Track20 is built around using all available data to better inform monitoring and planning. A key focus of this work is to make greater use of service statistics- this includes working to improve the quality of service statistics and to improve how the data is used to monitor progress.

Improving Estimates of Population Indicators

Reliance on large-scale surveys to track FP program results leaves many countries in a difficult situation with regard to developing annual estimates of FP2020 core indicators and other indicators appropriately measured at the population level. In the absence of direct measurements from annual surveys, many countries are compelled to project trends in key indicators (e.g., mCPR, Unmet Need) since the last large-scale survey into the future until data from the next large-scale survey become available.

Track20, in collaboration with UNPD & Amherst has developed a methodology/tool (FPET) for developing annual estimates of key family planning indicators: CPR, mCPR, Unmet Need and Demand Satisfied. In an effort to provide better estimates in the periods between surveys, Service Statistics can be used in the model to inform trends and improve projections.

How it works

FPET uses population level values from representative surveys to inform estimates of CPR, mCPR, Unmet Need and Demand Satisfied. While we cannot produce these values using service statistics, we can develop a proxy for mCPR based on service statistics that can be used in FPET. We call this proxy the "EMU" or Estimated Method Use. Using service statistics in conjunction with population data and CYP factors, we develop an estimate of the number of women using a modern method in a given year and compare that value and trend to available survey values for mCPR.

If the service statistics data meets a series of criteria, we develop the EMU, and incorporate the data into our FPET estimates to produce more accurate/up to date estimates and projections of key population indicators.

Requirements for using Service Statistics in FPET
  • Consistent levels of reporting over time
  • At least 3 years of consistent data
  • For at least one year, overlap between Service Statistics and a survey
  • At least one year of Service Statistics reported after the most recent survey