p–ISSN: 2723 - 6609 e-ISSN: 2745-5254

Vol. 5, No. 12 Desember 2024 http://jist.publikasiindonesia.id/

Jurnal Indonesia Social Teknologi, Vol. 5, No. 12, December 2024 5837

Knowledge, Attitude, and Supervision Correlation Analysis in

Improving Work Safety Using The Pearson Method at A Tin

Ore Refining Company in The Jelitik Industrial Area of

Bangka Regency

Ivana Ardhia Larasati

, Rika Ernawati

, Shofa Rijalul Haq

, Nurkhamim

Tedy Agung Cahyadi

Universitas Pembangunan Nasional Veteran Yogyakarta, Indonesia

1,2,3,4,5

Email:

[email protected]

*Correspondence

ABSTRACT

Keywords: knowledge;

attitude; supervision;

Correlation

Occupational safety is a crucial aspect, especially in the mining

sector, which has a high risk of accidents due to various factors.

These factors include knowledge, attitude, and supervision.

Although PT Pemurnian Oreh Timah in the Jelitik Industrial

Estate recorded a few incidents of work accidents that occurred in

the last 1 year, the risk of work accidents still exists. This study

aims to analyze the level of close relationship between

independent variables, namely knowledge and attitude,

knowledge and supervision, and attitude with supervision in the

Engineering Department of PT Pemurnian Oreh Timah in the

Jelitik Industrial Estate, Bangka Regency. Pearson's correlation

test method was used to analyze questionnaire data from 68

workers in the company's engineering department. The results

showed that there was a weak relationship (r = 0.311) between

knowledge and attitude, as well as a moderate relationship (r =

0.502) between knowledge and supervision. A moderate

relationship (r = 0.484) was also found between attitude and

supervision. These findings show that these factors are

interrelated, where it is hoped that effective supervision can

increase workers' knowledge and attitudes towards occupational

safety.

Introduction

Occupational safety is a crucial aspect in every industry, especially in the mining

sector, which has a high risk of accidents. Work accidents affect not only workers but

also companies and society at large. Mondy and Noe (2005) define occupational safety

as an effort to protect employees from injuries caused by work-related accidents. Based

on Law No. 1 of (1970), work accidents are unexpected and unplanned events that can

cause chaos in a process and result in loss of property, property, and people. Based on

research by Bird and German (1990), work accidents are defined as unexpected events

that result in losses to people, property, and the environment. The International Labour

Organization (1998) identified factors that contribute to work accidents, including worker

Ivana Ardhia Larasati, Rika Ernawati, Shofa Rijalul Haq, Nurkhamim, Tedy Agung Cahyadi

Jurnal Indonesia Social Teknologi, Vol. 5, No. 12, December 2024 5838

factors, management factors, and work environment factors. Workers' knowledge and

attitudes play an important role in preventing work accidents.

Notoatmodjo (2013) explained that knowledge is obtained through the five senses

and is influenced by internal factors such as education, interests, experience, and age, as

well as external factors such as economy, environment, and culture. Winardim (2020)

defines attitude as a mental state that is studied and organized based on experience, which

affects a person's reaction to others, objects, and situations. Zuchdi (2022) added that

attitudes are influenced by personal experiences, culture, important people, mass media,

educational and religious institutions, as well as emotional factors. Effective supervision

is key to ensuring that work is carried out in accordance with safety plans and standards.

Sarwono (1991) emphasized the importance of inspection, checking, matching,

inspection, and control activities in supervision.

The Jelitik Industrial Estate in Bangka Regency is one of the industrial estates that

has a high potential for work accidents. This area has abundant tin sand content and is the

location for several companies engaged in mining, processing, and refining tin ore,

including PT Pemurnian Oreh Timah. Although the company has recorded a few incidents

of work accidents in the past 1 year, potential risks still exist. This study aims to analyze

the close correlation between knowledge and attitude, knowledge and supervision, and

attitude with supervision in the Engineering Department of PT Pemurnian Ore Timah in

the Jelitik Industrial Estate, Bangka Regency. By examining these factors in depth, it is

hoped that the results of the research can make a significant contribution to the

development of more effective accident prevention strategies, as well as improve the

implementation of safety practices in accordance with industry standards. In addition, the

results of this study can enrich the occupational safety literature and provide practical

guidance for companies in an effort to minimize risks and promote a safe work

environment.

Method

This research uses a quantitative approach with a correlational descriptive design,

conducted in the Engineering Department of Tin Ore Refining Company in the Jelitik

Industrial Estate, Bangka Regency. The research sample consisted of 68 engineering

department workers who were selected by purposive sampling. Data was collected

through observation in the form of in-depth interviews with the Head of K3 and the

company's summit, as well as a questionnaire consisting of 15 questions for each variable

(knowledge, attitude and supervision). Instruments and classical assumption tests have

tested the questionnaire data. The test of the instrument in the form of validity and

reliability uses the Eviews 10 program, with a table value of 0.2387 and an Alpha

Cronbach value of > 0.05, where if the result is > table, the data is declared valid and

reliable. After that, a data normality test and a heteroskedasticity test were carried out.

The data normality test in the Eviews 10 program using the Jarque-Bera method is seen

with the condition that if the probability value > 0.05, the data is normally distributed.

Meanwhile, the heteroskedasticity test uses the scatterplot graph method and the White

Knowledge, Attitude, and Supervision Correlation Analysis in Improving Work Safety Using The Pearson

Method at A Tin Ore Refining Company in The Jelitik Industrial Area of Bangka Regency

Jurnal Indonesia Social Teknologi, Vol. x, No. x, Month 2024 5839

method resulting from the output of the Eviews 10 program (Friera et al., 2024). After

that, the data was analyzed using the Pearson Product Moment Correlation Test (r) to test

the level of the close relationship between the stated variables and the correlation

coefficient (r). Correlation analysis (r), according to Sugiyono (2015), only measures the

strength of the relationship between variables without having an impact on the causal

relationship. However, correlation analysis can show the direction of the relationship, i.e.,

positive or negative relationship. The characteristics of the correlation coefficient of r are:

1) If the correlation value (r) is above 0.05, there is a relationship between variables; 2)

Perfect/very strong relationship when r = +1 (indicates a positive direction) or r = -1

(indicates a negative direction), 3) The relationship between variables is weak/slight when

r is close to 0. The interpretation of the degree of correlation tightness (r) is shown in

Table 1.

Table 1 Guidelines for the Degree of Correlation Coefficient Relationship

Correlation Value

Relationship Level

0,00 – 0,20

Very Weak

0,20 – 0,40

Weak

0,40 – 0,60

Medium

0,60 – 0,80

Strong

0,80 – 1,00

Very Strong

Source : Sugiyono (2018)

Results and Discussion

An instrument first tested the questionnaire data in the form of a validity and

reliability test using the Eviews program to measure the validity or validity of a question

and check the consistency of the measuring tool from each questionnaire with the

provision that if the results > the table, the variable was declared valid and reliable with

a table of 0.2387. The validity test results are presented in Table 2, and the reliability test

results are presented in Table 3. Table 2 shows that the results in each question item in

each variable are greater than the table, so all questions in the questionnaire from each

variable are declared valid.

Table 2 Validity Test Results

Question

No.

hasil

table

Information

Validity of

Knowledge

Attitude

Validity

Validity of

Supervision

0,2948

0,39428

0,285351

0,2387

Valid

0,3355

0,28142

0,25031

0,308

0,30981

0,264446

0,24

0,54373

0,357268

0,2827

0,37443

0,353945

0,2732

0,43715

0,371693

0,2827

0,25571

0,295686

0,308

0,35761

0,306511

0,308

0,44268

0,37246

0,2472

0,28899

0,30308

0,3355

0,26278

0,331634

0,293

0,27017

0,283143

Ivana Ardhia Larasati, Rika Ernawati, Shofa Rijalul Haq, Nurkhamim, Tedy Agung Cahyadi

Jurnal Indonesia Social Teknologi, Vol. 5, No. 12, December 2024 5840

0,3626

0,31172

0,386074

0,2809

0,3182

0,280156

0,3643

0,27548

0,374446

Source: processed data, 2024

Table 3 shows the reliability coefficient (Alpha) for the three variables of the study:

knowledge (0.2618), attitude (0.3986), and supervision (0.2884), where all the values of

the Alpha coefficients of the three variables are more than the table, which means that the

three variables can be trusted to be used as a data collection tool or to measure a

predetermined object, in this case, to measure the degree of correlation between

independent variables. Although these values are relatively low, all three variables are

still categorized as "Reliable," indicating that the measurement instrument has sufficient

consistency in assessing these variables. Attitude values show better consistency than

knowledge and supervision.

Table 3 Reliability Test Results,

No.

Variable

Reliability Coefficient

(Alpha)

Information

Knowledge

0,2618

Reliable

Attitude

0,3986

Reliable

Supervision

0,2884

Reliable

Source: processed data, 2024

After the data was declared valid and reliable, a classical assumption test in the

form of a normality test and a heteroskedasticity test was carried out using the Eviews

program to produce valid parameter values. The Jarque-Bera method was used in the

Eviews program to perform a normality test. The normality test aims to test whether the

regression model of the distributed variable is normal or not, with the provision that if the

probability value > 0.05, the data is normally distributed. The data results showed that the

probability value of the knowledge variable was 0.139390, the attitude variable was

0.117604, and the supervision variable was 0.131498, where all the Jarque-Bera

probability values > 0.05, meaning that all variables were normally distributed. The data

results are presented in Table 4.

Table 4. Results of the Jarque-Bera Normality Test

Source: processed data, 2024

The heteroskedasticity test aims to test whether, in the regression model, there is an

inequality of variance from residual from one observation to another. A good regression

model is that heteroskedasticity does not occur. The scatterplot graph method and the

White method generated from the output of the Eviews program can be used to detect the

presence or absence of heteroskedasticity. In the scatterplot method, if the image shows

that the dots are randomly spread and scattered both above and below the number 0 on

the Y axis, then it can be concluded that there is no heteroskedasticity. In the White

Jarque-Bera Test

Knowledge

Attitude

Supervision

Probability

0.139390

0.117604

0.131498

a. Test distribution is Normal.

Knowledge, Attitude, and Supervision Correlation Analysis in Improving Work Safety Using The Pearson

Method at A Tin Ore Refining Company in The Jelitik Industrial Area of Bangka Regency

Jurnal Indonesia Social Teknologi, Vol. x, No. x, Month 2024 5841

method, if the probability value of the model shown in Obs*R-squared > 0.05, then it can

be concluded that there is no heterokedastition. The results of the heterokedastition test

of the scatterplot method are shown in Figures 1 – 3, and the table of heterokedasticity

test results of the White method is shown in Table 5. The test results showed that there

was no heteroskedasticity in both methods.

0 1 2 3 4 5 6 7 8

Figure 1. Results of Heteroskedasticity Test of

Knowledge Variable Scatterplot Method

0 1 2 3 4 5 6 7 8

Figure 2. Results of the Heteroskedasticity Test of the Scatterplot

Method of Attitude Variables

Ivana Ardhia Larasati, Rika Ernawati, Shofa Rijalul Haq, Nurkhamim, Tedy Agung Cahyadi

Jurnal Indonesia Social Teknologi, Vol. 5, No. 12, December 2024 5842

0 1 2 3 4 5 6 7 8

Figure 3. Results of Heteroskedasticity Test of Surveillance

Variable Scatterplot Method

Table 5 shows the results of heteroscedasticity analysis using the White method on

the Eviews program for all three variables, showing that there is no heteroscedasticity

problem in the research data. For the knowledge variable, the Obs*R-squared value of

0.0586 with a probability greater than 0.05 indicates that the data does not experience

heteroskedasticity. The attitude variable also produced an Obs*R-squared value of 0.5580

with a probability above 0.05, indicating that the data on this variable did not show

heteroskedasticity. Finally, for the monitoring variable, the Obs*R-squared value of

0.9685, which is greater than 0.05, indicates that the data also did not experience

heteroskedasticity. Thus, all variables in this study meet the assumption of

homoskemasiness or do not experience heteroskedasticity.

Table 5. Results of the White Method Heteroskedasticity Test,

a. Knowledge

Heteroskedasticity Test: White

F-statistic

2.957865

Prob. F(2,65)

0.0590

Obs*R-squared

5.672502

Prob. Chi-Square(2)

0.0586

Scaled explained SS

3.786962

Prob. Chi-Square(2)

0.1505

Source: processed data, 2024

b. Attitude

Heteroskedasticity Test: White

F-statistic

0.567366

Prob. F(2,65)

0.5698

Obs*R-squared

1.166735

Prob. Chi-Square(2)

0.5580

Scaled explained SS

0.628904

Prob. Chi-Square(2)

0.7302

Source: processed data, 2024

Knowledge, Attitude, and Supervision Correlation Analysis in Improving Work Safety Using The Pearson

Method at A Tin Ore Refining Company in The Jelitik Industrial Area of Bangka Regency

Jurnal Indonesia Social Teknologi, Vol. x, No. x, Month 2024 5843

c. Supervision

Heteroskedasticity Test: White

F-statistic

0.044451

Prob. F(2,65)

0.9566

Obs*R-squared

0.092879

Prob. Chi-Square(2)

0.9546

Scaled explained SS

0.064063

Prob. Chi-Square(2)

0.9685

Source: processed data, 2024

After the data was declared valid, reliable, and normal, and no heteroskedasticity

occurred, the Pearson correlation test was carried out. Data analysis was carried out using

the Pearson Product Moment Correlation Test to test the relationship between

independent variables, namely knowledge and attitude, knowledge and supervision, and

attitude with supervision, in the engineering department of a tin ore refining company.

The results of the analysis are presented in Table 6.

Table 6. Results of Analysis of Relationships Between Independent Variables

Variable

Relationship

Correlation

Value

Correlation

Description

Relationship

Level

Knowledge –

Attitude

0.311946

Weak

Knowledge –

Supervision

0.502067

Medium

Attitude –

Supervision

0.484784

Medium

Source: processed data, 2024

In Table 7 above, the results of Pearson's correlation analysis show that there is a

correlation between the independent variable and the varying degree of closeness in the

Engineering Department of PT Pemurnian Oreh Timah in the Jelitik Industrial Estate,

where the correlation between knowledge and attitude (r = 0.311) and the moderate

correlation between knowledge and supervision (r = 0.502), as well as attitude and

supervision (r = 0.484). These findings suggest that knowledge and attitudes are important

factors in improving occupational safety, although the relationship between knowledge

and attitudes is relatively weak. The moderate relationship between knowledge and

supervision, as well as attitude and supervision, shows that effective supervision can

improve workers' knowledge and attitudes toward occupational safety. Knowledge is

influenced by internal factors such as education, interests, experience, and age

(Notoatmodjo, 2012), as well as external factors such as economy, environment, and

culture. Attitude, as a determinant of behavior, is influenced by personal experiences,

culture, important people, mass media, educational and religious institutions, as well as

emotional factors (Arifuddin et al., 2023; Zuchdi, 1995). Effective supervision, as the

process of monitoring employees' work activities to ensure that the company remains on

track towards achieving its goals and making corrections if necessary (Mone et al., 2018;

Siagian, 2003), can help in increasing workers' knowledge and attitudes towards

occupational safety, as well as encouraging them to implement correct occupational

safety practices.

Ivana Ardhia Larasati, Rika Ernawati, Shofa Rijalul Haq, Nurkhamim, Tedy Agung Cahyadi

Jurnal Indonesia Social Teknologi, Vol. 5, No. 12, December 2024 5844

Conclusion

Based on the findings and analysis, this study demonstrates a relationship between

knowledge, attitudes, and supervision in the context of occupational safety. While the

correlation between knowledge and attitude is weak, the moderate correlations between

knowledge and supervision, as well as between attitude and supervision, highlight the

pivotal role of effective supervision in enhancing workplace safety. These results suggest

that targeted interventions focusing on supervision can significantly improve workers'

knowledge and attitudes towards safety. For example, implementing regular training

sessions guided by supervisors can strengthen workers’ understanding of safety protocols.

Additionally, fostering a culture of continuous feedback and monitoring ensures

adherence to safety measures, thereby reducing accident risks. To improve occupational

safety in the mining sector, it is essential to integrate these findings into comprehensive

safety management practices. This includes creating tailored supervision strategies that

align with workers' educational and cultural backgrounds, deploying advanced

monitoring technologies to enhance supervision, and establishing policies that encourage

proactive safety behavior.

Bibliography

Arifuddin, A., Zuchdi, D., Rosana, D., Arovah, N. I., Setiawan, C., Prihatni, Y., Setiawan,

A., Nur, A. F., Dyastuti, N. E., & Arifuddin, H. (2023). Strengthening of early

children’s character education stunting children in Indonesia. Journal of Education

and Health Promotion, 12(1). https://doi.org/10.4103/jehp.jehp_1857_22

Bird, F. E., & Germain, G. L. (1990). Loss control: Safety, health, and environmental

management. International Labour Office.

Friera, A., Syahrani, D., Sitanggang, D., Indriani, S., & Silaban, P. S. (2024). Konsep

Dasar dan Penerapan Pengolahan Data Menggunakan Aplikasi Eviews 10.

MANTAP: Journal of Management Accounting, Tax and Production, 2(1), 52–60.

https://doi.org/10.57235/mantap.v2i1.1521

International Labour Organization (ILO). (1998). Occupational safety and health

management systems. International Labour Office.

Mondy, R. W., & Noe, R. M. (2005). Human Resources Management (Edisi ke-9).

Prentice Hall.

Mone, E., London, M., & Mone, E. M. (2018). Employee engagement through effective

performance management: A practical guide for managers. Routledge.

Notoatmodjo. (2012). Promosi Kesehatan dan Perilaku Kesehatan. Rineka Cipta.

Pemerintah Pusat. (1970). Undang-undang (UU) Nomor 1 Tahun 1970 tentang

Keselamatan Kerja.

Knowledge, Attitude, and Supervision Correlation Analysis in Improving Work Safety Using The Pearson

Method at A Tin Ore Refining Company in The Jelitik Industrial Area of Bangka Regency

Jurnal Indonesia Social Teknologi, Vol. x, No. x, Month 2024 5845

Sari, T. T., & Luhur, P. N. (2020). Motivasi Memediasi Pengaruh Pelatihan dan

Lingkungan Kerja Terhadap Kinerja Karyawan PT. Telkom Witel Yogyakarta.

Kajian Bisnis Sekolah Tinggi Ilmu Ekonomi Widya Wiwaha, 28(1), 71–88.

https://doi.org/10.32477/jkb.v28i1.41

Siagian, S. P. (2003). Manajemen Sumber Daya Manusia. PT Gramedia Pustaka Utama.

Sugiyono. (2015). Metode Penelitian Pendidikan pendekatan Kuantitaif, Kualitatif, dan

R&D. Alfabeta.

Sugiyono. (2018). Metodologi Penelitian Kuantitif. CV. Alfabeth.

Usman, U., & Abbas, T. (2022). Pelatihan Pengolahan Data Menggunakan Aplikasi

Eviews Pada Himpunan Mahasiswa Prodi Ekonomi Pembangunan. Jurnal

Pengabdian Kreativitas (JPeK), 1(2), 24. https://doi.org/10.29103/jpek.v1i2.9390

Winardim, D. (2004). Psikologi Industri dan Organisasi. PT Gramedia Pustaka Utama.

Zuchdi, D. (1995). Psikologi Kepribadian. PT Gramedia Pustaka Utama.