The dataset for validation of factors affecting teachers’ decision to integrate character values into curriculum

The objective of this dataset is to examine the effect of factors of pedagogical content knowledge (PCK) and teachers’ beliefs (TB) on teachers' decisions (TD) to select character values to integrate into the curriculum in primary school in Indonesia. The data propose that PCK factors and teachers' beliefs (TB) factors significantly influence TD. PCK factors consist of content knowledge (CK), pedagogical knowledge (PK), and pedagogical content knowledge (PCK). While, TB factors consist of Attitude (ATT), Subjective norm (SN), and perceived behavioural control (PBC). The survey approach obtained 50 responses from one public school and two private schools in Indonesia. After adapting the survey instrument, face and content validity were conducted. Further, to examine the validity and reliability of the measurement model, a Partial Least Squares Structural Equation Model (PLS-SEM) was applied. For this purpose, the statistical process presents the load of the reflection indicator, the reliability of internal consistency, and the validity of convergence and discrimination. The dataset consists of demographic information, PCK, beliefs on character values integration, and teachers' decision to select character values to integrate into the curriculum. The dataset is beneficial to curriculum developers, school principals, and teachers for measuring factors affecting teachers' decisions to integrate character values into the curriculum.


a b s t r a c t
The objective of this dataset is to examine the effect of factors of pedagogical content knowledge (PCK) and teachers' beliefs (TB) on teachers' decisions (TD) to select character values to integrate into the curriculum in primary school in Indonesia. The data propose that PCK factors and teachers' beliefs (TB) factors significantly influence TD. PCK factors consist of content knowledge (CK), pedagogical knowledge (PK), and pedagogical content knowledge (PCK). While, TB factors consist of Attitude (ATT), Subjective norm (SN), and perceived behavioural control (PBC). The survey approach obtained 50 responses from one public school and two private schools in Indonesia. After adapting the survey instrument, face and content validity were conducted. Further, to examine the validity and reliability of the measurement model, a Partial Least Squares Structural Equation Model (PLS-SEM) was applied. For this purpose, the statistical process presents the load of the reflection indicator, the reliability of internal consistency, and the validity of convergence and discrimination. The dataset consists of demographic information, PCK, beliefs on character values integration, and teachers' decision to select character values to integrate into the curriculum. The dataset is beneficial to curriculum developers, school principals, and teachers for measuring factors affecting teachers' decisions to integrate character values into the curriculum. ©

Value of the Data
• Valid and reliable dataset are important to support studies regarding character values integration • in education.
• Practically, the data are beneficial for curriculum development centers, teachers, and school leaders to integrate proper character values into curriculum. • The dataset could be adopted, adapted, or extended for future researchers interested in conducting research with similar topics.

Data Description
This dataset proposes that pedagogical content knowledge (PCK) factors and teachers' beliefs factors significantly influence teachers' decision (TD) to integrate character values into the curriculum. PCK factors include content knowledge (CK), pedagogical knowledge (PK), and pedagogical content knowledge. Teachers' beliefs factors include attitude (ATT), subjective norms (SN) and perceived behavioural control (PBC). CK is a professional competency to master widely and deeply learning content, including mastery of curriculum content, taught subject matter at school. PK is described as teachers' ability to meaningfully deliver the lesson integrated with character values to students in the classroom. Meanwhile, a competency to combine two knowledge, CK and PK, to become part of the teaching process is defined as PCK. ATT is defined as teachers' feelings or mental state about the decision to select character values to integrate into the curriculum. At the same time, SN is described as teachers' beliefs that other people support them in determining character values to incorporate into the curriculum. PBC is defined by teachers' views regarding the availability of resources to assess character values to integrate into the curriculum. BI refers to teachers' intention to select character values to incorporate into the  ( Table 1 ). While the main survey has six exogenous and two endogenous constructs ( Fig. 1 ). Six exogenous are three constructs included in PCK measured from 1 = strongly disagree to 5 = strongly agree are CK (3 items), PK (3 items), and PCK (4 items), adapted from previous academic research [1 , 2] , and three constructs included in teachers' beliefs which are ATT (7 items), SN (5 items), and PBC (7 items)). The last constructs were two endogenous constructs, namely BI (2 items) and TD (5 items) [3][4][5] . Table 2 exhibits the Mean, Standard Deviation, Skewness and Kurtosis of the data.  Table 3 provides the information on the three assessments of the measurement model (reflective indicator loadings, internal consistency reliability, and convergent validity). Tables 4 and 5 show the discriminant validity by evaluating the Fornell-Larcker criterion and cross-loading. Fig. 2 exhibits the measurement model of the dataset. The raw dataset and instrument are accessible on https://data.mendeley.com/datasets/dd7hnsk4xf/3 .

Experimental Design, Materials and Methods
We applied 2-phase procedures in this study for scale development. Phase 1 is the adaptation and translation of the research instrument. We adapted the research instrument referring to previous literature sources, followed by the translation of the scale. The scale was translated from English to Indonesian and Indonesian to English using a reverse translation method involving two experts. In phase 2, face and content validity were conducted with two discussion sessions. The first session was a discussion with three users to ensure that the instrument was easy to understand by the sample respondents. The next session was a discussion with three experts to evaluate the scale for the appropriateness of context and setting. We then did an online survey based on google form to collect the data from March to April 2022 through simple random sampling. We randomly selected a subset of participants from the population who are primary school teachers in three Indonesian provinces, Balikpapan, Batam, and Depok. After receiving all responses, we converted the data into Microsoft Excel. Firstly, we assessed the normality by calculating Skewness and Kurtosis in SPSS 25, in which the values should be between -2 to + 2 [6] . All Skewness and Kurtosis values are in the range of the threshold; Skewness (SN3, -186 to ATT1, 1.259) and Kurtosis (BI1, 0.879 to ATT5, -417) ( Table 1 ). We then reported the four assessments of the measurement model (reflective indicator loadings, internal consistency reliability, discriminant and convergent validity) using the approach of PLS-SEM in Smart PLS 3.2. The loading of the reflective indicator should be .708 or higher. Table 2 performs all loading values that fulfil the threshold (.760-.968). We dropped two items (PBC4 and PBC6) due to their low loading values. Cronbach's alpha and Composite Reliability (CR) of greater than .700 should be applied for the internal consistency [6 , 7] . The Cronbach's alpha values of this dataset range from .913 to .973; similarly, the CR values are between .935 and .979. The validity of convergent was reported through Average Variance Extracted (AVE); the value of .500 or higher is recommended [8] . The AVE values range from .743 to .937 ( Table 2 ). The discriminant validity was evaluated by using the Fornell-Larcker and cross-loading. The AVE values of a construct should be less than the shared variance of the Fornell-Larcker's other construct. The data showed that the values of every construct are less than its' shared variance ( Table 4 ). The discriminant validity is reported when loading on a construct is greater than those of other constructs; cross-loading values. The values for all indicators (bold) in each construct exceeded all their cross-loadings ( Table 5 ). The model consists of eight constructs with 35 indicators ( Fig. 2 ).

Ethics Statements
Informed consent was obtained for the data collection and the participation was voluntary. The survey was anonymous that did not include any personal information of the participants.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.