Trust is multifaceted and its meaning varies from individuals and groups (Forsyth et al., 2011). It is founded on function, ownership, shared expectations, and relationships. Home–school trust in particular is projected as an ultimate concern for school organizations positioned to help students learn (Goddard et al., 2001). Faculty trust in parents and students is a collective school property in the same manner as collective efficacy and academic prominence (Hoy et al., 2006a). Adams and Forsyth (2013) affirm that trust is a normative property of school groups established from shared perceptions of openness, honesty, benevolence, reliability, and competence. Eng et al. (2014) add that trust is the confidence in investing in education that motivates involvement in children’s education. It is built on the confidence between trustor and trustee through communication (Li et al., 2016) and shared knowledge centered on the understanding of present activities and previously established responsibilities between the two parties (Borawski et al., 2002).

Focusing on the mutuality of trust, Hoy et al. (2006b) posit that faculty trust is a reciprocal relationship in which teachers and parents trust each other to consistently act in students’ best interests. It also includes reciprocal relationships among colleagues, principals, students, and parents. Bower et al. (2011) contend that faculty trust is the lubricant that ensures continued parental commitment and relational engagement in the school. Trust makes a parent feel confident that their child’s teacher is acting in a way that will benefit the parent–teacher relationship or a similar goal such as students’ academic success. On the other hand, trust causes teachers to believe that their colleagues, parents, and the principal are doing their best to achieve the shared educational goals of the students (Adams and Christenson, 2000), and this trust can be measured through the context of the parent–teacher relationships (Houri et al., 2019). It is, therefore, safe to argue that trust is built on the confidence placed upon another person to act in a manner that will benefit either the relationship or a similar goal, and could be facilitated by relationship factors such as commitment (Simpson, 2007).

In a trusting relationship, there is a degree of dependability among parties. Liu et al. (2016a) argue that trust is a mental condition involving the acceptance of vulnerability built on the expectations of favorable outcomes from others. It has been asserted that trust involves confidence that expectations will be met (Van Houtte, 2006). This means that there is a tone of dependability between trust referents. It also indicates that parents, teachers, principals, and students believe that the other party will be responsible enough to play their roles (Titrek, 2016). Thus, faculty trust is a collective form of trust in which the faculty has an expectancy that the word, promise, and actions of another group or individual can be relied upon and that the trusted party will act in the best interests of the faculty (Forsyth et al., 2011). It is the confidence that another person will act in a way to benefit or sustain the relationship or the implicit or explicit goals of the relationship to achieve positive outcomes for students (Kikas et al., 2011).

Since the focus of our paper was home–school trust, only studies that examined trust between home and school at any of the K-12 levels were included. Articles that investigated trust within post-secondary school contexts were excluded.

Additionally, only quantitative and mixed methods studies were included. Pure qualitative studies were excluded because we intended to examine how trust has been measured (statistically) as well as determine how potential gaps in its measurement may signal gaps in its conceptualization. To effectively do this, a careful reflection on the commonly used trust scales was necessary. It would have been impossible to understand the degree of trust measurement through qualitative themes, considering the diversity and lack of homogeneity among interview guides and findings. In the case of mixed-methods studies, we only examined the quantitative component which statistically computed trust. Although the levels of trust may vary from one grade level to another (Adams and Christenson, 2000), this paper did not focus on specific levels. The idea was to understand how researchers compute and present levels of trust (whether high or low) regardless of school or grade levels.

Only peer-reviewed journal articles were included. We limited our review to only journal articles as a way to narrow the scope while ensuring the quality of the literature. Peer-reviewed journals are generally believed to provide high-quality articles. The reviewed literature covers empirical studies from 2000 to date. Gray research, theses, conference papers, and unpublished materials were excluded, as it was difficult to ascertain their quality and authenticity.

We conducted a comprehensive literature search to locate studies related to home–school trust was carried out on three databases such as (1) SCOPUS, (2) EBSCO (ASC/BSC/ERC), and (3) Web of Science since they encompass educational and psychology journals where the studied construct is originated. The search was guided by the following terms; “trust and teachers,” “trust in schools,” “trust in parents,” “trust in a family school partnership,” “parents’ trust in schools,” “parent–teacher trust,” “family–school trust,” “students’ trust in teachers,” “faculty trust,” “trust and school achievement,” and “trust and school culture.” Articles in languages other than English were excluded. Furthermore, the references and bibliographies of the searched studies were screened for other related papers. Those related papers were additionally tracked in the databases.

The literature search to locate relevant studies was based on refinement of the search results in terms of (1) time (January 2000–May 2020); (2) language (English); (3) publication type (Journal articles); and (4) field (social sciences and psychology). A total of 3,552 non-duplicated titles were obtained. Both authors, separately, scanned through and reviewed the titles using the pre-established exclusion criteria (above). This process resulted in the elimination of 2,229 titles. The reasons behind the elimination are displayed in Figure 1.

The same process and criteria were adhered to by both researchers during the abstract review stage, where 522 articles were retained for full-text review. Finally, articles that reached the full-text review stage were screened according to the following criteria (a) full text in English (some abstracts were in English but full text in other languages); (b) trust as one of the variables and is measured statistically; (c) context (trust has to be examined within or between school and/or home); and (d) K-12 study sample. This process finally yielded 79 articles that were then reviewed in line with the study’s purposes. All inconsistencies were resolved during this process through constant review and discussion until consensus was reached.

Following the final selection of relevant literature, the authors evaluated the articles based on the following: (a) author information and year of publication, (b) provision of definition, and (c) components of definition (see Supplementary Table 1). We then summarized and tabulated home–school trust common elements across definitions by grouping them based on the theoretical models (i.e., process, state, and relationship roles) as indicated in Figure 2. Through these groups, we were able to capture a multidimensional conceptualization and nature of trust. In examining the major trust referents and their relationships, we categorized them according to the trust-flow structure and relationship. Home–school trust referents were distinguished based on three attributes: trust from home, trust from school, and trust within home and school. Trust from home reflected trust extended by the family members (parents and students). Trust from school, on the other hand, entailed trust extended by school members (teachers and principals). Trust within home and school is the trust within-family participants and/or within-school participants (for example, faculty trust in principals/colleagues and parent trust in students).

For the review’s measurement component, all scales were assessed and coded using categories empirically derived from the examination of scale items and their psychometric properties (see Supplementary Table 2). The conceptualization themes underpinning the analysis process provide clarity regarding the relationship between trust measurement and conceptualization. Armed with this framework, we then used inductive thematic analysis (Creswell, 2009) to generate categories that depict the conceptualization and measurement of trust in trust research. The data are narratively presented and organized based on the research questions.

Fifty-one studies (64.6%) examined trust as a unidimensional construct comprising (1) Trust-within-school [faculty’s trust in principals (12), faculty’s trust in colleagues (9); and principals’ trust in teachers (2)]; (2) Trust-from-school [faculty’s trust in clients – students and teachers as a combined unit – (5), faculty’s trust in students (3), faculty’s trust in schools (1), and faculty’s trust in parents (1)]; (3) Trust-from-home [students’ trust in teachers (9), parents’ trust in teachers (5), and parents’ trust in schools (1)]; and (4) Trust-within-home [parents’ trust in students (2), and students’ trust in their peers (1)].

Also, 23 articles (29.1%) identified trust as a multidimensional construct involving multiple trust referents, for example, faculty’s trust in principals, colleagues, and clients (17), students’ and parents’ trust in principals (1), parents’ trust in teachers and school (1), faculty’s trust in principals and colleagues (2), parents’ trust in principal and school (1), and student’s trust in teachers and colleagues (1). The reciprocity of trust between home and school was examined in five studies (6.3%). Of the 79 reviewed articles, 65 (82%) provided a conceptual view of trust, 8 (11%) did not offer any, while 5 (7%) were not clear (see Supplementary Table 1). The resulting themes in the conceptualization of trust can be categorized as (1) trust as a process; (2) trust as a state; (3) trust as relationship roles; and (4) shared goals/expectations (see Figure 2).

To fully understand the operationalization of home–school trust, it was also crucial to cast light on its measurement and examine the relationship between trust conceptualization and trust measurement. This review found a total of 32 scales purporting to measure trust or some aspects of it. Six scales were, however, dropped from the analysis for lack of clear dimensional focus. Therefore, only 26 measures were analyzed. Supplementary Table 2 shows the 26 scales in the final sample alongside the 6 that were excluded (at the bottom of the table). The table lists all the studies associated with these scales and other related coding details.

More than half of the studies provided trust definitions, none of which was unanimously accepted as a definition of home–school trust. Nonetheless, the following essential elements can be gleaned from the various conceptualizations of trust: (1) it involves the risk of vulnerability on the part of the trustor; (2) it is based on the trustor’s confidence in the trustee; (3) the trustee’s trustworthiness characteristics are critical in the decision to trust; (4) it occurs in the course of a relationship between the trustor and trustee; and (5) it takes place within a context of expected outcomes or shared goals. The latter part involves specific roles or tasks where behavior can be taped onto. Also, some definitions relate it to social exchanges between the trustor and the trustee.

From the reviewed articles, we established that trust has been conceptualized under four pillars namely, the trustor’s role, the trustee’s role, shared goals, and relationships among trust referents. Romero (2015) affirms that trust includes a trustee and a trustor, who undertake a certain crucial “role in settings involving vulnerability; where confidence in another’s goodwill and expertise is important” (p. 217). Scrutinizing the trustee’s role – trustor’s confidence that the trustee will be benevolent, competent, honest, open, and reliable – we found that most definitions clearly state that the trustor’s confidence is built on the perception that the trustee is capable (competent) and also possesses caring and trustworthiness traits (honesty, openness, and reliability).

However, the trustor’s perception of the trustee’s trustworthiness appears to be wrongly mistaken for the former’s willingness to risk vulnerability. Having confidence in another’s perceived trustworthiness does not automatically result in one’s willingness to risk vulnerability. When deciding to trust, the trustor may assess the trustee’s trustworthiness (Lerkkanen et al., 2013) but still take the final step regarding the nature and degree of vulnerability they would want to extend.

Even though some scholars have theorized that the “willingness to risk is the degree of confidence one has in a situation of vulnerability” (Hoy and Tschannen-Moran, 1999, p. 187), having confidence that the other party is trustworthy, and eventually extending trust (risking vulnerability) are two separate processes. Tschannen-Moran and Gareis (2015) argue that teachers’ perceptions and interactions with the principal are only a step toward the decision of whether to risk their vulnerability. We observed that most studies focus on measuring the intention to trust (willingness to risk vulnerability based on the confidence in the other party’s trustworthiness) rather than on the actual trusting (risk-taking) behavior.

Although risking vulnerability is associated with other motives such as desperation, obedience, impulsivity, innocence, or self-assurance (Hoy and Tschannen-Moran, 1999), trust requires action (Nienaber et al., 2015) because it is reciprocal. Teachers can feel trusted when the principal entrusts them with managerial tasks since by doing so, the principal exposes their vulnerability to teachers. Focusing on trust intentions is not enough since there is no guarantee that the actual trusting behavior will occur. For instance, the principal may be willing to risk vulnerability but never actually attempt it. In that case, teachers may not feel trusted since what the principal has expressed is merely their (principal) perception of the teachers’ trustworthy attributes. In a school setting, trust is demonstrated when leaders delegate a certain degree of power to their subordinates (Nienaber et al., 2015). Indeed, trust intentions arise from perceptions of trustworthiness. However, the most tangible evidence of trust is in the actual behavior.

Besides, trust and vulnerability are closely related. Deb and Chavali (2010) assert that “trust is consistently related to the vulnerability of the trustor because without the vulnerability of the trustor upon the trustee, trust becomes irrelevant” (p. 44). Further, Goddard et al. (2001) contend that “where there is no vulnerability, there is no need for trust” (p. 7). This implies that confidence is insufficient to be the only component in defining trust. Perceptions remain perceptions until one’s vulnerability is exposed to another party, and at that point, trust is formed. Trust assumes a state of vulnerability on the part of those who trust and furthers their willingness to take risks (Walker et al., 2011). This infers that there is an evaluation of vulnerability prior to the decision to risk it. Yet, the literature is silent about what kind of vulnerability is at stake in trusting relationships.

Accordingly, vulnerability assessment comes first, and the trustor has to ascertain whether it is safe to expose it to the other party. Lewis and Weigert (1985) argue that “we choose whom we will trust, in which aspects and under what circumstances, and we base the choice on what we take to be ‘good reasons,’ constituting evidence of trustworthiness” (p. 670). In the same vein, Hoy and Tschannen-Moran (1999) assert that trust is manifested on account of the nature of vulnerability to be risked. Therefore, we argue that trust requires risking a particular type of vulnerability.

Many of the reviewed studies confuse trust and trustworthiness and have used them interchangeably. Yet these two constructs are not the same. While trust refers to the risking of vulnerability based on the other party’s trustworthy behavior, trustworthiness refers to the characteristics of, and conditions around the person or thing being trusted, that facilitate that trust (Farnsworth et al., 2019). Thus, the so-called “facets of trust” are actually facets of trustworthiness and not trust. Facets are dimensions, sides, characteristics, or aspects of something. It is, therefore, grossly inaccurate to assume that benevolence, reliability, competency, openness, or integrity are components of trust. These are the trustworthiness attributes of the trustee as perceived by the trustor.

We argue that the said facets of trust should be more appropriately termed as bases of trustworthiness. On the other hand, trustworthiness is one of the bases of trust as it influences the decision to trust. Facets of trust should essentially reflect the components of trust (intrapersonal, relational, or collective) based on the context and nature of vulnerability (whether passive or active) (Poza et al., 2014; Nienaber et al., 2015). Trust is to the trustor while trustworthiness is to the trustee. Therefore, terming trustees’ characteristics as dimensions of trust weakens the conceptualization by shifting trust to the trustee. Without a doubt, the trustor’s willingness to risk vulnerability may be constructed on the confidence that the other party (trustee) is trustworthy. It is for this reason that trustworthiness can only be the basis for the trustor’s confidence in the trustee. However, what signifies the degree of trust is the trustor’s willingness to risk vulnerability. To demonstrate that the trustee characteristics may not always influence the decision to trust (Amemiya et al., 2020), note that,

Students may have a history of perceiving institutional bias and unfairness but express willingness to trust a particular teacher. These students may initially see their teacher as an exception to their broader theory of institutional injustice. However, when their teacher disciplines them, this punitive action may be perceived as confirmation of their theory that the school and its specific actors are unjust (p. 673).

While trustworthiness attributes (bases) are in some cases the major determinants of the decision to trust, research is still inconclusive concerning which particular ones are more instrumental in that decision. Most of the reviewed studies acknowledge all five common bases of trustworthiness (benevolence, honesty, reliability, openness, and competency) as collectively constituting trust while other studies focus on only some of them. Even though there is an ongoing debate regarding the nature and number of bases required to compute trust, we argue that the major concern should be the misconception of facets of trust and those of trustworthiness. As stated earlier, the said facets of trust are in fact the bases of trustworthiness and may influence trust indirectly through a mediator namely, confidence. This fits well with trust’s definition of willingness to risk vulnerability based on the confidence that the trusted party is trustworthy. The confidence (whether low or high) will lead to the decision of what degree of vulnerability can be risked. Likewise, confidence is built on those bases of trustworthiness. It will be irrational to suppose that the bases of trustworthiness have a direct influence on trust.

The confusion of facets of trust, trustworthiness, confidence, and vulnerability risking points to the weakness in the conceptualization of trust. Based on the reviewed literature, we highlight the stages of trust formation which can clarify the misperceptions in the home–school context. The first stage in trust formation is the assessment of trustworthiness. It is during this stage that the bases of trustworthiness (benevolence, honesty, reliability, openness, and competency) are considered. If assessed as positive, the perceived bases of trustworthiness will then boost the trustor’s confidence in the trustee. Thereafter, working together with other factors such as propensity to trust and shared goals, the built-up confidence will influence the actual risking of vulnerability. Trust is formed when the risk is being taken without any misgiving. It bears repeating that the concept of trust incorporates risking vulnerability traded against the confidence one has in the other party. Understanding the crucial role of this aspect is a giant step toward a more valid measure of trust.

We noticed that the statistical computation of trust is based on the perception of trustworthiness. Vulnerability risk is silent in almost all computations of trust, thereby justifying the notion that empirical examinations of mutual trust have not always aligned with its conceptualization (McAllister, 1995). Even in studies that acknowledge vulnerability, trust is still computed largely based on the five “facets of trust.” While we took notice of a few exceptional studies that acknowledged and incorporated both trustworthiness characteristics and vulnerability items in their scales as crucial components of trust (Hoy and Tschannen-Moran, 1999; Goddard et al., 2001), the measurement of vulnerability in these studies is shockingly cursory, and the items do not appear to measure vulnerability. Examples of such items purporting to measure vulnerability included the following: “trustor trusts trustee,” “trustor trusts the trustee to support them,” and “trustor is suspicious of the trustee.” These items are too general and do not precisely capture the aspect of vulnerability being risked in a trusting relationship.

Moreover, the computation of trust by aggregating scores on items intended to measure vulnerability with those intended for trustworthiness raises validity concerns. This is because trustworthiness influences the degree of vulnerability to be risked. Therefore, the influencing and influenced factors should not be averaged together to compute trust. Trust scales must examine what aspects of vulnerability the trustor is risking, rather than wholly focus on the trustee’s trustworthiness.

Measuring trust should also incorporate the relationship bases between trustor and trustee. This is because the trustor’s risking of vulnerability is not only influenced by the trustee’s trustworthiness but also the relationship between the trustor and trustee. Tschannen-Moran and Hoy (2000) observe that trust complexity extends to the relationship context of referents. They argue that “trust is multifaceted and may have different bases and degrees depending on the context of the trust relationship” (p. 551). The teacher–parent relationship, for example, is more complex (Keyes, 2002) than the teacher–student relationship. For decades research in home–school interactions has been attempted to establish relationship grounds between home and school (Fiore, 2001; Murray et al., 2008, 2013; Epstein, 2010) yet variations still exist. Parents have been perceived as uneducated or uncaring (Murray et al., 2008) in such a manner that their relationship with teachers and schools is treated at the clientele level.

On the other hand, researchers establish that a partnership-like relationship referred to as the family–professional collaborative relationship between educators and parents fosters their engagement and trust (Dirks and Ferrin, 2001; Murray et al., 2013). A partnership is built through practices such as establishing trust, stable relations, mutual respect and understanding, reciprocal communication, involvement in decision making, and efforts to use the school as a communal center (Henderson and Mapp, 2002). Apart from those practices, parents seek out advice from trusted members of the community such as educators to support their children’s development (Poza et al., 2014). Given that, it is controversial to measure trust between these referents without considering their trust relationship prominence. Understanding the nature and degree of trust relies very much on the relationship between parties in such a way that that it explains the requirements needed for trust to develop and flourish.

Bryk and Schneider (2002) contend that teacher–student trust in elementary schools operates primarily through teacher–parent trust. Additionally, research shows that teachers’ trust in students should be examined as a unified concept (trust in clients) in lower levels of education since the existing relationship between teachers and students is based on that of parents (Adams and Christenson, 2000; Goddard et al., 2001; Karakuş and Savas, 2012). However, during the schooling process, teachers, parents, and students work together with different levels of needs. Research shows that the student–teacher trust relationship is based on competence (Lawson, 2018), fair treatment (Amemiya et al., 2020), and care, whilst for parents, it is built on professional relationship, that is reliability, competence, and honesty (Janssen et al., 2012). We agree that students can be treated as clients, as they are an integral part of the parent–teacher/school relationship. Nonetheless, we argue that the differences in the relationships should not be ignored.

Although it is theoretically possible for teachers and students to be examined in a unified construct, we still cannot ignore different needs and levels of interdependence especially at higher grades (Van Maele and Van Houtte, 2009). This is evident in studies where these two referents have been examined separately (Adams and Christenson, 2000; Lee, 2007; Dönmez et al., 2010; Lerkkanen et al., 2013; Romero, 2015). Grounded in those differences, we, therefore, suggest that trust in parents and students should be examined separately not as a unified construct, especially in higher grades.

Furthermore, even though the trustee is perceived to be trustworthy, the decision of risking vulnerability cannot be guaranteed. This is because there are other factors besides trustworthiness that influence the trustor’s decision to risk. For example, the propensity to trust (Korsgaard et al., 2014), expectations (Fox et al., 2015), and the relationship between the trustor and trustee (Tschannen-Moran and Hoy, 2000).

This review has revealed discrepancies in the conceptualization, measurement, and operationalization of trust in recent home–school trust literature. Based on the discussion above, we, therefore, propose a shift in the conceptualization and measurement of trust.

Regarding trust conceptualization, the vulnerability to be risked by the trustor should be considered as an essential pillar in the trust formation process and is the ultimate evidence of trust. The assessment of the trustee’s trustworthiness characteristics, the shared goal, and the state of the relationship with the other party may lead the trustor to the decision of either surrendering their vulnerability or not. However, what eventually signifies the presence of trust or lack of it is the actual act of risking vulnerability. Our proposed definition of trust is, therefore,

The extent to which the trustor is willing to risk a particular aspect or degree of vulnerability, triggered by the propensity to trust, shared goals, the relationship between the trustor and the trustee, and the confidence that the trustee is trustworthy.

This definition is represented in a simple model (see Figure 3) for a clear conceptualization and understanding of trust. The model includes all key aspects of trust from trustor’s role (propensity to trust), trustor’s role (trustworthiness characteristics), trustor’s and trustee’s expectations, and their relationship. It also mirrors elements to be considered for the measurement of trust.

In the same vein, we propose two approaches through which trust can be computed. In the first approach, trust can be examined through the trustor’s behavior, that is, the nature and degree of vulnerability to be risked. Researchers should determine the types of vulnerabilities within the home–school context and assess the extent to which the trustor is willing to risk them. The second approach is closely related to the current method. However, in this particular approach, trust should be measured based on all the factors that influence the decision to risk vulnerability. These include: (1) the perception of the trustee’s trustworthiness, (2) shared goals/expectations, and (3) the nature of the relationship between the trustor and the trustee. Measuring trust through this method will, however, require the researcher to do the following: (a) control for the trustor’s propensity to trust; (b) establish relationship types (for example, under the teacher–principal referent, the relationship can be that of a leader and followers, supervisor and subordinates, or collegial) to understand their influence on vulnerability risking. This will also enable researchers to draw a line between trust, respect, and fulfillment of obligations; (c) revisit the facets of trustworthiness by identifying critical antecedents of trustworthiness, and determine which ones are more associated with vulnerability risking; and (d) investigate home–school expectations to understand whether schools and families have a common understanding of the shared goal. As trust is reciprocal, we recommend the computation of the reciprocity of trust by examining both from-school and to-school trust, which will include all key players in home–school trust.

The findings of this review cannot be generalized due to some limitations arising from our inclusion/exclusion criteria. First, the review excluded all gray research, books, dissertations, and symposia papers due to, among other reasons, validity concerns. Additionally, articles in languages other than English were not included. These excluded sources might probably have immensely contributed to this review.

Second, the review only dealt with papers using K-12 samples to the exclusion of those with post-high school samples. However, the post-secondary educational levels might have enriched our findings, especially the nature and types of relationships between trust referents. Finally, our review was limited to quantitative studies. Undoubtedly, qualitative studies in home–school trust have wider scope in capturing the concept of trust. Nonetheless, both qualitative and quantitative methods provide data with distinct understandings of trust, which might have been proven difficult to synthesize together. We recommend that future reviews consider expanding the scope of this current one by turning to qualitative studies to provide a deeper understanding of home–school trust.