Font Size: a A A

Empirical Studies On The Health Effect Of NCMS

Posted on:2016-05-17Degree:MasterType:Thesis
Country:ChinaCandidate:F ZhengFull Text:PDF
GTID:2309330461975886Subject:Statistics
Abstract/Summary:PDF Full Text Request
In order to improve the health level of rural residents, guarantee rural residents’ basic medical service, reduce the financial burden caused by disease, chinese government launched a subsidized voluntary health insurance program for rural residents, called New Cooperative Medical System (NCMS) in 2003. Now, more then 10 year pasted, the policy of NCMS has already coverd most residents in rural areas. However, does NCMS really leads to better outcomes, especially for improving the health level of rural residents, the conclusion is inconsistent from scholars at home and abroad.The data of this paper comes from China Health and Retirement Longitudinal Survey (CHARLS). It is consisted of survey data of Zhejiang and Gansu province in 2008 and its tracking data in 2012. Since the target of this paper is to explores the impact of the NCMS on health, the target variable of this paper is health effect. In this paper, we use two ways to measure the health effect, one is subjective health effect namely self-evaluation health, the other is objective health effect namely Activities of daily living (ADL scores). According to lots of references and CHARLS’ survey and data, we select 20 independent variables which including demographic background, household information, medical service utilization, geographical factors and so on.In this paper, we use different models to anylisis different measurements of health effect. To subjective health effect, we chose Ordered-Probit model while to objective ones we chose Propensity Score Matching Difference in Difference(PSMDID). In comparation with binary classification problems, Ordered-Probit model can deal with more classifica- tions, even they are ordered. There are five levels in self-evaluation health, so we chose Ordered-Probit model. Propensity Score Matching method can solve the problem of s-election bias and omitted variable problems. While Difference in Difference method use two years’data to eliminate the influence of time. PSMDID can solve these problems in the same time, so we chose PSMDID to analysis the effect of NCMS on ADL score.This paper provides certain reference value to the research of the effect of NCMS on health. We found that rural residents who take part in NCMS have better self-evaluated health and lower ADL score, which means NCMS has a positive effect on subjective self-evaluation health, as well as objective ADL scores. This means NCMS has certain effect in the view of health effect.
Keywords/Search Tags:NCMS, Health effect, Ordered-Probit model, PSMDID
PDF Full Text Request
Related items