Activity Number:
|
151
- #LeadwithStatistics in the Social Sciences
|
Type:
|
Topic Contributed
|
Date/Time:
|
Monday, July 30, 2018 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Bayesian Statistical Science
|
Abstract #329737
|
Presentation
|
Title:
|
Estimating Unmet Need for Contraceptive Methods in the World's Poorest Countries
|
Author(s):
|
Leontine Alkema* and Niamh Cahill and Chuchu Wei
|
Companies:
|
University of Massachusetts Amherst and University College Dublin and University of Massachusetts Amherst
|
Keywords:
|
Family planning;
FP2020;
Sustainable Development Goals;
Bayesian hierarchical model;
Data amalgamation;
Non-sampling errors
|
Abstract:
|
Estimates and short-term projections of family planning indicators such as contraceptive use and unmet need for family planning are needed for monitoring and planning by national programs and global initiatives such as the Sustainable Development Goals and FP2020. Data on family planning indicators in the world's poorest countries are available from surveys but these surveys are typically only carried out every three to five years and reporting is subject to non-sampling errors. Family planning service statistics that are produced as a byproduct of service delivery information provide annual data but are subject to bias due to measurement and conversion challenges. We present the Family Planning Estimation Tool (FPET), which is a Bayesian hierarchical time series model that is used to produce estimates and short term projections of family planning indicators for all countries in the world. We highlight recent methodological developments that allow for the inclusion of survey and service statistics data for producing estimates for the world's poorest countries, accounting for bias and non-sampling errors.
|
Authors who are presenting talks have a * after their name.