The research that was conducted produced several data on variables’ distribution including ergonomic risk factor, lifestyle, work fatigue, and well-being. The ergonomic risk factor consist of work posture indicator (listed in Table 1). While lifestyle consists of several indicators as nutritional status, recommended dietary allowance, dietary habit, exercise habit, smoking habit, alcohol consumptions, and sleep quality. Finally, the well-being variable includes 6 indicators such as autonomy dimension, personal growth, positive relation with other, purpose in life, and self-acceptance.
In Table 1, the pottery craftsmen’s ergonomic risk factor in the Karanganyar Tourist Village, Magelang, showed that the majority of them had the work posture assessment results in the low and medium categories (28.4% each), the majority were in the age category of 46–55 years (28.4%), and most of them have work experience in the category of > 10 years (77.9%). These three variables were factors that influence the ergonomic risk of the respondents.
The variables examined as markers of lifestyle (Table 2) were nutritional status (as defined by BMI), nutritional adequacy, eating habits, exercise habits, smoking habits, alcohol consumption habits, and sleep quality. The measuring findings, which reflected the respondents’ lifestyles as pottery craftsmen in the Karanganyar Tourist Village, revealed that the majority of them had normal nutritional status, good nutritional adequacy, and good eating habits. Furthermore, the majority of responders did not exercise, did not smoke, did not use alcohol, and had decent sleep quality.
Work fatigue was objectively tested with a response timer and subjectively quantified with the IRFC questionnaire. According to the measurement results showed in Table 3, the majority of respondents (57.9%) had mild objective and moderate subjective work fatigue (46.2%).
Well-being variable was constructed from six dimensions, such as autonomy (high scores indicate the individual’s ability to act independently, unaffected by societal influences), environmental mastery (effective utilization of opportunities and a feeling of control over one’s surroundings, including managing daily affairs and creating advantageous situations), personal growth (ongoing development, openness to new experiences, and acknowledgement of personal improvement over time), positive relations with others (active participation in meaningful relationships characterized by mutual empathy, closeness, and affection), purpose in life (a strong sense of direction and belief in the meaningfulness of life), and self-acceptance (a positive self-perception). All six dimensions must be achieved well to achieve overall good well-being. The research results showed that the majority of respondents achieved well-being in the moderate category (64.2%) (listed in Table 4).
Analysis of the influence of ergonomic risk factor and lifestyle on work fatigue and Well-Being among pottery craftsmen in Karanganyar tourist village, Magelang
In this study, the data analysis will be conducted step by step. First, it begins with the evaluation of the measurement model (outer model) showed in Fig. 1 and Fig. 2 while the evaluation of the structural model (inner model) performed in Fig. 3 and Fig. 4. The outer model describes the relationship between each indicator and its latent variable, while the inner model illustrates the relationships among latent variables.
First outer model analysis
Second outer model analysis
First inner model testing
Outer model analysis
The measuring model assessment sought to investigate the link between each indicator and its latent variable, known as the outer model. Loading factor values and Average Variance Extracted (AVE) values were evaluated in the outer model analysis. The loading factor reflected the relationship between a given question item’s score and the score of the indicator construct gauging that construct. If the loading factor value produced is greater than 0.6 and the AVE value is greater than 0.5, the construct is legitimate [19]. The outer model analysis findings are shown in the Table 5 below:
According to the Table 5, numerous items have loading factor of < 0.6 (showing that they do not contribute to the formation of the sociodemographic latent variable), hence the observed variables had to be eliminated and retested to get new loading factor values. After deleting the incorrect observed variables, the study model is showed in Fig. 2.
Table 6 shows the value of AVE, composite reliability, and Cronbach Alpha on each variable with the following results:
Inner model analysis and hypothesis testing
The Inner Model (Evaluation of the Structural Model) was a model that offered an overview of the links between latent variables and other latent variables (showed in Fig. 3 and Fig. 4). This model might be bootstrapped to determine the amount of variability of variables in explaining other latent variables with the R-Square value.
Hypothesis testing in this research will use the T-Statistic values showed in first inner model testing as it performed in Fig. 3. At a significance level of 5%, a variable can be considered to have a significant effect on another latent variable if the T-Statistic value is > 1.96 or the P-Value is < 0.05. If the T-Statistic value is < 1.96 or the P-Value is > 0.05, then the research variable does not have a significant effect on the other latent variable. Path coefficients aim to measure the extent of the relationship between latent variables. If the coefficient value approaches + 1, the relationship between variables is positive, indicating a stronger predictive value for the dependent variable. The model that has been bootstrapped in this study showed in Fig. 3.
According to Fig. 3, the P-value between ergonomic risk factor (ERF) and fatigue, as well as the P-value between ERF and well-being (WB), were both greater than 0.05, indicating that the results of the inner model testing were non-significant. Table 7 has more extensive information as follows.
The non-significant hypothesis in the model was removed in the second phase of the inner model testing. Because ERF exhibited no link to both fatigue and well-being variables, it was eliminated from the model, and the final model is presented in Fig. 4 as follows.
The SRMR value in Table 8 shows both estimated and saturated model in 0.070 and 0.072 respectively, this indicates that the model is already fit.