Electronic Theses and Dissertation

Mobilitas penduduk adalah salah satu faktor utama yang memengaruhi jumlah kasus terkonfirmasi COVID-19. ‘Google Mobility Report’ menyediakan data terbuka yang mencatat pergerakan penduduk di 6 area yaitu area retail dan rekreasi, toko bahan makanan dan apotik, taman, area transportasi umum, tempat kerja, dan area permukiman. Sementara itu ‘Bing COVID-19’ dari Microsoft menyediakan data kasus terkonfirmasi sampai level sub-nasional (provinsi). Penelitian ini bertujuan memahami pengaruh faktor pergerakan penduduk pada kasus terkonfirmasi COVID-19 di Indonesia dengan model regresi data panel. Penelitian ini menggunakan 34 provinsi di Indonesia sebagai unit cross-section. Sementara unit time series berupa data harian mobilitas penduduk dan kasus terkonfirmasi COVID-19 dari 1 April 2020 sampai 30 September 2022. Untuk melihat hubungan yang lebih jelas, unit time series ini selanjutnya dipisahkan ke dalam 3 periode yaitu masa sebelum program vaksinasi COVID-19, setelah diterapkannya program vaksinasi COVID-19, dan sesudah tercapainya herd immunity. Terdapat tiga model estimasi regresi data panel yaitu Common Effect Model (CEM), Fixed Effect Model (FEM) dan Random Effect Model (REM). Pemilihan model terbaik dilakukan dengan uji Chow, uji Hausman dan uji Lagrange Multiplier (LM). Uji signifikansi parameter dilakukan menggunakan uji serentak, uji parsial dan melihat nilai Koefisien Determinasi (R2). Hasil yang diperoleh menunjukkan bahwa model yang tepat yaitu Random Effect Model (REM) untuk data pada ketiga periode. Hasil pemodelan menunjukkan mobilitas memang berpengaruh nyata pada naik-turunnya kasus COVID-19 di Indonesia. Mobilitas di area retail dan rekreasi, toko bahan makanan dan apotek, pusat transportasi umum, dan area permukiman berpengaruh positif dan signifikan terhadap kasus terkonfirmasi COVID-19 pada periode sebelum vaksinasi. Sesudah diterapkannya program vaksinasi COVID-19, mobilitas di area toko bahan makanan dan apotek, pusat transportasi umum, dan area permukiman masih berpengaruh berpengaruh positif dan signifikan terhadap kasus terkonfirmasi COVID-19. Sedangkan data setelah tercapainya herd immunity menunjukkan mobilitas tidak lagi sepenuhnya mempengaruhi jumlah kasus terkonfirmasi. Bahkan arah pengaruh di area toko bahan makanan dan apotek, taman, tempat kerja dan area permukiman menjadi negatif terhadap kasus terkonfirmasi COVID-19. Artinya, setelah herd immunity, berapapun besar mobilitas penduduk tidak lagi menambah jumlah kasus. Jumlah kasus terus menurun karena sebagian besar warga telah mendapatkan vaksin. Koefisien determinasi (R2) sebelum dan sesudah program vaksinasi serta setelah tercapai herd immunity berturut-turut sebesar 10%, 15% dan 9%

Population mobility is one of the main factors affecting the number of confirmed cases of COVID-19. The 'Google Mobility Report' provides open data that records population movements in 6 areas, namely retail and recreation areas, grocery stores and pharmacies, parks, public transportation areas, workplaces, and residential areas. Meanwhile 'Bing COVID-19' from Microsoft provides confirmed case data to the sub-national (provincial) level. This study aims to understand the influence of population movement factors on confirmed cases of COVID-19 in Indonesia using a panel data regression model. This study uses 34 provinces in Indonesia as a cross-sectional unit. While the time series unit is in the form of daily data on population mobility and confirmed cases of COVID-19 from 1 April 2020 to 30 September 2022. To see a clearer relationship, this time series unit is then separated into 3 periods, namely the period before the COVID-19 vaccination program, after the implementation the COVID-19 vaccination program, and after supposedly achieving a herd immunity. There are three panel data regression estimation models, namely the Common Effect Model (CEM), Fixed Effect Model (FEM) and Random Effect Model (REM). Selection of the best model is done by Chow test, Hausman test and Lagrange Multiplier (LM) test. Parameter significance test was carried out using simultaneous tests, partial tests and looking at the value of the Coefficient of Determination (R2). The results obtained show that the appropriate model is the Random Effect Model (REM) for data from all three periods. The modeling results show that mobility does have an effect on the rise and fall of COVID-19 cases in Indonesia. Mobility in retail and recreation areas, grocery stores and pharmacies, public transportation centers, and residential areas has a positive and significant effect on confirmed cases of COVID-19 in the pre-vaccination period. After the implementation of the COVID-19 vaccination program, mobility in the area of grocery stores and pharmacies, public transportation center, and residential areas still has a positive and significant effect on the confirmation of COVID-19. Meanwhile, data after the achievement of herd immunity shows that mobility no longer fully affects the number of confirmed cases. Even the direction of influence in grocery stores and pharmacies, garden, workplace and residential areas becomes negative for confirmed cases of COVID-19. This means that after herd immunity, no matter how large the mobility of the population is, it will no longer increase the number of cases. The number of cases continues to decline because most residents have received the vaccine. The coefficient of determination (R2) before and after the vaccination program and after achieving herd immunity were 10%, 15% and 9%.