Table 2:

Household survey samples in 2012 and 2015

Sample characteristicGombe State, Nigeria*EthiopiaUttar Pradesh, India
201220152012201520122015
No. of clusters408040408080
Households
 Target no. to be surveyed§200060002000300060006000
 No. surveyed184459392118300052586466
 Response rate, %929910610088108
Resident women
 No. of women age 13–49 yr listed271881252153348086419446
 No. interviewed202174191934317081208522
 Response rate, %74**9190919490
 With birth in 12 mo preceding survey 3491100277404604584
  • * Sample selected from the 11 local government areas of Gombe State where the Society for Family Health was in operation. For more detail, see Appendix 1 (available at www.cmaj.ca/lookup/suppl/doi:10.1503/cmaj.190219/-/DC1).

  • Sample selected from 59 woreda (implementation districts) across 4 regions of Ethiopia (Oromia, Tigray, Amhara and Southern Nations Nationalities and Peoples), where a subset of activities by the Last 10 Kilometers project was in operation. For more detail, see Appendix 1.

  • Sample selected from 6 districts of Uttar Pradesh (Jhansi, Hardoi, CSM Nagar, Maharanjganj, Sultanpur and Raebarailly), where the Community Mobilisation Project was in operation. For more detail, see Appendix 1.

  • § Each cluster was a village or enumeration area that was segmented into units of 50 households (Gombe in 2012; Ethiopia in 2012) or 75 households (Gombe in 2015; Ethiopia in 2015; Uttar Pradesh in 2012 and 2015), with 1 segment being selected at random for the study. The protocol was to interview all household residents within the selected segment. In Ethiopia in 2012 and Uttar Pradesh in 2015, a larger number of households was available in the selected segment than had been anticipated, which made the response rate appear to be greater than 100%.

  • No women aged 13 or 14 years reported a recent live birth.

  • ** In Gombe State, the 2012 survey coincided with a period of unrest due to insurgency in the region, which may have affected the response rate that year.