The influence of professional attitude, welfare, self-sustaining development, and job satisfaction on teacher performance

ABSTRACT


INTRODUCTION
Educational development is a priority and a crucial issue, particularly in terms of improving the quality of human life and national development [1], [2].Education development involves various stakeholders, including teachers as a strategic profession that can maximize educational success [3].As a result, maximum effort and focus are required to support teacher self-development [4].A teacher's professionalism can be seen in various ways, including knowledge mastery, teaching skills, and so on.This must be noted to improve educational quality [5], [6] as teachers must be active, dynamic, technologically literate, and competent in their fields [7], [8].The competence can be observed in their performance, particularly in planning, implementing, and evaluating learning [9] which is consistent with another ISSN: 2089-9823  The influence of professional attitude, welfare, self-sustaining development … (Tri Murwaningsih) 273

RESEARCH METHOD
This was quantitative study employed correlational design.Correlational study was research that aims to explain, describe, and analyze the influence or relationship between the variables studied [43].Correlational research involves the process of collecting data by finding out whether there is a relationship between two or more variables.
Population briefly as a research subject which includes the generalization area of the number and certain characteristics to be concluded [44].The population in this study is all teachers at state vocational high schools majoring in the business and management department with a total of 21 schools.This research is located in the Surakarta Residency, Indonesia.In the residency, there are six regencies and one municipality.In the process of determining the sample, the researcher uses a proportional stratified random sampling technique.Sampling is a way of collecting data or research [45].In this study, researchers used a proportional stratified random sampling technique.The technique is defined as a sampling technique in a diverse population.This technique is carried out by taking samples from the population based on levels and randomly [46] to obtain a representative sample.While the research sample is teachers who have the certificate of educators and other requirements with a total of 155 respondents.
Data collection instruments are the tools used to collect data [47].The questionnaire instrument is used in this study.This technique is a series or list of questions or statements that are organized systematically, then sent to be filled in by the respondent [48].Questionnaires are an efficient data collection technique when the variables to be measured are known with certainty and what can be expected and more appropriate to use if the number of samples includes a large area.The questionnaire contains closed statement sentences.There are five questionnaires, namely the teacher performance questionnaire, the teacher professional attitude, teacher welfare, self-sustaining development, and job satisfaction.The indicators for each variable are described in Table 1.This study used a Likert scale in the drafting process of questionnaires.The specified scale includes a scale of 1 to 5 with details 1 (strongly don't understand/disagree), 2 (don't understand/disagree), 3 (doubtful), 4 (understand/agree), and 5 (Completely understand/agree).The gradation used in this study is 5, as shown in Table 2.The researcher first tested the validity and reliability of the instrument testing tools.Validity is the degree of precision and accuracy of an instrument in carrying out its measuring function [47].An instrument is said to be valid if the tool performs the measuring function correctly.The validity test used the product moment formula because it only tested the validity of the contents.Meanwhile, the reliability test used the Cronbach Alpha test with a record Cronbach Alpha value of >0.60 [50].The questionnaire instrument trial was conducted on 30 respondents from three schools in the research population area.Table 3 presents the results of the validity and reliability of the instrument.Based on Table 3, the teacher's professional attitude obtained 12 valid items, teacher welfare obtained 20 valid items, self-sustaining development obtained 13 valid items, job satisfaction obtained 10 valid items, and teacher performance obtained 49 valid items.Each indicator was assessed by the principal and students.The test results also showed a correlation value of more than 0.30 so it can be said to be valid.
Reliability is called consistency, dependability, trustworthiness, stability, and constancy [51].Reliability describes the measurement process can be trusted.The instrument is said to have good reliability if the instrument obtains the same results in several measurements.The reliability test showed a coefficient value that was greater than Cronbach's Alpha criteria of 0.60 so it can be said that the variable was feasible to use in this study.A cross-section is used in data collection because it is collected simultaneously and only once.The primary data is in the form of a questionnaire, while the secondary data is in the form of documents.Questionnaires were given to teachers, school principals, and research sample students.
Data analysis is the process of processing data from all samples/data sources [52].Data analysis was carried out by grouping data, tabulating data, presenting data, and testing the hypotheses that had been proposed.Structural equation modeling (SEM) analysis technique is used to process the data.SEM analysis technique with linear structural relationship (LISREL) is used in the analysis of research results.The steps taken are: i) Developing a theory-based model; ii) Developing flowcharts to show causality; iii) Change the diagram into equation form; iv) Selecting input matrices and estimation techniques; v) Evaluate the good-offit criteria; vi) Interpreting the Sobel test model; and vii) Using the interview method to obtain additional data and explain the results of the hypothesis [53].

RESULTS AND DISCUSSION
There are several stages in the procedure for collecting and processing data with the structural equation model (SEM) technique.First, a sample of 155 people fulfilled the SEM requirements.Second, the researcher conducted a normality test with a record value of the critical skewness ratio and a multivariate value of 0.096.It means the data is normally distributed because it is less than 2.58 and the P-value is 0.924>0.05.Thus, the assumption of normality is met.Third, the researcher conducted a linearity test with regression analysis between variables.Table 4 contains the results of the linearity test.The results in Table 4 show that all variables are linear because the value is Sig.<0.05 so that the data can be used for subsequent analysis.Fourth, the researcher tested for outliers using the Mahalanobis distance method.If the Mahalanobis distance value is greater than the x 2  This means that all samples can be analyzed.Fifth, the Multicollinearity assumption requires that there is no perfect correlation between the independent variables.However, in this research, there was only one independent variable, namely teacher professionalism.So, researchers did not do the multicollinearity test.
The next stage is to assess the unidimensional and reliability of the construct with a reliability limit value of 0.70.The second reliability measure with variance extracted shows the amount of variance from the indicators extracted by the developed latent construct.Variance-extracted values are recommended at a level of at least 0.5.In this research, testing the significance of the loading indicator is done by analyzing the value of the critical ratio.If the critical ratio value is greater than the critical value at a significance level of 0.05 (critical value=1.96),then the indicator is significantly an indicator formed.If the reliability coefficient is 0.60, it can be said that the variable is reliable.
This evaluation is carried out by confirmatory factor analysis (CFA).This is because at this stage the model will confirm whether the observed variables reflect the factors being analyzed or not.First, confirmatory factor analysis constructs on the teacher's professional attitude.The results of the confirmatory analysis of the teacher's professional attitude are presented in Figure 1.
Figure 1.Confirmatory factor analysis of teacher's professional attitude variable Figure 1 produces five manifests that measure the significance of the teacher's professional attitude variable.The measurement is based on the results of confirmatory factor analysis.To make it clearer, Table 5 shows the construct validity calculation.Table 5 shows the calculated T value and the lambda coefficient are valid and the value of the reliability construct of the teacher's professional attitude is reliable because the value is ≥0.6.Then, to reduce the limitations of Chi-Square, we use relative or normed Chi-Square (χ2/df) using LISREL analysis criteria.The results are presented in Table 6.6 provides information that the teacher professional attitude construct has met the specified requirements, namely the Chi-Square coefficient of 1.668 is smaller than the cut-off of 2.0 and the P-value is 0.15399 greater than 0.05.Thus, it can be said to be a perfectly intact construct.Second, confirmatory factor analysis of teacher welfare constructs.The results of the confirmatory analysis of the teacher welfare construct are presented in Figure 2.

Figure 2. Confirmatory factor analysis of teacher welfare variables
The results of the single model confirmatory analysis show that there are five manifests.All of these manifests measure the significance of the teacher's welfare variable construct.The complete results can be seen in Table 7.The table shows the value of the lambda coefficient or loading factor of each manifest is valid.Meanwhile, the reliability value of the construct obtained 0.79.Chi-Square is used to reduce limitations by using the LISREL analysis criteria as presented in Table 8.Based on Table 8, it can be seen that the teacher welfare construct meets the Chi-Square requirements of 1.7883 which is smaller than the cut-off of 2.0 and the P-value of 0.097 is greater than 0.05 with 6 degrees of freedom.So, the existing manifests are perfectly intact constructs.Third, confirmatory factor analysis of self-sustaining development.The results of the confirmatory analysis of self-sustaining development are presented in Figure 3.The results of the single model confirmatory analysis in Figure 3 show that there are three manifests that measure the significance of the variable construct of sustainable selfdevelopment.To see the validity of the construct can be seen from the value of lambda (λ), coefficient of determination (R 2 ) and T-Value.The complete results can be seen in Table 9.Based on Table 9, it can be seen the construct validity of each indicator by comparing the T-value with the T-table on the degrees of freedom of each variable.The T-table value at 0 degrees of freedom with an alpha of 5% is 0.10.In addition, validity can also be seen from the loading factor value, namely the loading factor for each dimension is greater than 0.4.Thus, it can be concluded that the lambda coefficient or loading factor for each manifest is valid.For the level of reliability of the construct above, it is obtained 0.671.It can be said that the variable of sustainable self-development is reliable because it is ≥0.60.Tests using the LISREL analysis criteria with the construct as a single model show the results as presented in Table 10.Based on the table, it is known that the Chi-Square coefficient of 0.000 is less than the cut-off of 2.0 and the P-value of 1.00 is greater than 0.05.Thus, it can be said to be a perfectly intact construct.Fourth, confirmatory factor analysis of teacher job satisfaction construct.The results of the confirmatory analysis of job satisfaction are presented in Figure 4.The results of the single model confirmatory analysis above show that there are four manifests that measure the significance of the construct variable of teacher job satisfaction.To see construct validity, it can be seen from the lambda value (λ), the coefficient of determination (R 2 ) and the T-value.Complete results can be seen in Table 11.  .In addition, validity can also be seen from the loading factor value, namely the loading factor for each dimension is greater than 0.4.Thus, it can be concluded that the lambda coefficient or loading factor of each manifest is valid.For the reliability level of the construct above, it is obtained 0.72, so it can be said that the teacher's job satisfaction variable is reliable because it is ≥0.60.Tests using the LISREL analysis criteria with the construct as a single model show the results as presented in Table 12.Based on Table 12, it is known that the construct of teacher job satisfaction has met the specified requirements, namely a low Chi-Square coefficient of 2.095 and a P-value of 0.12 which is greater than 0.05.Fifth, confirmatory factor analysis teacher performance construct.The results of the confirmatory analysis of the teacher performance construct are presented in Figure 5.

Figure 5. Confirmatory factor analysis of teacher performance variables
In this study, the construct of teacher performance consists of six indicators.The results of the single model confirmatory analysis in Figure 5 show that there are six manifests that measure the significance of the teacher performance variable construct.The complete results of the confirmatory factor analysis of teacher performance variables can be seen in Table 13 Table 13 shows that overall, the lambda coefficient or loading factor of each manifest is valid.The reliability value reached 0.86 (reliable because ≥0.60).Tests using the LISREL analysis criteria with the construct as the single model for teacher performance variables show results as in Table 14.Based on Table 14, it is known that the Chi-Square coefficient of 1.564 is smaller than the cut-off of 2.0 and the P-value of 0.12 is greater than 0.05.Thus, it can be said that the manifest in the performance variables above is a perfect intact construct.The next analysis process is carried out by testing the goodness of fit.The complete results are presented in Table 15.Table 15 shows that the test model does not meet the criteria, especially the tests of absolute fit indices and parsimony fit indices.The results of the Chi-Square test show that the null hypothesis is rejected so that there is no difference between the model (population covariance) and the observed sample covariance.Based on the test results, it turns out that the x 2 /df value of 1.9546 is smaller than the cut-off =2.From the data, it can be concluded that the model can be said to be a fit.
Furthermore, hypothesis testing is carried out to consider the significance value of the coefficients on each path.The significance value will be considered if the critical ratio is above 1.96 and the T-count is above 0.4.Table 16 shows the results of the hypothesis testing.To test the hypothesis that has a direct effect is done by looking at the results of the coefficient (standardized) and significance (p) on each path of the direct relationship.As for testing the indirect effect, a test will be carried out on the coefficient (standardized) of the indirect relationship with the Sobel test.Analysis using the structural equation model (SEM) was carried out to obtain the results of an analysis of the relationship between exogenous variables and endogenous variables either directly or indirectly.Based on the results of the analysis of

Teacher professional attitude toward teacher welfare, self-sustaining development, job satisfaction, and teacher performance
First, the results regarding the teacher's professional attitude towards teacher welfare, self-sustaining development, job satisfaction, and teacher performance.The results also show several things.The first result is that the teacher's professional attitude influences teacher welfare because the coefficient value of the standardized regression weight is 0.66 and the t-count is 5.27.The second result is that the attitude of the teaching profession does not affect self-sustaining development because the coefficient value of the standardized regression weight is 0.21 and the t-count is 1.14.The third result is that the teacher's professional attitude does not affect teacher job satisfaction because the coefficient value of the standardized regression weight is 0.06 and the t-count is -0.29.The fourth result is that the teacher's professional attitude does not affect teacher performance because the coefficient value of the standardized regression weight is 0.09 and the t-count is 0.55.
The results of the study show that teachers' professional attitude has a significant effect on teacher welfare.The results are due to several things, such as the attitude of obeying school rules.The impact is that teachers get additional income so that their needs are met and more prosperous.This study is in line with the previous research [54], [55] stating that certification is more about increasing teacher welfare than increasing professionalism.
Then, the teacher's professional attitude does not influence self-sustaining development because of a lot of administrative workloads, strict school rules, other expensive expenses, and retirement age.The results of this study are concerned to the previous findings [56] showing that teacher professional development is not positively related to teacher welfare.However, these results are in contrast to Adkoli's research [57] stating that the progress of professionalism provides a good future for teachers.
The next result is that the teacher's professional attitude does not affect teacher job satisfaction.This is due to teachers' incompatibility with their teaching assignments and lack of teaching expertise.The results of this study are same match with the results of other studies [41] showing that attitudes toward schools do not give teachers job satisfaction, but the results are not same with Buric and Moe's findings [58] showing that teacher enthusiasm affects job satisfaction.
The final result is that the teacher's professional attitude does not affect teacher performance.This result is due to the feeling of comfort in the presence of professional allowances and the additional workload outside of the main teaching duties.The results of this study are aligned with previous findings [59] showing that the professional attitude of teachers does not make teacher performance effective.

Teacher welfare on self-sustaining development, job satisfaction, and teacher performance
Second, the results regarding teacher welfare on self-sustaining development, job satisfaction, and teacher performance.The results also show several things.The first result is that the teacher's welfare influences the teacher's self-sustaining development because the coefficient value of the standardized regression weight is 0.64 and the t-count is 2.94.The second result is that teacher welfare influences teacher job satisfaction because the coefficient value of the standardized regression weight is 0.82 and the t-count is 2.38.The third result is that teacher welfare does not affect teacher performance because the coefficient value of the standardized regression weight is -0.24 and the t-count is -0.59.
The second finding is about teacher welfare on self-sustaining development, job satisfaction, and teacher performance.There are three findings at once.First, teacher welfare towards self-sustaining development.This is due to the availability of funds, motivation, as well as facilities and infrastructure such as seminars, workshops, education, and training related to the profession.The results of this study are aligned ISSN: 2089-9823  The influence of professional attitude, welfare, self-sustaining development … (Tri Murwaningsih) 281 with the findings of Maba et al. [60] stating that teacher welfare affects school education services.Second, teacher welfare affects job satisfaction.Previous findings [61]- [65] also show similar data, namely that there is a relationship between satisfaction and employee welfare.Third, teacher welfare does not affect teacher performance.The results of this study are aligned with the previous findings [66] stating that an increase in teacher allowances has no impact on teacher qualifications.

Self-sustaining development on job satisfaction and teacher performance
Third, the results regarding self-sustaining development on job satisfaction and teacher performance.The results also note several things.The first result, self-sustaining development does not affect teacher job satisfaction because the coefficient value of the standardized regression weight is 0.06 and the t-count value is 0.21.The second result is self-sustaining development influences teacher performance because the coefficient value of the standardized regression weight is 0.61 and the t-count value is 2.40.The results of the first study show that self-sustaining development does not affect teacher job satisfaction.This is due to the lack of time for self-development.Besides, government policies do not support teacher selfdevelopment.This research is in line with other findings [67] showing that job satisfaction does not affect the professionalism of employees at public hospitals.However, this finding was refuted by Toropova et al. [68], Ismail and Rishani [69] because their research proves that professional development affects job satisfaction.The second result is that self-development affects teacher performance.The results of this research are aligned to previous study [70], [71] showing that career development has proven effective in improving employee performance.

Teacher job satisfaction on teacher performance
Fourth, job satisfaction does not affect teacher performance because the coefficient value of the standardized regression weight is 0.42 but the t-count value is 1.53.The final finding shows that job satisfaction does not affect teacher performance.This is due to the lack of smooth disbursement of professional allowances and the lack of teacher dedication.This research is in line with other findings [72], [68] showing that job satisfaction has no impact on employees' performance.However, other researchers [73], [74] argue the finding because they can prove that job satisfaction affects teacher performance.

CONCLUSION
It is possible to conclude that teacher professional attitude has an impact on teacher welfare but has no impact on self-sustaining development, job satisfaction, or teacher performance.Teacher well-being influences self-sustaining development and job satisfaction, but not teacher performance.Self-sustaining development improves job satisfaction but has no effect on teacher performance, and job satisfaction improves teacher performance but has no effect on teacher performance.
There are several limitations to this study, including a lack of focus on teachers' age characteristics.As a result, teachers may perceive instruments with varying outcomes.Future researchers should pay close attention to sample characteristics.The teacher's condition is then in a different area, so the teacher may have different insights.As a result, future researchers are encouraged to conduct a similar study in the same area.Furthermore, other researchers can conduct research with elementary or junior high school teachers as subjects.
These findings provide theoretical and practical implications.The resulting theoretical implications are in the form of the availability of theories about the attitude of the teaching profession, teacher welfare, sustainable self-development, job satisfaction, and teacher performance.Meanwhile, the practical implication of these findings is that the research results can be used as material for further studies for future researchers, for example conducting an in-depth analysis of all the variables of this study or just a few variables.Future researchers can also upgrade research variables related to teacher performance, such as teacher wages, principal managerial, work discipline, leadership style, and others.

Table 1 .
Research variable indicators

Table 3 .
Summary of instrument testing results

Table 4 .
Linearity test results

Table 5 .
Construct variables validity of teacher professional attitude

Table 6 .
The Conformity level of the single model of teacher professional attitude J Edu & Learn, Vol.17, No. 2, May 2023: 271-284 276 Table

Table 7 .
Construct validity of teacher welfare variables

Table 8 .
The conformity level of the single model of teacher welfare

Table 9 .
Construct validity of self-sustaining development variables

Table 11 .
Construct validity of teacher job satisfaction variables Based on Table 11, it can be seen the construct validity of each indicator by comparing the T-value with the T-table on the degrees of freedom of each variable.The T-table value at degrees of freedom 2 with an alpha of 5% is 2.920.All indicators have a T-count greater than T-table

Table 12 .
The conformity level of the single model of teacher job satisfaction

Table 13 .
. Construct validity of teacher performance variables The influence of professional attitude, welfare, self-sustaining development … (Tri Murwaningsih) 279

Table 14 .
The conformity level of teacher performance single model

Table 15 .
The goodness of fit test results

Table 16 ,
the following hypothesis testing can be carried out:

Table 16 .
The results of research hypothesis testing