Analisis Meta Regresi Untuk Menjelaskan Heterogenitas Hasil Beberapa Penelitian Pada Faktor Ventilasi Dan Pencahayaan Yang Mempengaruhi Tuberkulosis Di Indonesia

Khalimah, Nely Hidayatul (2015) Analisis Meta Regresi Untuk Menjelaskan Heterogenitas Hasil Beberapa Penelitian Pada Faktor Ventilasi Dan Pencahayaan Yang Mempengaruhi Tuberkulosis Di Indonesia. Undergraduate thesis, Institut Technology Sepuluh Nopember.

[img]
Preview
Text
1311100024-Undergraduate Thesis.pdf - Published Version

Download (1MB) | Preview

Abstract

Angka kematian tuberkulosis di Indonesia tergolong tinggi dan sering dikaitkan dengan faktor ventilasi dan pencahayaan. Beberapa penelitian tuberkulosis yang dipengaruhi faktor ventilasi dan pencahayaan melaporkan hasil yang bertolak belakang. Terdapat penelitian yang melaporkan ventilasi dan pencahayaan berpengaruh tinggi terhadap TB, sedangkan penelitian lain melaporkan pengaruh rendah terhadap TB, sehingga perlu mengkaji ulang penelitianpenelitian tersebut untuk mendapatkan effect size yang tepat dan mengestimasi varians diantara seluruh penelitian. Selain itu, hasil penelitian yang berbeda memungkinkan dapat ditelusuri penyebabnya. Variabel yang diduga dapat menjelaskan hasil penelitian yang berbeda adalah pengaruh ventilasi, pengaruh pencahayaan, lokasi, desain dan tahun penelitian. Hasil uji homogenitas pada meta analisis menunjukkan varians antar penelitian heterogen, sehingga model random effect meta analisis lebih tepat digunakan. Heterogenitas antar penelitian harus ditelusuri penyebabnya, salah satu metode untuk menelusuri sumber heterogenitas adalah meta regresi. Hasil meta regresi pada ventilasi menunjukkan bahwa variabel pengaruh ventilasi dan desain penelitian dapat menjelaskan heterogenitas diantara effect size, sedangkan meta regresi pada pencahayaan menunjukkan bahwa lokasi,tahun penelitian dan pengaruh intensitas pencahayaan dapat menjelaskan heterogenitas diantara effect size pengamatan. ===================================================================================================== Mortality rate of tuberculosis in Indonesia is high and often associated with factors ventilation and lighting. Some research tuberculosis which is influenced by ventilation and lighting reported contradictory results. There are studies which report that ventilation and lighting give high effect against TB and lighting, while another study reported a lower effect against TB, so need to review this researchs to get the right effect size and estimating variance among all researchs. Besides that, the different results of researchs can be traced cause. Variables that could be expected to explain the different results of researchs is the effect of ventilation, effect of lighting, location, design of study and years of researchs. Test results on a meta-analysis showed that homogeneity of variance between studies are heterogeneous, so the random effect model of meta-analysis is more appropriate to use. Heterogeneity between studies should be explore the cause, one method to discover the source of heterogeneity is meta regression. Results of meta-regression showed that variable effect of ventilation and design of study can be explain the heterogeneity between effect size, while metaregression on lighting shows that location, years of research and the influence of lighting intensity can be explain the heterogeneity between effect size.

Item Type: Thesis (Undergraduate)
Additional Information: RSSt 519.536 Kha a
Uncontrolled Keywords: Effect size, Heterogenitas, Meta Analisis, Meta Regresi, Pencahayaan, Tuberkulosis, Ventilasi
Subjects: Q Science > QA Mathematics > QA278.2 Regression Analysis
Divisions: Faculty of Mathematics and Science > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User: Mr. Tondo Indra Nyata
Date Deposited: 19 Nov 2018 04:09
Last Modified: 19 Nov 2018 04:09
URI: https://repository.its.ac.id/id/eprint/59970

Actions (login required)

View Item View Item