Comment 1
Random sampling, also known as probability sampling, requires that each member of the study population have an equal opportunity to be selected as a study subject. Selecting members of the population randomly enable members of the population to be selected equally. Random sampling enables each facet of the study population to be represented without bias (Little, 2016). For instance, a questionnaire is provided in an unsystematic way to a population so each individual in the population has a chance to submit information for that analysis. Again, many outcomes in statistics are derived from probability theory. Probability always assumes that random processes are functionally (Little, 2016). For any statistical outcomes to be valid, there is need to use samples which can be treated as random variables. Simple random samples make it easier to do that. There is less chance of results of analysis being skewed using random sampling.
However, there are limitations with random samples. It requires individuals within the population to volunteer and provides information. The people who usually participate may have the strongest opinion on the topic being analyzed. This brings the question of accuracy of the representation of such population (Little, 2016). There are two main limitations of random sampling. These are biased outcomes and laborious and time-consuming. Biased results are brought by the fact that the population of interest produces the people to be a sample. If the question is about things affecting them, they may be biased in providing information (Little, 2016). Random sampling is also time-consuming and involves a lot of things to be done hence laborious.
To have a truly random sample, one will need to have access to the target population and be able to assign each participant randomly to a condition. This is ideal since only random assignment to people by researchers before manipulation can provide causal interpretability of results (Little, 2016). To prevent problems and limitations with random sampling, there is need to ensure that you have a group which is bigger and each member has different views.
Comment 2 ( this is an answer for the teacher about my answer) concerning this question. Explain each sampling technique discussed in the “Visual Learner: Statistics” in your own words, and give examples of when each technique would be appropriate.
Thanks you Rosy for your answer. When doing statistical interpretation of any data, we always hope to achieve a good understanding of the estimated population parameters. Naturally, we do this using observed data from various samples.
But how do you realistically determine the sample size based on confidence level?