Sampling Size Sampling Strategy and Sample Size

Sampling Size

Sampling Strategy and Sample Size for a Qualitative Research Plan

A grounded theory perspective on the victimization of women and human trafficking

The formulas for determining appropriate sampling sizes for quantitative research studies are fairly formalized, given the need to make generalizations about the macro-level population based upon the findings of these experimental or quasi-experimental studies. However, the purpose of qualitative studies is to come to conclusions based upon the researcher's inductive findings regarding a specific group of people. For an ethnographic study, the 'sample size' of the population might constitute a very small tribe. A narrative qualitative study might involve discussing the findings of studying a small class of children. The purpose of qualitative studies is to let the experiences of individuals speak for themselves, whether this takes the form of a case study, a narrative of a specific handful of participants, or simply 'making meaning' from a culture. Although I would hope that my grounded theory study on women who had been trafficked into the sex trade or economic slavery would yield findings that would be helpful in giving assistance to such women in the future and contribute to my reader's general understanding of the topic, my ultimate purpose is still to let these women speak, and the theory will derive from their experiences. My hypothesis as a researcher should not drive and shape the findings.

In general, with quantitative research, larger sampling sizes are preferable, given the 'law of large numbers' or the idea that large samples are less prone to exhibit errors because of anomalies. But "there is a point of diminishing return to a qualitative sample -- as the study goes on more data does not necessarily lead to more information. This is because one occurrence of a piece of data, or a code, is all that is necessary to ensure that it becomes part of the analysis framework" (Mason 2010:1). The point of qualitative research is to embark up on the intensive study of specific phenomenon, not to create broad generalizable conclusions. However, it is also true that it is not necessarily the case that 'smaller is better,' either. "Qualitative samples must be large enough to assure that most or all of the perceptions that might be important are uncovered" (Mason 2010:1).

The principle of sampling size in qualitative sampling is one of saturation, or the concept that new data "does not shed any further light on the issue under investigation" which will be determined by "heterogeneity of the population; the number of selection criteria; the extent to which 'nesting' of criteria is needed; groups of special interest that require intensive study; multiple samples within one study; types of data collection methods use; and the budget and resources available" (Mason 2010: 1-1.1). In the case of my research study, the population of women who are frequently victims of trafficking is heterogeneous to some degree in that women from many different nations may be trafficked and there may be 'special' groups which deserve additional consideration, such as women who are trafficked into the sex trade vs. domestic and economic servitude or women who are compelled to remain with their abductors through physical violence and threats of physical violence vs. women who remain with their abductors primarily due to psychological intimidation.

By and large, most women who are trafficked come from poorer, developing world nations. Most are duped into their situation with promises of a better opportunity than exist in their homeland. But there are other factors to consider when creating a sampling size, namely: "the scope of the study, the nature of the topic, the quality of the data, the study design" the scope of the study, the nature of the topic, the quality of the data, the study design" (Mason 2010: 1.1). For example, my study will by virtue of necessity limited to women who have been saved from their situation and who are willing to talk to a researcher. That itself will be a limit upon the population I can study. It is true that the women I study may come from many different nations and may have different experiences of the physical process of trafficking itself based upon their location and a different cultural worldview of their experience. National assumptions regarding gender, the role of women in society, the role of women in relationship to men and the family, and a host of other psychological factors will influence the women's experiences. However, I will be limited by the quality and type of translators I can obtain to interpret the women's words, given that they are unlikely to be able to speak English fluently. Thus, my study population may also be limited simply by virtue of who is willing to speak with me and whom I feel I am able to conduct a competent interview of, based upon logistical constraints.

It should be noted that some experts in the field of qualitative research do give very specific numbers as to what they consider to be a reasonably-sized population for the purposes of drawing inductive conclusions. For ethnography and ethnoscience: 30-60 participants are suggested; for phenomenology 5-25. For grounded theory (my selected method) Creswell suggests 20-30 while Morse suggests 20-50 (Creswell 1998: 64; Morse 1994: 225, both cited by Mason 2010:1.2). Presumably the demand for larger numbers of grounded theory is rooted in the fact that a theory is supposed to be derived from the evidence, requiring a deeper and more expansive survey of possible experiences. However, the range from 20-50 is fairly broad.

Ideally, if I was able to find fifty women of different nationalities and different experiences and competent translators for the women in their native languages, I would prefer this sampling size. But realistically, I know I am more likely to have to confine my analysis from anywhere from 10-25, depending upon the number of women who are willing to talk to me and the logistical complications which arise regarding translation. However, I will explain the limits of the study in the introduction. This is one of the values of qualitative research -- because it does not attempt to come to a conclusion using the experimental method, but invests a great deal of trust in the reader's ability to draw effective conclusions based upon the evidence, a researcher can 'bracket' the problem of coming up with a large enough sample size and ask the reader to interpret the results in light of such limitations -- the researcher does not have to discard the entire purpose and intent of the study, simply because the sample size is not ideal.