Validity and Reliability of Organizational Trust Instrument

Organizational trust is a multifaceted construct with various definitions and measurement items according to the definition given by the researchers. Hence, this research was done to assess the aspects of organizational trust that exist in schools, as well as its validity and reliability. All constructs utilized were modified from the Omnibus Trust-Scale. The three constructs that make up the overall 26 items are trusted in colleagues, trust in clients, and trust in principle. Note that the content validity process involved a total of six experts. Subsequently, the instrument's validity was assessed via the Content Validity Index (CVI) and adapted Kappa coefficient. The study results established that 22 items have a threshold value greater than 0.8, and four items have a value lower than 0.8. Three items were modified for the next validation process, and one was dropped. Consequently, all 25 items were kept after the validity process was completed, and a pilot test was employed to conduct a reliability analysis. Moreover, 130 heads of committees from primary school were given the questionnaire during the pilot test to calculate Cronbach's Alpha value. Because the corrected item-total correlation had a low value, one item was eliminated following the reliability analysis, leaving 24 items in place. This instrument provides a new perspective on measuring organizational trust in education, especially in the school context.


Introduction
Trust refers to interpersonal relationships between communities that are critical to an organization's functionality (Dzul et al., 2021). An individual's or group's behavior, attitude, and readiness to take chances that other people are trustworthy, competent, benevolent, honest and open are indications of trust (Hoy & Tschannen-Moran, 1999). Because of its complexity, scholars believe that trust cannot simply be defined but may be found in various situations, including philosophical, economic, individual, and organizational factors (Choong et al., 2018). Additionally, one can judge someone's or a group's level of trustworthiness by observing their actions, communication, commitments, and words in both spoken or written declarations (Tschannen-Moran & Hoy, 1998). One of the key factors of a school organization's effectiveness is trust within the school organization itself (Tschannen-Moran, 2004). Social relationships like trust can create a positive school climate (Bryk & Schneider, 2003), which affects how teachers work, learn, and focus on the task at hand (Tschannen-Moran & Hoy, 1998). Other than that, a positive working relationship at school can foster openness and cooperation while allowing teachers to play a role and showcase their abilities (Tschannen-Moran & Hoy, 1998).
Trust is needed when teachers collaborate, leading them to discover new teaching strategies (Goddard & Goddard, 2001). It also provides teachers with a sense of security in experimenting with new teaching practices (Bryk & Schneider, 2003), influences teachers' teaching performances (Fitria, 2018;Liou et al., 2016), and encourages teachers to innovate (Cai & Tang, 2021). Note that distrust among administrators, teachers, and students can lead to an uncomfortable situation, increased conflict, and a failure to develop students' cognitive and social-emotional functions (Tschannen-Moran & Hoy, 1998). This is because trust is a multi-dimensional construct that has three different aspects: (i) trust in clients, (ii) trust in colleagues, as well as (iii) trust in principle. These elements will promote a culture of high trust in the school's working environment.

Literature Review
The aspects of trust, organizational trust's dimension, instrumentation, validity, and reliability from the literature are to be reviewed in this section.

Elements of Trust
Five components of trust were identified by Tschannen-Moran & Hoy (2000): (i) benevolence, (ii) reliability, (iii) competence, (iv) honesty, as well as (v) openness. In addition, the degree of trust shared by the parties and the type of their connection influence the significance of each component. On the other hand, Vodicka (2006) offers an alternative viewpoint on the components of trust, which are consistency, compassion, communication, and competence. He contends that to promote compassion, leaders must show faith in their subordinates' talents, acknowledge their efforts, utilize common courtesies like saying "thank you" and "help" as well as extending forgiveness. Correspondingly, leaders must provide feedback on both personal and organizational performance. Table 1 describes the definition of elements of trust. Table 1 Definition of elements of trust Elements of trust Definition

Benevolence
The most basic element in a trust virtue. Have confidence that the administrator is willing and will protect the teacher.

Reliability
Teachers rely on others to protect and serve consistently. Dependence presents that a person is confident that their needs will be met positively and can benefit others.

Competence
The ability to perform things based on established standards. An individual depends on the competence and skills of other individuals to meet their needs. An individual will not be trusted if they do not possess the skills to carry out their duties.

Honesty
Character, integrity, and authenticity are leadership behaviors that form the basis of trust. Having a sense of personal accountability and not lying about the truth and blaming others.

Openness
Putting yourself at risk when sharing information with other individuals and not hiding information. Information shared may be community-related or personal. Openness shows mutual trust. Meaning, confidence that the info will not be misused and the recipient can feel the same confidence Source: Hoy et al (2006);Hoy & Tschannen-Moran (1999).

Dimension of Organizational Trust
Three aspects of trust were identified by Hoy & Tschannen-Moran (2003): (i) trust in clients, (ii) trust in colleagues, as well (iii) trust in principle.

Trust in Principle
According to Altinkurt and Yilmaz (2012), trust in the principle is strongly correlated with the teacher's trust in the principal's honesty, ability to maintain commitments made, ability to form positive relationships, care for difficulties, and ability to protect both the principle and teachers' secrecy. Consequently, the committee leader is encouraged to work when the teacher acts in a way that is supportive of the principle by demonstrating concern for ideas and feelings. Furthermore, Canipe (2006) argued that trust in principle positively correlates with team productivity, effective communication, and decision-making processes.

Trust in Colleagues
The extent of trust between teachers and their peers is referred to as trust in colleagues. This relationship of trust is highly dependent on the integrity and openness of the counterpart in interacting, the way of treatment, as well as the supportive attitude (Tarter et al., 1989;Tschannen-Moran & Hoy, 1998). Here, building trust between instructors and other colleagues can help them cooperate and collaborate effectively (Cranston, 2011) and mutual respect for the competence and expertise of other colleagues (Tschannen-Moran, 2009). Other than that, trust between colleagues will form a safe and comfortable relationship. Even if the teachers do not know each other, they can still work and learn together. The safe and comfortable environment will allow teachers and peers to share, ask questions, discuss ideas related to teaching and learning, collaborate, and have reflective dialogue (Cranston, 2011).

Trust in Clients
Trust in clients is influenced by the teacher's trust in the parents and students in terms of student work and parental support, as well as their assertions. The teacher also plays a key role in connecting students and parents with the school. This is because the teachers interact with the students daily, becoming a link in communication between parents and the school (Goddard et al., 2001). Furthermore, the collaborative relationship between the school, as well as parents, in terms of decision-making is strongly predicted by trust in clients (Hoy & Tschannen-Moran, 1999). The teacher also needs to believe in the students and parents to achieve the school's goals. Note that the relationship between students and teachers is important for the success of teaching and learning (Tschannen-Moran, 2009). Their relationship is likened to a parent-child relationship that needs to be based on trust (Bryk & Schneider, 2002). Teachers who believe their students are competent and trustworthy will build a learning environment that helps them succeed academically.

Instrumentation
This organizational trust instrument was adapted from Hoy & Tschannen-Moran (2003). This instrument comprises 26 items in three dimensions. This instrument has four negative items, which are item 2, item 5, and item 8 in the dimension of trust in principle and item 11 in the dimension of trust in colleagues. The researcher modified these four negative items to positive items and retained them for content validity assessment from experts. Item 2, "I am suspicious of most of the principal's actions," was changed to "I am confident of most of the principal's actions." Subsequently, item 5, "Principal of this school does not show concern for teachers," was changed to "Principal of this school shows concern for teachers." Moreover, the phrase 'The principal does not tell teachers what is really going on' in item 8 has been substituted with 'The principal tells teachers what is really going on.' Item 11, "Teachers in this school are suspicious of each other," was changed to "Teachers in this school trust each other," but it resembles item 9. Therefore, the item was dropped from the instrument, resulting in 25 items being kept for the content validity procedure.

Validity
The validity of an instrument refers to the extent to which it measures what it is designed to measure (Ary et al., 2010). Adapted instruments are needed to be tested and validated because they are applied in a new context (Fraenkel et al., 2012). CVI was utilized to assess the content validity of the organizational trust instrument. Two types of CVI exist: (i) CVI for Items (I-CVI) as well as (ii) CVI for Scales (S-CVI) (Yusoff, 2019). Consequently, expert opinions on questionnaire items were sought to create a CVI score. The suggested expert numbers and their implications pertaining to the acceptable CVI cut-off scores are presented in Table 2. At least three experts are advised to do the content validity assessment (Shrotryia & Dhanda, 2019) and no more than ten people (Polit & Beck, 2006). To demonstrate content validity, the S-CVI value must be at least 0.8, and the I-CVI value must not be less than 0.78 (Lynn, 1986;Shrotryia & Dhanda, 2019).  (1986) According to Table 2, an acceptable I-CVI value for two experts should be a minimum of 0.80. Meanwhile, an acceptable I-CVI value should be 1.00 for three to five experts. Besides, for six to eight experts, an acceptable I-CVI value is at least 0.83. Other than that, for at least nine experts, a CVI of at least 0.78 is considered acceptable. Therefore, the I-CVI proposed by Lynn (1986) and its acceptance value of 0.83 for six experts were employed to assess the content validity of the questionnaire utilized in this research. A scale of 4 is used for content validity, scale 1=relevant, 2=somewhat relevant, 3=relevant, and 4=very relevant. Davis (1992) asserts that this scale is crucial for calculating the I-CVI. Additionally, I-CVI is scored 1 when the evaluator gives the item a score of 3 or 4.

Reliability
Reliability is often utilized interchangeably with stability and internal consistency (Creswell, 2010;Pallant, 2001;Sekaran, 1992). When assessing the internal consistency of a construct, Cronbach's Alpha value is frequently employed (Cronbach, 1946;Norusis, 1977). It is common practice to use a Cronbach's Alpha value greater than 0.60 as an indicator of an instrument's reliability (Majid, 1990;Pallant, 2001;Siti Rahayah, 2003). Furthermore, Sekaran (1992) stated that an Alpha value of 0.60 to 0.80 is deemed acceptable, whereas an Alpha value greater than 0.80 is considered good. Note that a reliability rating of less than 0.60 is seen as low and unsatisfactory. The researcher utilized Cronbach's Alpha values to examine the questionnaire's reliability in light of the above explanation. Here, 130 primary school committees participated in a pilot test to determine the organizational trust questionnaire's reliability. Table 3 displays the Alpha coefficient's interpretations.

Methodology
Six experts were used for expert validation, and 130 heads of the committee participated in a pilot test for reliability assessment. Selecting the experts was in accordance with each person's level of experience and experience in the study's field (Muhamad Saiful Bahri Yusof, 2019). Therefore, this study utilizes the services of six experts in the field of education management who have the expertise, experience, and knowledge in the field of study. Details of the qualifications and expertise of each appointed expert are displayed in Table 4. The acceptable CVI score is at least 0.83 because this study employed six experts to ensure the questionnaire's content validity (Lynn, 1986).

Content Validity
Six experts contributed to the validation of the organizational trust questionnaire at this stage. On a scale of 1-4, six experts evaluated each item on the dimension of trust in the principal of the organizational trust, and their scores are presented in Table 5. In addition, the calculation of the Kappa coefficient and S-CVI/Ave value was also conducted for the organizational trust item. The items in the dimension of trust in principles have an I-CVI score between 0.83 and 1.00. Meanwhile, the value of the kappa coefficient ranges from 0.82 to 1.00. Ultimately, the S-CVI (average) pertaining to the dimension of trust in principle was 0.92, and all eight items were retained. Table 5 Ratings of trust in the principle S-CVI/Ave = 0.92 (accepted). I-CVI = content validity index for items; S-CVI = content validity index for scales; Pc = Probability of chance agreement Table 6 indicates the ratings on trust in colleague items. Each item has an I-CVI that falls between 0.83 and 1.00. On the contrary, each item in this part has a Kappa statistic that falls between 0.82 and 1.00. Overall, S-CVI (Average) for trust in a colleague is 0.98. As a result, all seven items were kept. Table 6.

Ratings of Trust in colleagues
S-CVI/Ave = 0.98 (accepted). I-CVI = content validity index for items; S-CVI = content validity index for scales; Pc = Probability of chance agreement. Table 7 presents the ratings on trust in client items. Each item's I-CVI falls between 0.83 and 1.00. In contrast, each item in this part has a Kappa statistic that falls between 0.82 and 1.00. S-CVI (Average) for trust in colleagues is 0.97 overall. Ten items were kept as a result.  Table 7 Ratings of Trust in clients S-CVI/Ave = 0.97 (accepted). I-CVI = content validity index for items; S-CVI = content validity index for scales; Pc = Probability of chance agreement As a summary, the S-CVI (Average) for all three dimensions is between 0.92 (Table 5), 0.98 (Table 6), and 0.97 (Table 7), respectively. The findings of the validity investigation utilizing the S-CVI (Average), Kappa statistics, as well as I-CVI are presented in Table 8. Following the validity analysis procedure, one item from the original instruments' 26 items was eliminated. Due to their values above 0.83, 25 items were kept in these three dimensions and were accepted (Polit et al., 2007;Polit & Beck, 2006;Lynn, 1986). Additionally, the study's modified kappa index (Kappa Coefficient) for six experts ranges from 0.82 to 1.00 (Polit et al., 2007). Overall, this instrument's items all have excellent levels of content validity and are accepted in their entirety within the study context. Table 8 Findings with respect to Validity Analysis according to I-CVI, S-CVI (average)

as well as Kappa Statistics
Reliability Table 9 exhibits Cronbach's Alpha coefficient value regarding the organizational trust instrument. It is crucial to highlight that the overall organizational trust instrument's value is .946. This showed the value is above .60, which has good internal consistency and a high level of reliability.  Table 9 Value of Cronbach's Alpha for Organizational Trust Questionnaire Table 10 displays the analysis of Cronbach's Alpha if the item is deleted for all items, which ranges in value between 0.931 to 0.946. This demonstrates that the organizational questionnaire had a high level of internal consistency. Item discrimination, which assesses how an item corresponds to the overall score, was calculated via the corrected item-total correlation. A redundant item will be deemed to have a correlation value below 0.3, and it shall be taken out from the questionnaire (Ismail et al., 2020). Aside from item 10, which possesses a low correlation value of -0.57, all of the items in this questionnaire have values of corrected item-total correlation greater than 0.3. Other than that, the questionnaire initially had 25 items prior to the reliability analysis. However, one item was eliminated because of the poor corrected item-total correlation value after the reliability analysis, leaving 24 questions in the questionnaire that may be utilized in the study area.

Discussion
In summary, there are three constructs with 26 items utilized to measure the reliability and validity of the organizational trust instrument. Six experts were employed to confirm the organizational trust instrument's content validity. As per the items' necessity and relevance, these experts checked them off. Every construct's S-CVI/Ave total suggests that the items possess a high level of content validity. Correspondingly, this measurement instrument greatly contributes to the measurement of organizational trust at school. The lack of studies related to organizational trust that occurs in schools can be overcome with instruments that can be utilized to measure organizational trust in the field of education, especially in schools (Dzul, 2021).

Limitations of The Study
The present research possesses several limitations. Firstly, the data was only gathered from national government schools as part of the School Transformation 2025 program (TS25), which included cohorts 1 until 4. Future studies might incorporate data from different kinds of schools in Malaysia. This would allow generalizations to be made about the study's findings. Second, it is possible to test the validity of other kinds, including face, construct, and criterion. Additionally, to enhance the validity and reliability of an instrument, a Structural Equation Model (SEM) can be employed in conjunction with quantitative analysis techniques like Exploratory Factor Analysis (EFA), Confirmatory Factor Analysis (CFA), as well as path analysis.

Conclusion
This survey was created to find out the heads of the committee's thoughts on organizational trust in primary schools. The findings of this study have three major implications. First, the findings presented that field experts and heads of committees in TS25 primary schools regarded the organizational trust instrument as highly valid and reliable. In other words, this instrument had been approved for use in primary schools. Second, the study's findings show that, trust in the school community can help the school function more productively (Tschannen-Moran & Hoy, 1998) and promote teachers' professionalism (Tschannen-Moran, 2009). Hence, administrators, teachers, students, and parents should work together to create an environment of mutual trust, efficiency, honesty, and openness in the school community