Teleworking—An Economic and Social Impact during COVID-19 Pandemic: A Data Mining Analysis
Abstract
:1. Introduction
2. Literature Review
3. Research Strategy
4. Research Methodology
- questions regarding personal and professional information of the employee;
- questions about home accommodation of the employee for TL;
- the support that the employer has offered to the employee during TL;
- impact of TL over the employee regarding social and professional aspects;
- factors that may change the point of view over the TL after the pandemic will pass;
- work efficiency and satisfaction during TL;
- economic consequences of TL;
- collecting the data from the questionnaire that was sent to employees in order to be filled in with their opinion about TL. In this regard, economic and social information was collected.
- establishing the degree of filling for each attribute and eliminating the empty records or with very few data; By degree of filling is meant how many values an attribute had completed.
- Group A (social)—personal information about the employee, information about last level of studies, geographic region from Romania where he/she lives, age, and sex;
- Group B (social)—information about the employee’s professional activity such as status on labor market, if he/she is employed or not, if he/she performed TL in 2020, salary level, activity domain;
- Group C (economic)—information on home accommodation for teleworking, if any additional investments were made for TL, if any additional expenses were made for TL regarding services like electricity, telephony, internet, water and sanitation, heating and air conditioning and also their total estimation, how much the respondent was informed by the government about the TL regulations;
- Group D—support provided by the employer for teleworking with information about the aspect that identifies if the employer made efforts to facilitate work during TL and if he/she has offered continuously support for employees;
- Group E (economic)—efficiency of teleworking vs. normal activity with information about how the employee has considered the efficiency of the TL over the normal activity;
- Group F—information about the employer (company) regarding the number of employees and if the regulations given by the state had impact on supporting the TL for the employees;
- Group G (economic)—employee and TL in relation to basic activities with information about how the TL is appropriate for the employee to carry out main activities;
- Group H (economic)—consequences of teleworking with information about the additional hours allocated for TL and the variation of the income;
- Group I (social)—the impact of teleworking from a social point of view with information about communication efficiency during TL, main advantages and disadvantages of TL, main problems identified by the employees;
- Group J—factors ranking with information about several types of factors (like cultural, economic, natural, psychological, social) that are most affected by the TL;
- Group K (social)—satisfaction degree with information about if the employee felt satisfaction during TL;
- Group L—impact of TL after the pandemic with information about the fact if the employee is willing to continue in TL after the pandemic is gone or he/she wants to return to the office for full time;
- establishing the degree of filling for each attribute;
- ranking the attributes related to a specific class attribute: GainRatioAttributeEval algorithm as attribute evaluator was used together with Ranker search methods; a ranking score was obtained for each attribute used, based on a class attribute; the scores are offered in a decreasing order and the attribute with the greatest score is the most important (or relevant) to the class attribute; Some algorithms (e.g. “GainRatioAttributeEval”, “PART”, “J48”) are required to obtain intermediate results. They were mentioned to make the analysis more transparent.
- cluster algorithms: Simple K-Means to determine the characteristics for each group for a certain number of employees group. Before using the Simple K-Means algorithm, the EM (expectation maximization) algorithm was used in order to identify the most appropriate number of groups (clusters); The EM algorithm was used to identify the number of clusters in the dataset. Then the Simple K-Means algorithm was used for cluster analysis.
- classification algorithms: The PART algorithm for generating a decision rules list and the J48 algorithm to obtain profiles for employees using classification decisions trees. Both algorithms are related to a certain class attribute.
5. The Modeling Phase and the Main Results
5.1. Clustering Approach
5.1.1. First Cluster Experiment
- the job particularities and its activities and how well these can be carried out in a TL way;
- how much the home of the employee can be transformed into a working place or office and the accommodation degree;
- the employer and how much effort it puts into facilitating the working activities of the employee;
- the state and the laws that it gives in order to help and sustain the employer during this TL period of time;
- the implication of the employer in offering a continuously support to the employers during the hole teleworking period.
- have graduated a faculty as last level of study.
- are between 26 and 40 years old.
- have a salary amount between 2501 and 4000 RON (equivalent of 500 and 800 Euro);
- they work in the economic domain in a company that has between 50 and 250 employees;
- have made a lot of initial investments between 501 and 1000 RON (equivalent of 100 and 200 Euro) for TL home accommodation;
- have had additional expenses per month between 100 and 500 RON (equivalent of 20 and 100 Euro), this being for electricity, telephony, internet, water and sanitation, heating and air condition;
- home accommodation was done properly;
- the employer made effort to facilitate the teleworking and to offer continuously support in order to sustain it;
- the regulations offered by the state have a high impact in helping the employers.
- have graduated a high school as last level of study;
- are between 18 and 26 years old;
- have a salary amount between 1300 and 2500 RON (equivalent of 260 and 500 Euro);
- they work in the economic domain in a company that has between 10 and 49 employees;
- have not made any initial investments for TL home accommodation
- did not have any additional expenses or very few per month, under 100 RON (equivalent of 20 Euro), this being for electricity, water and sanitation, heating and air condition;
- home accommodation was done properly;
- the employer made effort to facilitate the teleworking and to offer continuously support in order to sustain it;
- the regulations offered by the state have a high impact in helping the employers.
- have graduated a faculty as last level of study;
- are between 40 and 55 years old;
- have a salary amount between 2501 and 4000 RON (equivalent of 500 and 800 Euro);
- they work in services domain in a company that has between 10 and 49 employees;
- have not made any initial investments for teleworking home accommodation;
- did not have any additional expenses or very few per month, under 100 RON (equivalent of 20 Euro), this being only for electricity;
- home accommodation was not done properly;
- the employer made no effort to facilitate the teleworking and to offer continuously support in order to sustain it;
- the regulations offered by the state have a high impact in helping the employers.
5.1.2. Second Cluster Experiment
- the efficiency that the employee felt he had during the TL;
- if the employee felt that the communication efficiency was influenced by the TL;
- the income level during the pandemic being in TL;
- the employee identifies disadvantages and problems for teleworking;
- if the employee must work additional hours.
- the work efficiency is best in TL versus normal working way;
- teleworking does not imply additional working hours;
- the communication efficiency was improved by the teleworking;
- the work satisfaction was high during teleworking;
- the main advantage is considered to be the time saved that was spent in traffic;
- the main disadvantage of TL is considered to be the lack of human interaction;
- the communication level has decreased between team members and this it is considered to be an important problem in order to develop daily activities and tasks;
- the income increased during the TL.
- the work efficiency varies based on the activity type;
- teleworking does imply additional working hours;
- the communication efficiency was not improved by the teleworking;
- the work satisfaction was average during teleworking;
- the main advantage is considered to be the time saved that was spent in traffic;
- the main disadvantage of teleworking is considered to be the lack of human interaction;
- the lack of the direct contact between employees and customers it is considered to be an important problem;
- the income remained the same during the teleworking.
- the income remained the same during the teleworking.
- the work efficiency is best during the normal working way;
- teleworking does not imply additional working hours;
- the communication efficiency was not improved by the teleworking;
- the work satisfaction was average during teleworking;
- the main advantage is considered to be the time saved that was spent in traffic;
- the main disadvantage of teleworking is considered to be the lack of human interaction;
- the communication level has decreased between team members and this it is considered to be an important problem in order to develop daily activities and tasks;
- the income remained the same during the teleworking.
5.1.3. Third Cluster Experiment
- the work efficiency is best during teleworking;
- the work satisfaction is average;
- main advantage is the saved time that was spent in traffic;
- main disadvantage is the lack of human interaction;
- the order of the factors that can influence the point of view over the teleworking is: economic, social, psychological, cultural and natural;
- activity domain is in Services.
- the work efficiency is good only for certain activities;
- the work satisfaction is average;
- main advantage is the saved time that was spent in traffic;
- main disadvantage is the lack of human interaction;
- the order of the factors that can influence the point of view over the teleworking is: social, economic, psychological, cultural and natural;
- activity domain is in Economic.
- the work efficiency is no good at all in TL and they want to come back to the normal working way as soon as possible;
- the work satisfaction is average in TL;
- main advantage is the saved time that was spent in traffic;
- main disadvantage is the lack of human interaction;
- the order of the factors that can influence the point of view over the TL is: psychological, social, economic, natural and cultural;
- activity domain is Economic.
5.2. Classification Approach
5.2.1. First Classification Experiment
5.2.2. Second Classification Experiment
6. Findings and Discussion
- the employees that have considered having the best work efficiency: they work in the economic domain, in which many activities may be carried out in TL; they made additional investments to organize a space from home into an office (between 100 and 200 Euro) and had monthly additional expenses (between 20 and 100 Euro). They also received support from the employer in facilitating TL and from the state based on the adopted regulations. So, for this situation, all the three actors made efforts to improve the work efficiency—the employee made investments to transform his/her home, the employer offered support and the state has created the appropriate legal framework for TL.
- the employees that have considered not having a good work efficiency at all and wanted to come back to the office work in the Services domain, in which only some activities may be performed in a proper manner in TL; they have not made any additional investments for home accommodation and the monthly additional expenses are low (under 20 Euro). In this situation, even if they consider that the state has created a good legal framework for TL, they consider also that the employer made no effort to help them and also they were not open to transform the home into an office and to create the right accommodation.
- the employees that have considered having the highest work satisfaction: have the best work efficiency in TL, are able to communicate in an efficient manner with their colleagues and the distance in not a problem to them and, also their income has increased and no additional hours were made. So, in this category we may conclude that there are employees that were able to adapt quickly, to understand the situation and to overcome easily the challenges imposed by the TL. However, we may also say that their domain was not affected too much by the pandemic or it was developed even more (like the IT domain).
- the employees that have considered having an average work satisfaction and wanted to return to the office as soon as possible: they consider the office work the best manner to do the job and to accomplish the activities, the TL had a negative effect for the communication between colleagues, and the income remained the same.
- the employees that will want to continue working in TL are the ones that are considered to have the best efficiency working in TL and also consider that the TL is affecting the society most of all from an economic and social point of view. In this group, we may consider the employees that are more flexible and adaptable, are working in domains in which many activities can be done in TL, succeed in maintaining a good collaboration and communication with their colleagues. However, they are aware about the economic and social problems that are arising from a TL applied for a long period of time.
- the employees that will not want to continue working in TL are the ones that are considered to have the best efficiency working from the office and also consider that the TL is affecting the society most of all from a psychological and social point of view. In this group, we may consider the employees that are less adaptable, are working in domains in which few activities can be done in TL, and do not succeed in maintaining a good communication with their colleagues. However, they have identified some psychological and social problems that are arising from a TL applied for a long period of time.
7. Conclusions
- for the first cluster experiment that identifies the relationship between the work efficiency that employees felt that they had during TL and certain social and economic factors—first, the employees that have considered having the best work efficiency, in which case the involvement of all the actors is mandatory at a high level (the employee made investment to transform his/her home, the employer has offered support and the state has created the appropriate legal framework for TL); second, the employees that have considered not having a good work efficiency at all, in which case even they consider that the state has created a good legal framework for TL, they consider also that the employer made no effort to help them and also they were not open to transform their home into an office and to create the right accommodation.
- for the second cluster experiment that identifies the relationship between the work satisfaction degree that employees felt during TL and certain social and economic factors—the employees that have considered having the highest work satisfaction: first, the employees that have considered to have the best work efficiency in TL, in which case they were able to adapt quickly, to understand the situation and to overcome easily the challenges imposed by the TL; second, the employees that have considered having an average work satisfaction and want to return to office as soon as possible, in which case they have considered the office work the best manner to do the job and to accomplish the activities, the TL having a negative effect for the communication between colleagues. What the two groups have in common is that they have considered as the main advantage the time saved from staying in traffic and as main disadvantage the lack of human interaction.
- for the third cluster experiment that identifies the relationship between the impact of teleworking after the pandemic and several factors like works satisfaction and efficiency and some others factors point of view—first, the employees that will want to continue working in TL have considered to have the best efficiency working in TL and also have considered that the TL is affecting the society most of all from an economic and social point of view, the employees from this group are more flexible and adaptable, being able to maintain a good collaboration and communication with their colleagues; second, the employees that will not want to continue working in TL are those that have considered to have the best efficiency working from the office and that the TL is affecting the society most of all from a psychological and social point of view, the employees from this group are less adaptable, are working in domains in which few activities can be done in TL and they are not able to maintain a good communication with their colleagues.
- for the first classification experiment that offers a detailed analysis of the clusters identified based on the work efficiency in the first cluster experiment and highlights the most common profiles of the employees—analyzing the most different two profiles confirms the fact that the employees that had the best work efficiency during TL were the ones that were willing to spend more money and to make additional investments and expenses. The other employees, for which the work-efficiency was not good at all, have encountered problems with the employer, not receiving any help from it during the TL, the specific activities were not suitable for TL, and did not make any additional expenses.
- for the second classification experiment that offers a more detailed analysis of the clusters identified based on the work satisfaction in the second cluster experiment and highlights the most common profiles of the employees—analyzing the different two profiles it is confirmed the fact that the employees that had the best work satisfaction during TL were the ones that have very high work efficiency, TL had a positive influence over the social communication efficiency and did not need any additional hours to complete their activities. The other employees with a low work satisfaction have considered that the office work is the best way to accomplish their tasks and the social communication was affected in a negative way during TL
8. Future Development
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Conflicts of Interest
Appendix A
Group ID | Attribute Group | Attributes (Factors) |
---|---|---|
A | Personal information about the employee (social) | 23_A_Last_level_of_studies {High school, Bachelor, Master, PhD} 26_A_Geographic_region {Bucuresti_Ilfov, Sud_Vest_Oltenia, Nord_Est, Sud_Muntenia, Sud_Est, Centru, Vest, Nord_Vest} 27_A_Age {between_18_and_26, between_26_and_40, between_40_and_55, over_55} 28_A_Sex {M, F} |
B | Information about the employee’s professional activity (social) | 01_B_Status_on_labor_market {Employed_at_a_public_institution, Employed_at_a_private_company, Entrepreneur, Others} 02_B_Employed {Full_time, Part_time, Others} 03_B_TLW_in_2020 {Yes_Full_time, Yes_Part_time, No} 22_B_Salary_level {under_1300_RON, between_1300_and_2500_RON, between_2501_and_4000_RON, between_4001_and_5500_RON, over_5500_RON} 24_B_Activity_domain {Agriculture_and_forestry, Constructions, Economic, HR, Industry, Insurance, IT, Others, Services_Defense, Services_Education, Services_Financial_banking, Services_Medical, Services_Others, Services_Public_administration } |
C | Home accommodation for teleworking (economic) | 06_C_Home_accomodation_for_TLW {Yes, No} 07_C_Additional_investments_for_TLW {Yes, No} 071_C_Amount_of_the_additional_investments_for_TLW {no_investment, no_more_than_500_RON, between_501_and_1000_RON, between_1001_and_2000_RON, over_2000_RON} 08_C_Additional_expenses_for_TLW {Yes_quite_a_lot, Not_at_all_or_very_few} 0811_C_Additional_expenses_ELECTRICITY {Yes, No} 0812_C_Additional_expenses_TELEPHONY {Yes, No} 0813_C_Additional_expenses_INTERNET {Yes, No} 0814_C_Additional_expenses_WATER_and_SANITATION {Yes, No} 0815_C_Additional_expenses_HEATING_and_AIR_CONDITIONING {Yes, No} 082_C_Estimated_expenses_generated_by_TLW {under_100_RON, between_100_and_500_RON, between_501_and_1000_RON, over_1000_RON} 15_C_Information_degree_about_government_regulations_on_TLW {1,2,3,4,5} |
D | Support provided by the employer for teleworking | 05_D_Employer_effort_to_facilitate_TLW {Yes, No, I_can_not_say} 17_ D_Employer_continuously_support_for_TLW {Sustained, Not_sustained, I_can_not_say } |
E | Efficiency of teleworking vs. normal activity (economic) | 04_E_Efficiency_TLW_vs_normal_activity {TLW_best, NormActiv_best, Varies_based_on_activity } |
F | Information about the employer/company | 18_F_Regulations_impact_supporting_TLW_on_your_employer {Very_high_impact, High_impact, Low_impact, Very_low_impact, I_cannot_appreciate} 25_F_Company_number_of_employees {under_10, between_10_and_49, between_50_and_250, between_251_and_1000, between_1001_and_5000, over_5000} |
G | Employee—telework in relation to basic activities (economic) | 21_G_TLW_appropiate_in_doing_the_main_activities {All_the_main_activities, Only_part_of_the_main_activities, No} |
H | Consequences of teleworking (economic) | 09_H_TLW_means_also_additional_working_hours {Yes, No} 19_H_Income_level_during_the_pandemic {Increased, Decreased, Remained_the_same, Varied} |
I | The impact of teleworking (social) | 10_I_TLW_pozitiv_influence_over_the_communication_efficiency {Yes, No} 12_I_Main_advantage_of_TLW { Additional_free_time_obtained, Efficient_management_of_working_hours, Financial_savings, Improving_work_life_balance, Others, Saving_time_spent_in_traffic} 13_I_Disadvantages_of_TLW {Yes, No, I_can_not_say} 131_I_Main_disadvantage_of_TLW {Limited_access_to_resources, Lack_of_communication_between_employees, Lack_of_concentration, Lack_of_human_interaction, Decreased_productivity, More_allocated_resources, Decreased_salaries} 14_I_Problem_of_TLW { Time_delimitation_for_work_and_personal_problems, Limited_access_to_documents, Lack_of_direct_contact_employees_customers, Protection_of_confidential_data, Risk_of_not_advancing_in_the_career, Decreasing_the_communication_level_between_team_members, Others} |
J | Factors ranking | 161_J_Factors_influence_POV_about_TLW_1st_place {cultural, economic, natural, psychological, social} 162_J_Factors_influence_POV_about_TLW_2nd_place {cultural, economic, natural, psychological, social} 163_J_Factors_influence_POV_about_TLW_3rd_place {cultural, economic, natural, psychological, social} 164_J_Factors_influence_POV_about_TLW_4th_place {cultural, economic, natural, psychological, social} 165_J_Factors_influence_POV_about_TLW_5th_place {cultural, economic, natural, psychological, social} |
K | Satisfaction degree (social) | 11_K_Satisfaction_degree_for_TLW_activity {Low, Average, High} |
L | Impact of teleworking after the pandemic | 20_L_TLW_after_the_pandemic {Yes_full_time, Yes_only_certain_activities, No} |
1. What is your status on the labor market? | Count |
a. Employed in a public institution | 119 |
b. Employed in a private firm/company | 228 |
c. Entrepreneur | 6 |
d. Other status (like liberal professions, etc.) | 24 |
377 |
2. Are you employed: | Count |
a. full time | 284 |
b. part time | 86 |
c. other | 7 |
377 |
4. Do you consider that you performed the telework activity with the same efficiency as at the office? | Count |
a. work efficiency is the best during TLW | 105 |
b. work efficiency varies based on the activity type | 154 |
c. work efficiency is the best during normal activity | 118 |
377 |
5. Do you consider that your employer has made a sustained effort to facilitate your telework activity? | Count |
a. Yes | 193 |
b. No | 104 |
c. I cannot say | 80 |
377 | |
22. What is your salary level (net salary in RON)? | Count |
a. under 1300 | 29 |
b. 1.300–2.500 | 106 |
c. 2.501–4.000 | 137 |
c. 4.001–5.500 | 57 |
d. over 5.500 | 48 |
377 | |
23. What is the last level of studies completed? | Count |
a. High school | 141 |
b. Bachelor | 139 |
c. Master studies | 93 |
d. PhD studies | 4 |
377 | |
24. What is the field in which you work/carry out activity? | Count |
a. Agriculture, hunting, forestry | 18 |
b. Constructions | 14 |
c. Economic | 91 |
d. Human Resources | 7 |
e. Industry | 27 |
f. Insurance | 18 |
g. IT | 36 |
h. Defense service | 3 |
i. Education service | 29 |
j. Financial and banking services | 35 |
k. Medical service | 4 |
l. Public administration service | 4 |
m. Other services | 75 |
n. Other | 16 |
377 | |
25. What is the number of employees in your company? | Count |
a. less than 10 | 52 |
b. 10–49 | 109 |
c. 50–250 | 106 |
d. 251–1000 | 61 |
e. 1001–5000 | 30 |
f. over 5000 | 19 |
377 | |
26. Region of development you come from: | Count |
a. Bucharest—Ilfov | 171 |
b. South-West Oltenia | 15 |
c. North-East | 26 |
d. South—Muntenia | 101 |
e. South-East | 47 |
f. Centre | 14 |
g. West | 2 |
h. North-West | 1 |
377 | |
27. What age group do you belong to? | Count |
a. 18–25 years | 125 |
b. 26–40 years | 114 |
c. 41–55 years | 128 |
d. over 55 years | 10 |
377 | |
28. What is your gender? | Count |
a. Male | 126 |
b. Female | 251 |
377 |
References
- Lister, K.; Harnish, T. Telework and its effects in the United States. In Telework in the 21st Century; Edward Elgar Publishing: Cheltenham, UK, 2019. [Google Scholar]
- Gaffney, A.W.; Himmelstein, D.U.; Woolhandler, S. Trends and Disparities in Teleworking During the COVID-19 Pandemic in the USA: May 2020–February 2021. J. Gen. Intern. Med. 2021, 36, 3647–3649. [Google Scholar] [CrossRef]
- Georgescu, S. Munca de la distanță devine noua normă pe piața muncii/TL Becomes the New Rule on the Labor Market. Available online: https://www.rfi.ro/social-125987-studiu-pwc-munca-distanta-parte-normalitate (accessed on 29 June 2021).
- Doherty, L.C.; Andrey, T.S.; Johnson, J.C. The Economic and Social Impacts of Telework, Telework, The New Workplace of the 21st Century. 2000. Available online: https://play.google.com/books/reader?id=_QrtAAAAMAAJ&pg=GBS.PA206&hl=ro&output=text&lr=&printsec=frontcover (accessed on 5 June 2021).
- Radulescu, C.V.; Ladaru, G.R.; Burlacu, S.; Constantin, F.; Ioanăș, C.; Petre, I.L. Impact of the COVID-19 Pandemic on the Romanian Labor Market. Sustainability 2021, 13, 271. [Google Scholar] [CrossRef]
- Stanworth, C. Telework and the information age. New Technol. Work Employ. 1998, 13, 51–62. [Google Scholar] [CrossRef]
- Azzalini, B.; Scarpa, A. Data Analysis and Data Mining: An Introduction; OUP USA: Oxford, UK, 2012. [Google Scholar]
- Lin, N.; Cercone, T.Y. (Eds.) Rough Sets and Data Mining: Analysis of Imprecise Data; Springer Science & Business Media: Berlin/Heidelberg, Germany, 2012. [Google Scholar]
- Peña-Ayala, A. Educational data mining: A survey and a data mining-based analysis of recent works. Expert Syst. Appl. 2014, 41, 1432–1462. [Google Scholar] [CrossRef]
- Chen, Y.L.; Weng, C.H. Mining fuzzy association rules from questionnaire data. Knowl.-Based Syst. 2009, 22, 46–56. [Google Scholar] [CrossRef]
- Matsumoto, T.; Sunayama, W.; Hatanaka, Y.; Ogohara, K. Data Analysis Support by Combining Data Mining and Text Mining. In Proceedings of the 2017 6th IIAI International Congress on Advanced Applied Informatics (IIAI-AAI), Hamamatsu, Japan, 9-13 July 2017; pp. 313–318. [Google Scholar] [CrossRef]
- Zhang, S.; Zhang, C.; Yang, Q. Data preparation for data mining. Appl. Artif. Intell. 2003, 17, 375–381. [Google Scholar] [CrossRef]
- Hirji, K.K. Exploring Data Mining Implementation. Commun. ACM 2001, 44, 87–93. [Google Scholar] [CrossRef]
- Wahyudi, I.; Tiswiyanti, W.; Nusifera, S. Business Sustainability in the Pandemic COVID-19: Study at MSMES in Jambi City. In The 3rd Green Development International Conference (GDIC 2020); Atlantis Press: Amsterdam, The Netherlands, 2021; Volume 205, pp. 363–366. [Google Scholar]
- Profiroiu, M.-C.; Radulescu, C.V.; Burlacu, S.; Guţu, C. Changes and trends in the development of the world economy. In Competitivitatea şi inovarea în economia cunoaşterii. Ediţia a 22-a, 25–26 septembrie 2020, Chişinău. Chişinău Republica Moldova: Centrul Editorial-Poligrafic al ASEM; 2020; pp. 324–330. ISBN 978-9975-75-985-4. Available online: https://ibn.idsi.md/sites/default/files/imag_file/324-330_4.pdf (accessed on 1 October 2021).
- Contreras, F.; Baykal, E.; Abid, G. E-Leadership and Teleworking in Times of COVID-19 and Beyond: What We Know and Where Do We Go. Front. Psychol. 2020, 11, 1–11. [Google Scholar] [CrossRef]
- Vrchota, J.; Maříková, M.; Řehoř, P. Teleworking in small and medium enterprises (Smes) before the onset of coronavirus crisis in the czech republic. Management 2020, 25, 151–164. [Google Scholar] [CrossRef]
- Dewi, A.; Nurmandi, A.; Rochmawati, E.; Purnomo, E.P.; Rizqi, M.D.; Azzahra, A.; Benedictos, S.; Suardi, W.; Dewi, D.T.K. Global policy responses to the COVID-19 pandemic: Proportionate adaptation and policy experimentation: A study of country policy response variation to the COVID-19 pandemic. Heal. Promot. Perspect. 2021, 10, 359–365. [Google Scholar] [CrossRef]
- Omrani, H.; Modroiu, M.; Lenzi, J.; Omrani, B.; Said, Z.; Suhrcke, M.; Tchicaya, A.; Nguyen, N.; Parmentier, B. COVID-19 in Europe: Dataset at a sub-national level. Data Br. 2021, 35, 106939. [Google Scholar] [CrossRef] [PubMed]
- Burlacu, S.; Patarlageanu, S.R.; Diaconu, A.; Ciobanu, G. E-government in the Era of Globalization and the Health Crisis caused by the Covid-19 Pandemic, between Standards and Innovation. SHS Web Conf. 2021, 92, 08004. [Google Scholar] [CrossRef]
- Ammons, S.K.; Markham, W.T. Working at hdome: Experiences of skilled white collar workers. Sociol. Spectr. 2004, 24, 191–238. [Google Scholar] [CrossRef]
- Redman, T.; Snape, E.; Ashurst, C. Location, location, location: Does place of work really matter? Br. J. Manag. 2009, 20, S171–S181. [Google Scholar] [CrossRef] [Green Version]
- Sarbu, R.; Alpopi, C.; Burlacu, S.; Diaconu, S. Sustainable Urban Development in the Context of Globalization and the Health Crisis Caused by the Covid-19 Pandemic. SHS Web Conf. 2021, 92, 01043. [Google Scholar] [CrossRef]
- Bailey, D.E.; Kurland, N.B. A review of telework research: Findings, new directions, and lessons for the study of modern work. J. Organ. Behav. 2002, 23, 383–400. [Google Scholar] [CrossRef]
- Favilukis, J.Y.; Lin, X.; Sharifkhani, A.; Zhao, X. Labor Force Telework Flexibility and Asset Prices: Evidence from the COVID-19 Pandemic. SSRN Electron. J. 2020, 20, S171–S181. [Google Scholar] [CrossRef]
- Katsabian, T. The Telework Virus: How the COVID-19 Pandemic Has Affected Telework and Exposed Its Implications for Privacy and Equality. SSRN Electron. J. 2020, 1–56. [Google Scholar] [CrossRef]
- Liu, S.; Su, Y. The impact of the COVID-19 pandemic on the demand for density: Evidence from the U.S. housing market. Econ. Lett. 2021, 207, 110010. [Google Scholar] [CrossRef]
- Buomprisco, G.; Ricci, S.; Perri, R.; de Sio, S. Health and Telework: New Challenges after COVID-19 Pandemic. Eur. J. Environ. Public Heal. 2021, 5, em0073. [Google Scholar] [CrossRef]
- Schur, L.A.; Ameri, M.; Kruse, D. Telework After COVID: A ‘Silver Lining’ for Workers with Disabilities? J. Occup. Rehabil. 2020, 30, 521–536. [Google Scholar] [CrossRef]
- Nguyen, M.H. Factors influencing Home-Based Telework in Hanoi (Vietnam) during and after the COVID-19 Era; Springer: Berlin/Heidelberg, Germany, 2021; Volume 48. [Google Scholar]
- Coggon, D.; Croft, P.; Cullinan, P.; Williams, A. Assessment of workers’ personal vulnerability to covid-19 using ‘covid-age’. Occup. Med. 2020, 70, 461–464. [Google Scholar] [CrossRef]
- Kikuchi, S.; Kitao, S.; Mikoshiba, M. Who suffers from the COVID-19 shocks? Labor market heterogeneity and welfare consequences in Japan. J. Jpn. Int. Econ. 2021, 59, 101117. [Google Scholar] [CrossRef]
- Cretan, G.C.G.; Gherghina, R.; Duca, I.; Postole, M.A.; Constantinescu, C.M. Determinants Of Employees’ Option For Preserving Teleworking After The Covid-19 Pandemic. Amfiteatru Econ. 2021, 23, 669–682. [Google Scholar] [CrossRef]
- Feinstein, R.E.; Kotara, S.; Jones, B.; Shanor, D.; Nemeroff, C.B. A health care workers mental health crisis line in the age of COVID-19. Depress. Anxiety 2020, 37, 822–826. [Google Scholar] [CrossRef] [PubMed]
- Barbieri, T.; Basso, G.; Scicchitano, S. Italian Workers at Risk During the COVID-19 Epidemic. Ital. Econ. J. 2021. [Google Scholar] [CrossRef]
- Shi, X.; Moudon, A.V.; Lee, B.H.Y.; Shen, Q.; Ban, X.J. Factors Influencing Teleworking Productivity—A Natural Experiment during the COVID-19 Pandemic. Findings 2020, 18195. [Google Scholar] [CrossRef]
- Martin, L.; Nguyen-Thi, U.T.; Mothe, C. Human resource practices, perceived employability and turnover intention: Does age matter? Appl. Econ. 2021, 53, 3306–3320. [Google Scholar] [CrossRef]
- Andargie, M.S.; Touchie, M.; O’Brien, W. Case study: A survey of perceived noise in Canadian multi-unit residential buildings to study long-term implications for widespread teleworking. Build. Acoust. 2021, 28, 443–460. [Google Scholar] [CrossRef]
- Kamath, P.; Patil, P.; Shrilatha, S.; Sowmya, S. Crop yield forecasting using data mining. Glob. Transitions Proc. 2021, 2, 402–407. [Google Scholar] [CrossRef]
- Bouckaert, D.S.; Remco, R.; Eibe, F.; Hall, M.; Kirkby, R.; Reutemann, P.; Seewald, A. WEKA Manualfor Version 3-6-2; University of Waikato: Hamilton, New Zealand, 2010. [Google Scholar]
- Bodea, C.-N.; Bodea, V.; Rosca, I.G.; Mogos, R.; Dascalu, M.-I. Student Performance in E-Learning Environments; IGI Global: Hershey, PA, USA, 2011. [Google Scholar]
- Kawashima, T.; Nomura, S.; Tanoue, Y.; Yoneoka, D.; Eguchi, A.; Shi, S.; Miyata, H. The relationship between fever rate and telework implementation as a social distancing measure against the COVID-19 pandemic in Japan. Public Health 2021, 192, 12–14. [Google Scholar] [CrossRef]
- Golden, T.D. Telework and the Navigation of Work-Home Boundarie. Organ. Dyn. 2021, 50, 100822. [Google Scholar] [CrossRef]
- Sunhare, P.; Chowdhary, R.R.; Chattopadhyay, M.K. Internet of things and data mining: An application oriented survey. J. King Saud Univ. Comput. Inf. Sci. 2020. [Google Scholar] [CrossRef]
No. | Analysis | Groups of Used Attributes | Class Attribute | Results of the Analysis and Research Question to Answer |
---|---|---|---|---|
I | Factors that are influencing the work efficiency during TL | A, B, C, D, F, G | E | a. Obtaining a score for the most important social and economic attributes related to the class attribute = > attribute ranking (Q1) b. Clustering—grouping the employees based on similarities => clusters for the employees (Q1) c. Classification analysis based on cluster assignment => employees’ profiles related to the group attributes (Q4) |
II | Factors that are influencing the work satisfaction during TL | H, I, E | K | a. Obtaining a score for the most important social and economic attributes related to the class attribute = > attribute ranking (Q2) b. Clustering—grouping the employees based on similarities => clusters for the employees (Q2) c. Classification analysis based on cluster assignment => employees’ profiles related to the group attributes (Q5) |
III | Analysis of factors that may change the mentality about TL taking based on the work satisfaction and efficiency during TL | B (activity domain), K, E, J, I (main advantage and main disadvantage) | L | a. Clustering—grouping the employees based on similarities => clusters for the employees (Q3) |
No. | Data Mining Approaches | Results of the Analysis and Algorithms Used |
---|---|---|
I | Clustering (3 experiments) | Obtaining a score for the most important social and economic attributes related to the class attribute => attribute ranking (Gain Ration Attribute Evaluator and Ranker Search Method) |
Clustering—grouping the employees based on similarities => clusters for the employees (Simple K-Means algorithm) | ||
II | Classification (2 experiments) | Classification analysis based on cluster assignment and a class attribute => employees’ profiles related to the group attributes (PART and J48 algorithms) |
Attribute | Full Data | Cluster 0 | Cluster 1 | Cluster 2 |
---|---|---|---|---|
Instances number: 377 | 165 | 127 | 85 | |
Status on labor market (group B) | Employed at a private company | Employed at a private company | Employed at a private company | Employed at a private company |
Employed (group B) | Full time | Full time | Full time | Full time |
Efficiency TLW vs. normal activity (group E) | Varies based on activity | TL best | Varies based on activity | Normal activityis the best |
Employer effort to facilitate TLW (group D) | Yes | Yes | Yes | No |
Home accommodation for TLW (group C) | Yes | Yes | Yes | No |
Additional investments for TLW (group C) | No | Yes | No | No |
Amount of the additional investments for TLW (group C) | No investment | Between 501 and 1000 RON | No investment | No investment |
Additional expenses for TLW (group C) | Yes, quite a lot | Yes, quite a lot | Not at all or very few | Not at all or very few |
Additional expenses ELECTRICITY (group C) | Yes | Yes | Yes | Yes |
Additional expenses TELEPHONY (group C) | No | Yes | No | No |
Additional expenses INTERNET (group C) | No | Yes | No | No |
Additional expenses WATER and SANITATION (group C) | Yes | Yes | Yes | No |
Additional expenses HEATING and AIR CONDITIONING (group C) | Yes | Yes | Yes | No |
Estimated expenses generated by TLW (group C) | Under 100 RON | Between 100 and 500 RON | Under 100 RON | Under 100 RON |
Information degree about government regulations on TLW (group C) | 3 | 3 | 4 | 3 |
Employer continuously support for TLW (group D) | Sustained | Sustained | Sustained | Not sustained |
Regulations impact supporting TLW on your employer (group F) | High impact | High impact | High impact | High impact |
TLW appropriate in doing the main activities (group G) | Only part of the main activities | Only part of the main activities | Only part of the main activities | No |
Salary level (group B) | Between 2501 and 4000 RON | Between 2501 and 4000 RON | Between 1300 and 2500 RON | Between 2501 and 4000 RON |
Last level of studies (group A) | High school | Bachelor | High school | Bachelor |
Activity domain (group B) | Economic | Economic | Economic | Services |
Company number of employees (group F) | Between 10 and 49 | Between 50 and 250 | Between 10 and 49 | Between 10 and 49 |
Geographic region (group A) | Bucuresti Ilfov | Bucuresti Ilfov | Sud Muntenia | Bucuresti Ilfov |
Age (group A) | Between 40 and 55 | Between 26 and 40 | Between 18 and 26 | Between 40 and 55 |
Sex (group A) | F | F | F | F |
Attribute | Full Data | Cluster 0 | Cluster 1 | Cluster 2 |
---|---|---|---|---|
Instances number: 377 | 100 | 149 | 128 | |
Efficiency TLW vs. normal activity (Group E) | Varies based on activity | TLW has the best work efficiency | Varies based on activity | Normal activity has the best work efficiency |
TLW means also additional working hours (Group H) | No | No | Yes | No |
TLW positive influence over the communication efficiency (Group I) | No | Yes | No | No |
Satisfaction degree for TLW activity (Group K) | Average | High | Average | Average |
Main advantage of TLW (Group I) | Saving time spent in traffic | Saving time spent in traffic | Saving time spent in traffic | Saving time spent in traffic |
Disadvantages of TLW (Group I) | Yes | Yes | Yes | Yes |
Main disadvantage of TLW (Group I) | Lack of human interaction | Lack of human interaction | Lack of human interaction | Lack of human interaction |
Problem of TLW (Group I) | Decreasing the communication level between team members | Decreasing the communication level between team members | Lack of direct contact employees customers | Decreasing the communication level between team members |
Income level during the pandemic (Group H) | Remained the same | Increased | Remained the same | Remained the same |
Attribute | Full Data | Cluster 0 | Cluster 1 | Cluster 2 |
---|---|---|---|---|
Instances number: 377 | 142 (38%) | 144 (38%) | 91 (24%) | |
Efficiency TLW vs. normal activity (Group E) | Varies based on activity | TLW efficiency is better | Varies based on activity type | Normal activity efficiency is better |
Satisfaction degree for TLW activity (Group K) | Average | Average | Average | Average |
Main advantage of TLW (Group I) | Saving time spent in traffic | Saving time spent in traffic | Saving time spent in traffic | Saving time spent in traffic |
Main disadvantage of TLW (Group I) | Lack of human interaction | Lack of human interaction | Lack of human interaction | Lack of human interaction |
Factors influence POV about TLW 1st place (Group J) | social | economic | social | psychological |
Factors influence POV about TLW 2nd place (Group J) | social | social | economic | social |
Factors influence POV about TLW 3rd place (Group J) | economic | psychological | psychological | economic |
Factors influence POV about TLW 4th place (Group J) | cultural | cultural | cultural | natural |
Factors influence POV about TLW 5th place (Group J) | natural | natural | natural | cultural |
Activity domain (Group B) | Economic | Other services | Economic | Economic |
TLW after the pandemic (Group L) | Yes but only for certain activities | Yes full time | Yes but only for certain activities | No |
a | b | c | <-- Classified as |
---|---|---|---|
150 | 11 | 4 | a = cluster0 |
11 | 98 | 18 | b = cluster1 |
9 | 22 | 54 | c = cluster2 |
TP Rate | FP Rate | Precision | Recall | F-Measure | ROC Area | Class | |
---|---|---|---|---|---|---|---|
0.909 | 0.094 | 0.882 | 0.909 | 0.896 | 0.918 | cluster0 | |
0.772 | 0.132 | 0.748 | 0.772 | 0.76 | 0.838 | cluster1 | |
0.635 | 0.075 | 0.711 | 0.635 | 0.671 | 0.755 | cluster2 | |
Weighted Avg. | 0.801 | 0.103 | 0.798 | 0.801 | 0.799 | 0.855 |
a | b | c | <-- Classified as |
---|---|---|---|
83 | 9 | 8 | a = cluster0 |
7 | 139 | 3 | b = cluster1 |
4 | 3 | 121 | c = cluster2 |
TP Rate | FP Rate | Precision | Recall | F-Measure | ROC Area | Class | |
---|---|---|---|---|---|---|---|
0.83 | 0.04 | 0.883 | 0.83 | 0.856 | 0.91 | cluster0 | |
0.933 | 0.053 | 0.921 | 0.933 | 0.927 | 0.955 | cluster1 | |
0.945 | 0.044 | 0.917 | 0.945 | 0.931 | 0.978 | cluster2 | |
Weighted Avg. | 0.91 | 0.046 | 0.909 | 0.91 | 0.909 | 0.951 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Belostecinic, G.; Mogoș, R.I.; Popescu, M.L.; Burlacu, S.; Rădulescu, C.V.; Bodislav, D.A.; Bran, F.; Oancea-Negescu, M.D. Teleworking—An Economic and Social Impact during COVID-19 Pandemic: A Data Mining Analysis. Int. J. Environ. Res. Public Health 2022, 19, 298. https://doi.org/10.3390/ijerph19010298
Belostecinic G, Mogoș RI, Popescu ML, Burlacu S, Rădulescu CV, Bodislav DA, Bran F, Oancea-Negescu MD. Teleworking—An Economic and Social Impact during COVID-19 Pandemic: A Data Mining Analysis. International Journal of Environmental Research and Public Health. 2022; 19(1):298. https://doi.org/10.3390/ijerph19010298
Chicago/Turabian StyleBelostecinic, Grigore, Radu Ioan Mogoș, Maria Loredana Popescu, Sorin Burlacu, Carmen Valentina Rădulescu, Dumitru Alexandru Bodislav, Florina Bran, and Mihaela Diana Oancea-Negescu. 2022. "Teleworking—An Economic and Social Impact during COVID-19 Pandemic: A Data Mining Analysis" International Journal of Environmental Research and Public Health 19, no. 1: 298. https://doi.org/10.3390/ijerph19010298