Non-probability sampling encompasses various methods for selecting participants from a population without ensuring that each individual has a known and equal chance of being included.
Some common types of non-probability sampling are:
- Convenience sampling:
- The researcher selects individuals who are conveniently accessible, often resulting in a sample that may not be representative of the entire population.
- It is quick and cost-effective, suitable for exploratory research or when a complete sampling frame is unavailable.
- Purposive or judgmental sampling:
- Participants are selected based on the researcher’s judgment and specific criteria related to the research objectives.
- It is useful when the goal is to capture specific perspectives or experiences, allowing for targeted sampling.
- Quota sampling:
- The researcher sets quotas for different subgroups and then selects individuals to fill these quotas, often using convenience or purposive methods.
- It provides control over sample composition but lacks true randomness.
- Snowball sampling:
- The sample grows through referrals, with participants referring others who meet the study criteria.
- It helps study hard-to-reach populations or those with specific characteristics but may result in a biased sample.
- Volunteer sampling (self-selection):
- Participants voluntarily choose to participate in the study.
- It is easy to implement but may result in a biased sample as volunteers may differ systematically from those who do not.
- Haphazard or convenience sampling:
- The researcher selects readily available individuals willing to participate, often leading to a convenient sample that may need more representativeness.
- It is quick and easy to implement but may need more generalizability.
- Quasi-random sampling:
- It is a hybrid approach that involves a mix of random and non-random selection.
- It allows for some level of control and randomness but may not meet the criteria for true random sampling.
List of recommended resources #
For a broad overview #
This blog entry on non-probability sampling by Statistics Canada gives an overview of the method of selecting units from a population using a subjective (i.e. non-random) method. It describes some of the commonly used non-probability sampling methods, along with crowdsourcing and web panels.
This article by Kassiani Nikolopoulou on Scribbr gives an overview of non-probability sampling, its various types and examples. It also details the advantages and disadvantages of non-probability sampling as well as differentiates it from the probability sampling method.
For in depth understanding #
This article by Arkadiusz Wiśniowski, Joseph W Sakshaug, Diego Andres Perez Ruiz and Annelies G Blom evaluates supplementing inferences based on small probability samples with prior distributions derived from non-probability data. It further demonstrates that informative priors based on non-probability data can lead to reductions in variances and mean squared errors for linear model coefficients.
This book explores the increasingly scientific endeavor of surveys and their growing complexity, as different information sources and data collection modes are combined. Chapter 22 of this book, written by Vasja Vehovar, Vera Toepoel & Stephanie Steinmetz, deals particularly with non-probability sampling.
Case study #
This report by Andrea Serra presents the results of a study which aims to understand the status of and challenges faced by Mozambican women’s small and medium enterprises (SME´s). The study was carried out through a quantitative survey and focus group discussions among a convenience sample of 70 business women.
This study by Govinda R. Timilsina, Gal Hochman and Iryna Fedets conducts a two-stage quota sample survey of 509 commercial and industrial firms of all regions of Ukraine to understand the barriers to energy efficiency improvements.