Event Website
https://publichealth.lsuhsc.edu/honorsday/2021/
Start Date
1-4-2024 9:00 AM
Description
Survey sampling techniques can be categorized into two types, which are namely probability and nonprobability sampling techniques. In probability sampling, each member of the population has a known, non-zero chance of being selected to a study. In nonprobability sampling, however, population members are selected using a non-random process; therefore, in nonprobability sampling not all members of the population have a chance of being in the sample. As such, it is difficult to know if the sample reflects the distribution of the larger population, creating the issue of non-generalizability of research findings. Consequently, nonprobability sampling is inferior to probability sampling. However, there are circumstances where probability sampling techniques may not be feasible, practical, or theoretically sensible. In such situations, nonprobability sampling methods are implemented inevitably. They are cost- and time-effective, easy to apply, and can be utilized with small populations. Convenience sampling, which is also known as Haphazard or Accidental sampling, is a type of nonprobability sampling and involves collecting a sample via the easiest possible manner, which often occurs at one location and time with whomever is available. In other words, convenience sampling involves getting participants wherever the researcher can find them and whenever is convenient. Thus, there is a high chance that the convenient sample is not representative enough of the population under study. In this study, we evaluate the bias and mean square error (MSE) of the sample mean obtained from convenience sampling and compare them with the ones obtained from simple random sampling using a simulation study under various scenarios. We show that the MSE of the sample mean from convenience sampling is much higher than that from simple random sampling for most of the scenarios. We conclude that convenience sampling should be limited to pilot studies where very little on the study topic is known or should only be considered as a last resort when implementation of probability sampling is absolutely not possible.
Recommended Citation
Mazumder, Harun; Grunblatt, Sarah; Zeng, Chaoyi; and Oral, E., "A Simulation Study to Compare the Bias and Efficiency of Sample Mean Obtained from Probability and Nonprobability Samples" (2024). School of Public Health Delta Omega Honors Day Poster Sessions. 10.
https://digitalscholar.lsuhsc.edu/dohd/2021/2021/10
A Simulation Study to Compare the Bias and Efficiency of Sample Mean Obtained from Probability and Nonprobability Samples
Survey sampling techniques can be categorized into two types, which are namely probability and nonprobability sampling techniques. In probability sampling, each member of the population has a known, non-zero chance of being selected to a study. In nonprobability sampling, however, population members are selected using a non-random process; therefore, in nonprobability sampling not all members of the population have a chance of being in the sample. As such, it is difficult to know if the sample reflects the distribution of the larger population, creating the issue of non-generalizability of research findings. Consequently, nonprobability sampling is inferior to probability sampling. However, there are circumstances where probability sampling techniques may not be feasible, practical, or theoretically sensible. In such situations, nonprobability sampling methods are implemented inevitably. They are cost- and time-effective, easy to apply, and can be utilized with small populations. Convenience sampling, which is also known as Haphazard or Accidental sampling, is a type of nonprobability sampling and involves collecting a sample via the easiest possible manner, which often occurs at one location and time with whomever is available. In other words, convenience sampling involves getting participants wherever the researcher can find them and whenever is convenient. Thus, there is a high chance that the convenient sample is not representative enough of the population under study. In this study, we evaluate the bias and mean square error (MSE) of the sample mean obtained from convenience sampling and compare them with the ones obtained from simple random sampling using a simulation study under various scenarios. We show that the MSE of the sample mean from convenience sampling is much higher than that from simple random sampling for most of the scenarios. We conclude that convenience sampling should be limited to pilot studies where very little on the study topic is known or should only be considered as a last resort when implementation of probability sampling is absolutely not possible.
https://digitalscholar.lsuhsc.edu/dohd/2021/2021/10