Probability/Non-Probability Sampling Designs
Probability sampling designs are those designs that adapt the aspect of randomization; thus, giving every object in the target population an equal chance of taking part in the study (Lund Research, 2012). Unfortunately, probability sampling is not ideal in this study of illegal immigrants because their records are not available. Examples of probability sampling designs include:
- Simple Random Sampling. This method makes use of random numbers using random tables; alternatively, they are generated from a computer. The use of random numbers is an easier way of attaining the desired number of the sample in comparison to the ballot method which is tedious and time-wasting. This is deemed the gold standard of probability sampling.
- Sequential Sampling. This kind of sampling entails selecting every nth member of the sample population based on the sampling interval. It is also known as systematic random sampling; it utilizes randomization in the selection of the first subject.
- Stratified Sampling. This method of sampling prevails when the studied population contains subgroups that are significant in a study. For example, a study that seeks to investigate the effect of socioeconomic status on health might utilize this kind of sampling.
- Cluster Sampling. This sampling method is employed when it is not feasible to draw individual subjects from the general population because the targeted population exists as clusters.
Non-probability on the other hand is conducted when probability sampling is not plausible. This is usually the case when no list of the sampling units is available, for example, in the case of homelessness. Thereby, it becomes arduous to employ probability methods of sampling because of the inability to utilize randomization. The following are the examples of non-probability sampling as discussed by Babbie (2011):
- Reliance on Available Subjects. This is also called convenience or haphazard sampling and it entails selecting a sample based on the ease of getting the sample and its frugality. Unfortunately, this method is not generalizable unless the researcher (s) is sure that the sampled population is an actual representation of the entire population. This method is not feasible in the study of illegal immigrants because they will not readily accept that they are illegal immigrants and any indication that there is a search to identify them would render the search futile; the immigrants would hide.
- Snowball Sampling. This is the sampling design that would be ideal for the study of illegal immigrants in the US. It is mainly used to sample members in a population that is hard to find and recruit for the purposes of research. Usually, the researcher identifies just one member of the cohort the researcher wishes to study and he or she helps in identifying other members of the cohort (Weathington, Cunningham, & Pittenger, 2010).
Other non-probability sampling methods include judgmental, purposive, quota, and self-selection.
Suitable Sampling Design for this Study
In reference to the discussion indicated above, the snowball sampling method remains the ideal choice of sampling for this study of illegal immigrants. Even though it is a difficult process trying to identify the first illegal immigrant, once identified, identifying subsequent immigrants will become easier.
Sampling Strategy
The recommended kind of sampling design is non-probability sampling based on reasons stated earlier. This research should outline clear objectives in the informed consent. In addition, the researcher should be guaranteed of his or her protection of research rights and that personal information would be stored in a secure and password-protected file to ensure no unauthorized party accesses the file. The identification of the first illegal immigrant will be made possible by networking and soliciting referrals from various individuals. However, discretion would be highly observed because the illegal immigrants are wary of being identified. I am under the assumption that illegal immigrants can discreetly identify each other and especially if they have a similar national or cultural background. Thus, the identification of the first immigrant will automatically lead to the other immigrants.
Selection of Sample
As it is, illegal immigrants live in the US in fear of being caught and deported back home. As a result, there is no list of these immigrants and it becomes difficult to apply probability sampling methods in this study. In alignment with Vander Ven (2015, p. 142) view on the labeling theory, “treating people like criminals has the unintended consequence of making them more criminal.” Therefore, labeling illegal immigrants as criminals and failing to give them a chance in society results in their engagement in these illegal acts. Whereas Vander Ven (2015) exercised ethnography, the subjects were not aware they were being observed, Therefore, I propose more ethical and valid research that seeks to obtain the consent of the research participants while ensuring their rights are observed.
I will identify the neighborhoods of individuals that are not of American origin and establish rapport with various individuals until I am able to identify one immigrant. It is not advisable to consider every person engaging in illegal acts as an illegal immigrant; thus, I will not be misguided by such a prejudiced view. Upon identification of the first illegal immigrant, I will verify this bit by asking him or her if he or she possesses any legal document warranting his stay in the US. He or she will refer more individuals without legal documents that ought to warrant their stay in the US. However, the identification of these illegal immigrants will not be used as a reason for their victimization.
References
Babbie, E. R. (2011). The Basics of Social Research (5th ed.). Belmont, CA: Wadsworth.
Lund Research. (2012). Sampling Strategy. Web.
Vander Ven, T. M. (2015). Fear of victimization and the interactional construction of harassment in a Latino neighborhood. In M. G. Maxfield & E. R. Babbie, Research Methods for Criminal Justice and Criminology (7th ed.). Stamford, CT: Cengage Learning.
Weathington, B.L., Cunningham, C. J., & Pittenger, D. J. (2010). Research methods for the behavioral and social sciences. Hoboken, NJ: John Wiley & Sons, Inc.