A Note on Segmented Multinomial Logit Models for the Analysis of Discrete Choice Experiment Data

2020 ◽  
Author(s):  
Henning Schaak
2020 ◽  
Vol 30 (Supplement_5) ◽  
Author(s):  
P Binyaruka

Abstract Background Informal payments are regressive. They can limit the access to quality healthcare, particularly of the most vulnerable, and are potentially catastrophic. Little is known in terms of providers' preferences for interventions. Methods We conducted a cross-sectional discrete choice experiment (DCE) among 432 health providers from 42 public health facilities (hospitals and health centres) in seven districts from Pwani region and five districts from Dar es Salaam region. The DCE attributes were derived from a scoping literature review, qualitative interview from 27 key informants from three districts, and through workshop with health providers, managers and policy makers. The final DCE survey tool included 12 unlabeled choice sets, each describing two hypothetical jobs that varied across six attributes: mode of payment, supervision at facility, opportunity for private practice, awareness and monitoring, measures against informal payment, and incentive payment for lack informal payment in the past 6 months. Multinomial logit and mixed multinomial logit methods were used to estimate preferences for the attributes. Results All attribute-levels, apart from supervision at the facility level, were significantly influencing health providers' choice decisions for job type (p < 0.001). The most preferred attributes were measures for awareness creation and monitoring -i.e. preferences were significantly higher for facility with noticeboard (coefficient 0.39, 95% CI 0.29 -0.48 ), followed by provision of receipts (0.34, 0.24 -0.44) and presence of hotline number for reporting corrupt practices (0.26, 0.17 -0.35). Opportunity for private practice was significantly preferred (0.38, 0.31-045) and job preference increases as salary top-up increases (0.06, 0.05-0.7). The less preferred attributes were cash payment for healthcare (-0.27, -0.35- -0.19) and disciplinary measures at the district (-0.15, -0.23 - -0.07) or facility level (-0.10, -0.17- -0.03).


Data in Brief ◽  
2020 ◽  
Vol 28 ◽  
pp. 105027
Author(s):  
Erik Fruth ◽  
Michele Kvistad ◽  
Joe Marshall ◽  
Lena Pfeifer ◽  
Luisa Rau ◽  
...  

Author(s):  
Nia Kurnia Sholihat ◽  
Masita Wulandari Suryoputri ◽  
Ade Martinus

Even though pharmaceutical care has been proven increasing patients’ quality of life, pharmacists still have barriers to implement it. Our study aims to examine factors affecting pharmacists in the community to implement pharmaceutical care using a Discrete Choice Experiment (DCE). The study was a cross-sectional study. A structured DCE questionnaire was administered to 90 community pharmacists in Banyumas district, Indonesia. Respondents were chosen using a simple random sampling method. According to the literature review and expert opinions, the following six attributes were selected: pharmacists’ confidence; willingness to implement pharmaceutical care; communication skill; knowledge and professional skill; availability of time; and availability of space in pharmacy. Eighteen choice sets were developed. Each choice sets comprised of two scenarios. Respondents were asked to choose the scenario they preferred the most. Data were analyzed using multinomial logit model. Of 90 questionnaires distributed, 67 were analyzed. Based on multinomial logit, all attributes had a significant effect on pharmacists’ preferences to implement pharmaceutical care. The findings suggested that pharmacist association should train their member to increase professional skills, as well as the management of pharmacy should provide enough space to perform pharmaceutical care.


2021 ◽  
Vol 16 (4) ◽  
pp. 279-282
Author(s):  
Stan Lipovetsky

The work presents various techniques of the logistic and multinomial-logit modeling with their modifications. These methods are useful for regression modeling with a binary or categorical outcome, structuring in regression and clustering, singular value decomposition and principal component analysis with positive loadings, and numerous other applications. Particularly, these models are employed in the discrete choice modeling and the best-worst scaling known in applied psychology and socio-economics studies.


Data in Brief ◽  
2021 ◽  
Vol 35 ◽  
pp. 106887
Author(s):  
Fresenbet Zeleke Abshiro ◽  
Girma T. Kassie ◽  
Jema Haji ◽  
Belaineh Legesse

2019 ◽  
Vol 111 (7) ◽  
pp. 1243-1260 ◽  
Author(s):  
Alex Roach ◽  
Bruce K. Christensen ◽  
Elizabeth Rieger

2019 ◽  
Author(s):  
Y Peters ◽  
E van Grinsven ◽  
M van de Haterd ◽  
D van Lankveld ◽  
J Verbakel ◽  
...  

2016 ◽  
Vol 18 (2) ◽  
pp. 155-165 ◽  
Author(s):  
Axel C. Mühlbacher ◽  
John F. P. Bridges ◽  
Susanne Bethge ◽  
Ch.-Markos Dintsios ◽  
Anja Schwalm ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document