items in the Being Proactive factor dealt with different situations within schools, so we believe that the internal consistency of this subscale has to be improved in the future. As expected, each factor of the PAC correlated with at least one aspect of the SSA; therefore, the concurrent validity of the PAC was supported. In sum, the validity and reliability of the PAC were supported to a certain level and the scale can be used to assess teachers’ actions toward home–school partnership.

The effect of the PAT and the PAC on teachers’ burnout

Our findings showed that the PAT and the PAC had effects on teachers’ burnout. First, Traditional Attitude had a positive effect and Respect Parents had a negative effect on teachers’ decreased sense of Personal Accomplishment (PA). In Japan, an increasing number of students with diverse family structures and cultural backgrounds go to school (MEXT, 2016), and teachers with traditional attitudes tend to have strong beliefs about what should be done to gain parents’ cooperation to fit their children into the school goals. Thus, when these strong beliefs are not met by the families, their sense of PA will likely be lowered. Contrastingly, teachers showing higher levels of respect toward parents’ opinions may have a higher sense of PA, mainly because, although this may increase their overload, they may tend to listen to parents’ wishes and invite them to talk in the school environment, and if this collaboration is successful and results in practical improvement for the children, they may experience higher levels of PA.

Second, the Difficulty factor showed positive effects and the Being Proactive factor showed negative effects on Depersonalization (DP). These results indicated that dealing with difficult parents may contribute to teacher’s burnout, but being conscientious (e.g., making efforts to talk in person and build a collaborative relationship with the parents) may prevent teachers’ from performing dehumanized behaviors.

Lastly, teachers’ Emotional Exhaustion (EE) was positively related to the Traditional Attitude and the Difficulty factors, while the Being Proactive and the Facilitating Conversation factors were negatively related to it. Like some aforementioned effects of the PAC/PAT on specific aspects of teachers’ burnout, having a traditional attitude and having difficulty toward dealing with parents were shown to be risk factors for EE, whereas being proactive toward dealing with parents and facilitating conversation with them were shown to be protective factors of EE.

In burnout research, problems related to “emotional labor,” which refers to the service workers such as nurses, teachers, and public service workers having to regulate their emotions according to social norms and job demands more often than others have been pointed out (Hochschild, 1983; Kubo, 2014). Further, while some teachers may face difficulty to collaborate with parents, they are still expected by many stakeholders to have positive attitudes toward parents to support children’s success in school. Such expectations may cause an emotional conflict and lead to EE. Therefore, to decrease their burden, teachers themselves should be supported by their colleagues and principles and mental health professionals who could help promote their well-being.

Implications

The aspects of the PAT and the PAC that were clarified in our study will be useful to general Japanese schools in their daily practices. In Japan, parent-teacher meetings, home visits, and individual parent-teacher meetings are usually conducted in most schools as efforts to promote home–school partnership. However, there are substantial individual differences, mainly because teachers are, usually, the ones responsible to decide on the amount of efforts that are actually made toward the promotion of such collaboration. Considering the situation Japanese school teachers are facing today, evaluating what is really effective in building home–school partnerships and what is not is recommended. For example, the parent meetings are usually conducted once per semester by each grade teachers. Many parents participate in such meetings when their children are at the first grades of the elementary school, but the number of participants decreases as their children grow older. Parents of upper grades and junior high school students are often troubled by problems such as dealing with early adolescence and their use of the Internet, but psychoeducational program on these issues toward parents are seldom carried out. To make parents meetings more effective, such content can be provided at schools. These programs do not necessarily have to be provided by teachers only. They can be complemented by school counselors who are assigned to each school to offer counseling and guidance. With multidisciplinary teams, schools could gain trust from the parents (Forsyth et al., 2002).

Second, it was indicated that some factors of the PAT and the PAC may be risk factors for specific teachers’ burnout symptoms. Particularly, teachers’ traditional attitudes seemed to be risk factors for their burnout. Teachers having a traditional attitude expect all families to look out for their children’s homework and will appropriately discipline them. Although this type of family involvement in the children’s education is important, some families are really struggling to support their households financially and are not able to thoroughly contribute to their children’s education. The Cabinet Office (2015) revealed that the rate of children living in relative poverty⁴ was 16.3%, that is, one in six children in 2012, and the rate was even higher for single-income families, which was 54.6%. Moreover, the number of child abuse incidents reported to the Child Consultation Center have drastically increased in recent years (Ministry of Health, Labour and Welfare, 2018). Thus, this creates a huge gap between the families’ needs and teachers’ expectations. In that sense, teachers’ traditional attitudes seem to not work effectively in the scenario where families have diverse backgrounds, which may then lower the teachers sense of PA and increase their EE.

Further, Difficulty also seemed to be a risk factor for teachers’ burnout. Previous literature shows that some parents are really difficult to deal with (Kasai, 2018; Onoda, 2008, 2015) and when teachers experience difficulties when dealing with specific parents, school principals, other experienced teachers, coordinators for children with special needs, and colleagues, along with mental health professionals (i.e., school counselors and school social workers) should provide systematic support to prevent teachers’ burnout and turnover.

Third, what was also interesting to see is that the Respect Parents, Being Proactive, and Facilitating Conversation factors worked as protective factors for teachers’ burnout. As abovementioned, teachers who frequently engage in actions that promote home–school partnership might experience higher workloads, but their burnout level is lower. This result implies that, if collaboration works out well and teachers are able to see children’s improvements in school, their work can be rewarding, and their motivation might increase. Further, Adams and Christenson (2000) showed that parents’ trust in teachers is higher than teachers’ trust in parents, and our findings demonstrated that, if teachers trust parents and take actions toward establishing appropriate partnerships, it is likely that they will be able to establish good home–school relationships, thereby producing good effects in children’s school success.

Limitations and recommendation for future research

Although we presented some candid findings and reflections, this study has some limitations. First, some factors of the PAT and of the PAC showed low internal consistencies in the reliability analyses, so it is necessary to further investigate reliability and validity of the constructs. Second, the sample size was modest and limited to Japanese teachers working in public elementary schools and middle schools. Therefore, the generalization of the results may be somewhat limited. Future research should utilize a sample with teachers in different types of schools because different schools have different types of parents and different type of home–school relationships. Moreover, private middle schools are starting to become popular in Japan, especially in urban areas, with parents paying extra tuitions to send their children to such schools, so some difficulties may arise in home–school relationships within such a setting. Finally, teachers in many countries have similar concerns regarding the proper establishment of good relationships with parents, and they also work with limited support from mental health professionals to deal with difficult parents. Thus, future research should investigate the similarities and differences among different countries in this regard and evaluate what works and what does not work to help us advance research on home–school partnership.

Acknowledgments

Part of this research was presented at the ISPA 39th conference in Amsterdam.

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability

The data that support the findings of this study are available on request from the corresponding author, (IJ).

Additional information

Funding

This work was supported by the Japan Society for the Promotion of Science [KAKENHI grant Foundation (C) Grant Number 17K04341 and (B) Grant Number 20H01758].

Notes on contributors

Junko Iida

Junko Iida Ph.D. is an Associate Professor in the Department of Human Sciences, University of Tsukuba, and a researcher and practitioner in school psychology.

Naoko Shimada

Naoko Shimada Ph.D. is a Researcher in the Psychological Institute, Rissho University, and a researcher and practitioner in school psychology.

Saori Yamasaki

Saori Yamasaki works for the Board of Education in Tottori Prefecture, Tottori, Japan, and works as a school counselor and a supervisor for school counselors.

Notes

¹ A Japanese school year starts in April and ends in March. The school year is usually divided into three terms: The First term begins in April and ends in July, the Second term begins in September and ends in December, and the Third term begins in January and ends in March. Nowadays, some schools have two semesters instead of three semesters to reduce teachers’ burden regarding their grading work and increase the time spent on students’ learning activities instead of testing.

² Since school counselors are part-time positions in Japan, they are not at the school site most of the time, so the roles of student guidance and counseling is usually assigned to teachers who are in the committee of guidance and counseling division each year. Thus, usually, each teacher is assigned many kinds of school roles other than that of teaching, such as a guidance position, a counseling position, a special educational needs position, a research promoting position, a PTA position, and a health promoting position, just to name a few.

³ Nevertheless, in the abovementioned research that dealt with teachers’ burnout in Japan (Tamura & Ishikuma, 2001), two items were omitted from the original scale (Kubo & Tao, 1994), and it has been pointed out that the factor analysis of this burnout scale often yielded different results (Kubo, 2014). Thus, since we used the original scale with the 17 items, we felt unsure about its factorial structure, so we conducted an exploratory factor analysis (EFA) for the teachers’ burnout scale (shown in the Results).

⁴ Relative poverty rate means the percentage of people who earn less than half of the middle rank (median number) of income for a regular job in Japan.