Efficiency of Respondent Driven Sampling to Study Street Children

Rapid social, economic and cultural changes along with intentionally or unintentionally consequences has provided the incidence and spread of social pathologies such as street children in large cities. Probability and non-probability sampling methods that are used to study this data are faced many problems such as coverage error and invalid statistical inferences, respectively.
Respondent driven sampling method is an effective method to penetrate the hidden and hard to reach layers of the target population. The main objective of this paper was to introduce this method, obtain unbiased proportion estimates, as well as evaluate effectiveness of this method in a study of street children in Tirana, Albania (2010).
According to the results of this survey, success of respondent driven sampling method for studying street children lies on knowing the exact size of social networks for these children, availability of survey sites, reaching to equilibrium in interested variables and selecting appropriate incentives.
Accomplishing accurate findings of street children without the presence of a representative sample would not be possible and it would not be achieved unless researchers use effective sampling methods to study these populations. Indeed driven sampling is recommended.