Non-probability sampling is a sampling technique that researchers can use to effectively collect data. The data is collected from individuals who are in a certain population. This method is however different from random sampling in the view that it does not give all the individuals in that population an equal chance of being selected for the research.
This method is always very effective in situations where researchers are bounded by either time, finances or even a workforce to accomplish the data collection(Rubin 2016). These limitations make it impossible to perform a random sample throughout the entire population. This therefore makes it necessary to employ an alternative form of sampling, and this is where non-probability sampling comes in.
Questionnaires are an effective method of data collection. They have several advantages, which makes them more efficient and effective to use. Questionnaires are affordable, as they are cheap to produce and distribute. Questionnaires also offer speedy results, especially with closed ended questionnaires. Questionnaires are also easy to use, and one does not require to be an expert to use them. Also offer anonymity to users, and users can fill them at their own time and discretion and help in respondents being more genuine with their answers.
Apart from being advantageous, questionnaires also have various disadvantages, which may make them inappropriate to use. One great disadvantage of questionnaires is the possibility of having low response rates. These response rates vary greatly, from 10% to 90% response rates. This wastes time and resources in the end and thus one should not overly rely in these for research data. It is also very difficult to probe responses in a questionnaire, and this loses the flavor of the response, especially in closed ended questionnaires. Also, questionnaires may not be filled by the intended people, as some respondents end up delegating other people to fill them up on their behalf, thus one does not receive correct responses(Andreau 2015).
Questionnaires also have various ethical issues. They include issues in their design, content and even privacy. Questionnaires should often have questions that prompt respondents to give certain specific answers instead of letting them decide. This can be solved by having open questions where the respondent is free to answer to their will. Another issue comes in anonymity of responses, where some organizations often disclose respondents’ details as well as the data collected to other organizations. This can be solved by failing to collect personal details of respondents, and this way their data will not be let out.
Measurement reliability refers to the extent at which data is consistent. There are three types of reliability which are; over time, across items and across different researchers. On the other hand, measurement validity is the extent to which certain measurements represent the variables that they are intended to represent(Doucet 2015).
Andreau, C. (2015). Questionnairs as a method of sampling. International Journal of Mathematics, 61-74.
Doucet, A. (2015). Measurement Validity Vs Reliability. Journal of Mathematics, 34-37.
Rubin, D. (2016). Non-probable Sampling. Journal of Statistics, 20-34.