The sampling process is often important in both quantitative and qualitative research. Whereas random sampling strategy is often linked to quantitative study method, non-random sampling strategy is tied to qualitative study method. Still, the selection of sampling method must always be founded on the generalization interest. Non-probability strategies, on the other hand, are free from randomization techniques. They are used in studies where it is impossible to randomize the study sample. Examples of non-probability sampling include purposive sampling, snowballing, and expert sampling among others. The mentioned sampling strategies are employed when the members of the population of interest have similar crucial variables and are heterogeneous (Teddlie & Yu 2007). However, probability sampling signifies the superlative strategy for choosing research participants.
According to (Onwuegbuzie and Collins (2007), Probability strategies ensure that all members of a given population have an equal chance of selection. The sampling method may be undertaken randomly, systematically or through stratification of the data of interest. The cluster sampling strategy which is an example of probability strategy is ideal when the area of study comprises of units rather than individuals. The method is quite simple and convenient to conduct and saves time as the sampling procedure is complete within a short time. Probability strategies have a high tendency of representativeness of all data in the population. (Onwuegbuzie & Collins (2007).
Data Collection Methods
The method of data collection can be qualitative or quantitative. Qualitative methods address the “why” and “how” in research. Some examples of qualitative research methods include focus groups, and interviews among others. On the contrary, the quantitative methodology is centered on the “what.” It is systematic and standardized. An example of such approach is survey. One preferable data collection method is the Questionnaire. It may be either oral or written. The strengths of the questionnaire are that it is an inexpensive and practical way of gathering information. Mobile and online surveys have no printing costs as they are sent via email to target groups. Also, the results from digital questionnaires give quick feedback in real time. That means a survey can be conducted in the shortest time possible. It also allows for data collection from a larger audience. The tool can be sent to anyone in any part of the world. The respondents are also not under any form of pressure, thus they can take their time to complete the questions. The benefits of anonymity of respondents allow honest response thus offering data accuracy. However, the limitations of Questionnaire is that it does not allow expression of emotions and feelings. The lack of face-to-face interaction may also lead to misinterpretation of questions leading to skewed results. There is also a risk that some questions may be unanswered. It may also be unsuitable for illiterate respondents or those with hearing or visual challenges.
Though the method requires a participant’s approval, the ethical issue involved could be the sensitivity of the questions which could trigger distress. However, the strategy will be offering a clear overview of the research and ensure the mood in which the interviewee is before administering the data collection method. Moreover, the questionnaire will have the option of ‘pass’ to ensure that participant’s right are not infringed.
Reliability and Validity
Reliability is the capacity of research to yield same results whenever it is repeated. If the test produces different findings whenever research is conducted using the same method, then the study cannot be termed as reliable. On the other hand, validity is the integrity of the research. Validity means the capacity of study research to be generalized to a study population. The relationship between reliability and validity is that for a study to be valid, it has to be reliable. However, reliability is not a condition for validity (Drost 2011). For instance, in my discipline of health science, my weighing scale could read 109 pounds anytime a particular client step on it despite the client’s actual weight 120 pounds. Hence, it shows that my weighing scale is reliable but not valid.
Drost, E. A. (2011). Validity and reliability in social science research. Education Research and Perspectives, 38(1), 105–124.
Onwuegbuzie, A. J., & Collins, K. M. (2007). A typology of mixed methods sampling designs in social science research. The Qualitative Report, 12(2), 281–316. Retrieved from http://nsuworks.nova.edu/tqr/vol12/iss2/9
Teddlie, C., & Yu, F. (2007). Mixed methods sampling: A typology with examples. Journal of mixed methods research, 1(1), 77-100.