But, these authors opine that all types of sampling techniques in qualitative research can be encompassed under a broader term, ‘purposeful sampling’. The authors stated that the qualitative research typically focuses ‘on relatively small samples, even single cases, selected purposefully’ (Patton, 1990, p. 169). Patton (2002) provided 16 different kinds of strategies for selecting information-rich cases. These strategies bring forth the complexity of sampling in qualitative research. The principle underlying these strategies is to select an information-rich case that is the sample/case is selected purposefully to fit with the purpose of the study. Patton did not provide any discussion on theoretical sampling, though some similarities can be found in his conforming and disconfirming cases. Purposeful sampling requires an access to a key informant which becomes the source for other samples. The strategies given by Patton (1990) are discussed below.

·         Extreme or Deviant Case Sampling: It involves selecting ‘illuminative cases’ (Patton, 2002) that illustrate a context in terms of outstanding successes or failures. That is it the approach focuses on the cases that have in-depth information. These cases may be unusual or peculiar or enlightening. This strategy would be particularly suitable for ‘realist syntheses’ (Suri, 2011) which examines how a program is likely to work under particular circumstances by analysing the successful and unsuccessful implementation of the program (Pawson, 2006). Say, for example, the objective of the study is to analyze the effectiveness of CSR programs, one might compare the CSR activities of different industries, or new CSR initiatives with that of well-established ones.

Past studies that have used extreme and deviant case sampling in their studies are Çetingöz (2012), Ersoy (2014), Lakhan, Bipeta, Yerramilli & Nahar (2017), and ?ahin (2008).  Lakhan et al., (2017) explored the common patterns of the consanguineous relationship in the parents of children with intellectual disability in India. The authors desire to explore whether intellectual disability which is inherited in families through consanguineous marriage can be the cause of intellectual disability in the children. Extreme or deviant case sampling was used to select cases from homes, camps, and clinical settings.  Similarly, Ersay (2014) employed extreme or deviant case sampling to select participants (teachers and students in this case) from two kinds of school, a low socioeconomic school and a high socioeconomic background school. The author wants to explore the challenges of citizenship education procedures in the social studies course in Turkey.

·         Intensity Sampling: Intensity sampling involves selecting samples that are excellent or rich examples of the phenomena of interest (Patton, 2002). It is similar to extreme case sampling but with less focus on the extremes. Intensity sample includes information-rich cases that exhibit intense but not extreme inputs. Intensity sampling looks for rich examples and not unusual cases. A mild sample won’t provide much to researchers for their study. So, a sample with sufficient intensity is required to make the study interesting. Intensity sampling involves prior information and judgment on part of the researcher. The researcher needs to do exploratory research to determine the nature of the variation in the study. For instance, if a researcher wants to have a comprehensive understanding of a phenomenon then it is crucial to examine cases where these changes were occurring thoroughly in the system over a sufficient period of time (Suri, 2011).

Several studies have used intensity sampling to conduct their qualitative studies are Hignett (2003), Falciani-White (2017), Issa (2006), Kashkalani, Maleki, Tabibi and Nasiripour (2017), Kleinn, Ramírez, Holmgren, Valverde and Chavez (2005), Mehra, Singh, Agarwal, Gopinathan and Nishchal (2015), Meland, Xu, Henze and Wang (2013), and Ragagnin, de Sena Júnior and da Silveira (2010). Kashkalani et al. (2017) used purposeful intensity sampling to identify the factors that are involved in determining the number of clinical faculty members required for medical schools in Iran. Similarly, Falciani-White (2017) used intensity sampling to select academic scholars from major three divisions of academia (humanities, natural sciences, and social sciences). The purpose was to understand how information behaviours function in the broader landscape of academic practice. Hignett (2003) also employed intensity sampling to choose participants from hospitals to examine the influence of organizational and cultural factors on the practice of ergonomics.

·         Maximum Variation (Heterogeneity) Sampling: In this approach, the key dimensions of variations are identified and then cases are selected that differ from each other as much as possible. This sampling technique yields—detailed descriptions of each case which are useful for capturing uniqueness, and the shared patterns that differentiate cases from each other. Purposeful sampling captures the central themes that span across a large sample or variation. Heterogeneity is an issue in small samples as individual cases vary from each other. The maximum variation sampling turns this problem into the strength by looking into common patterns that emerge from variation in a program. The variation in a small sample can be maximized by identifying the diverse characteristics to construct the sample. For instance, if a study looks into the effect of a new legislation in the State, specifically at different levels of management, and across rural and urban areas, there may not be enough resources to randomly select enough information across the state. The researcher can ensure a variation in geographical locations for the purposes of the study.