KOEKOEH HARDJITO, 090515569M (2007) APLIKASI ANALISIS REGRESI ORDINAL DENGAN PENDEKATAN LINIER DAN NON LINIER (LOGISTIK) PADA DATA KEPUASAN RESPONDEN TERHADAP PELAYANAN PUSKESMAS. Thesis thesis, UNIVERSITAS AIRLANGGA.

Preview	Text (ABSTRAK) gdlhub-gdl-s2-2015-hardjitoko-37421-7.ringk-n.pdf Download (618kB) | Preview
	Text (FULLTEXT) 2.pdf Restricted to Registered users only Download (1MB) | Request a copy

Abstract

This research is applied by using secondary data. From the data obtained are expected ordinal regression models with linear and non-linear approach. The variables discussed in this research consisted of predictor variables are reliability, responsiveness, assurance, empathy and tangible response variable is the level of patient satisfaction Data were analyzed by three approaches: categorical regression, multinomial logit regression and ordinal regression with PLUM. After each analysis approach could reveal significant independent variables were not significant and can be seen and the coefficient of determination (R2). The results showed that the significant independent variable with three approaches are as follows: At PLUM approach there are two significant variables are tangible and assurance. Multinomial approach raises three significant variables that Responsivennes, reliability and assurance, while the categorical approach, there are four significant variables are tangible, responsivennes, empathy, assurance. Assessment R2 use the indicator Cox and Snell, Nagelkerke, and McFadden Of the three smallest values obtained on PLUM approach, the greatest value on the categorical approach.

Actions (login required)

View Item