Non Probability Samples
Also: Sampling at Service Sites (SSS)
 |
Non-probability sampling differs from probability sampling in that it does not use random selection. The latter ensures that all units being studied (i.e., members of a population) have an equal opportunity of being selected for participation. While often considered the most rigorous sampling method, probability samples may not be feasible in certain research contexts due to limited time, resources or other constraints.
Non-probability samples can be useful in such circumstances and fall into two main categories: convenience and purposive sampling. Purposive sampling is more commonly used and involves targeting social groups, experts or key informants. Results of non-probability techniques are limited, however, as samples are not representative and confidence intervals cannot be generated.
Sampling at Service Sites (SSS) is an adaptation of non-probability sampling for measuring maternal mortality. Developed by Immpact, it samples respondents at service or other sites (such as antenatal care sites, child health services, or at market places) and asks sisterhood method questions. SSS uses high coverage sites, data on respondent characteristics and reliance of reporting on siblings to overcome some biases.
Identification of death
Adult respondent reports death using either the indirect or direct Sisterhood Methods.
Ascertainment of maternal/pregnancy related status
Adult respondent reports deaths (and sex and age at death) among all his/her brothers and sisters born to the same mother in response to sibling survivorship questions and direct sisterhood method questions.
Advantages:
- can be used in a range of settings with diverse populations
- a cost-efficient means of greatly increasing the sample, thus enabling more frequent measurement of pregnancy-related mortality
- requires less time for data collection and analysis
Limitations:
- does not generate representative samples
- confidence intervals cannot be calculated
- potential selection biases in respondents need to be assessed
- interviews must be kept reasonably short because respondents are busy
- incentives may be needed for respondents in markets in lieu of time taken for interview
- potential for repeat capture of the same respondent
Measurement requirements:
Data as required by use of the indirect or direct sisterhood methods. See above.
 |
 |
 |
 |
SSS Tool: Calculator for number of survey respondents required |
|