While doing a research study, it is mandatory for the researcher in taking the opinion of the people through some surveys and questionnaires, which will help the study to be authentic in nature. The researcher has to frame up the questions in such a manner so that it can help them in discussing the findings that they will get from the respondents. The use of the proper method of sampling will help them in targeting the right audiences who will help them in the completion of the study (Neuman and Robson 2014).
Population and sample size
Population consists of all the elements that are present within a set of data where as the sample refers to the particular observations that are required by the researcher from the entire population. The population is determined by the researcher from where they try to select a subset of participants for the survey that they wish to conduct. The sample size has a few numbers of observations that is decided by the researcher so that the process can be completed in a better manner and the responses can be collected by targeting a certain section of the participants. The population that is present to the researcher can help them in taking more number of observations from various sample population depending up on the survey that needs to be conducted (Riley et al. 2014).
The elements that are present in the population is that the various age groups and the different genders along with different income levels of the people. These are few of the factors that is present among the population along with the marital status of the people. The researcher has to segregate the participants whom they want to take part in the process of research. this is known as the sampling process (Brick 2015). The researcher has the choice in selecting the size of the sample on their own based on the method that they want to carry out the process of research. This is known as probability sampling. On the other hand, the sample that is collected by the researcher in a random manner is known as non-probability sampling where they just select the participants randomly and have no knowledge about the responses that will be collected. This is mostly done on an unbiased manner (Mercer et al. 2017).
The simple random method of sampling is one of the most familiar type of processes where mostly all the elements have a fair chance of selection and the sampling of the responses are done on a single stage where the elements get selected independently. This type of sampling process happens when the responses are collected in clusters (Etikan, Alkassim and Abubakar 2016).
The systematic process of sampling method starts when the elements are selected at a random manner firstly after which it is followed by the nth selection. For example, the researcher may choose the tenth response of the survey after which they may continuously choose the n+20th response from the survey sheet (Etikan, Musa and Alkassim 2016).
Another type o sampling method is known as the cluster sampling where the researcher has the option of selecting the sample through different stages. The selection of the elements is done in clusters such as the blocks of buildings or an entire school or university. After the responses are collected in clusters, the researcher then has to select the individual elements from those clusters through the process of systematic or random sampling (Sarstedt et al. 2017).
Probability and non-probability sampling
Probability sampling is the method in which the researcher is of the knowledge that all the participants have an equal and known chance of being selected in the process of research. This provides an equal opportunity to the researcher in selecting the participants. For instance, a population consisting of 100 participants has the odds to be chosen in the process of research. In case of non-probability sampling, these odds are not at all equal. For instance, the participants who are living close to the researcher will have a better option of being selected in the process of research. The probability sampling is the best way the researcher can create a sample that will represent the entire population (Uprichard 2013).
There are various types of probability sampling such as the simple random method of sampling where the subjects are selected at a random manner. This can be done by assigning different numerical values to the subjects from where the researcher gets the option of choosing any numerical value. This method helps in selecting the elements in an unbiased manner (Muhib et al. 2016).
The other type is known as stratified random sampling where the subjects are split in to mutual groups after which the researcher gets to use the simple method of sampling process. The systematic process of sampling method is where the first response is collected after which the researcher uses the nth term to select the other responses. The last type of sampling process is known as multi-stage random process of sampling, which is the combination of all the techniques that are mentioned above (Skinner and Wakefield 2017).
The advantages of probability sampling are that it is convenient for the researcher, as the process is simple and easy. It also helps the researcher in creating a sample as a representative out of the entire population. It also helps the researcher in creating layers and stratas that will help in making the process of sampling easier for the researcher (Hillyguys, Jackson and Young 2014).
The disadvantage is that the process may not work well if all the elements are not homogenous in nature. The process is also consumes time and is tedious for the researcher when they want to create a sample that is larger in size. The systematic process of sampling in particular is not as random, as the researcher may choose in the simple random technique of sampling (Dever and Valliant 2014).
The non-probability process of sampling is where the chances of being selected in the sample cannot be measured or calculated. This is almost the opposite of probability sampling where the researcher has the ability of calculating the elements. The process of non-probability sampling is based on the judgment that is subjective from the viewpoint of the researcher.
The advantages of this method of sampling are that is is effective in saving up cost and time. Due to its easy process, the researcher can use it easily in situations where probability sampling cannot help.
The disadvantage is that it is not possible for the researcher to know the population that is representing the sample for the process of research. The error margin and the confidence intervals also cannot be calculated by the researcher (Grafstrom, Saarela and Ene 2014).
The probability method of sampling can be classified as simple random sampling where the elements that are present in the entire population have a chance of being selected in the study of the research.
The systematic process of selection is where the representative people who are available are streamlined, for example the shoppers who are present in the outlet at a specific point of time.
The stratified method of sampling process is where the population is divided in to a certain way where the groups do not overlap each other and the collection of the samples have to be done from them by the researcher (Bernard, Wutich and Ryan 2016).
The non-probability process of sampling also has various techniques, which the researcher can choose to use in the process of the research. The first technique is known as convenience sampling where the researcher will have an easy access in choosing the participants for the process of research. The second technique is known as snowball sampling where the participants will recommend others who are meeting the requirements, as chosen by the researcher. An example of this would be when the researcher wants to collect the responses of all the members belonging to a particular club. The last technique is known as quota sampling where the participants are selected in a non-random manner according to a quota that is fixed by them. The quota sampling that is proportional in nature aims at matching the proportions of the characteristics that is found in the entire population. The non-proportional quota sampling on the other hand has less restriction in matching the level of the figures (Creswell et al. 2016).
The simple random method of sampling process helps the researcher in providing fair chances to the participants as they may get selected for the research process. This option can be chosen by the researcher, as it helps them in completing a frame of sampling where they provide a list of all the subjects that they want to measure out of the entire population. This sampling frame may be incomplete or missing, which needs to be filled up by taking up an alternative approach in statistics (Levy and Lemeshow 2013). The population taken up for the process of sampling has to be homogenous in nature where the researcher has to consider those samples, which will have the same effect on the research process. The researcher has to take steps so that it can ensure that the randomness is more in the sample. The randomness of the design of the research is based on the statistical inference that the researcher needs to take up. The researcher has to select the sample in such a manner so that they are willing to participate in the study (Elliot and Valliant 2017).
The systematic process of sampling is conducive in nature and focuses on a wider area of study where it takes in to consideration the arbitrary parameters that are present in the data as well. this type of study can be performed when the researcher has a shortage of budget and has less time as well. It helps the researcher in giving them a certain degree of control so that the process can be sensible (Callegaro et al. 2014). It can be formed with the help of hypothesis that is designed in a straight manner, as the parameters will be taken in to consideration that are fit for the study. The main problem with this type is that the population size that is available to the researcher can be approximated in a reasonable manner. This helps in creating the risk of choosing the cases that are common in nature, as the population do not exhibit a proper degree of randomness (Vehovar, Toepoel and Steinmetz 2016).
The stratified method of sampling also help the researcher in saving on their time and money, as a smaller group of people is selected from an entire population. The major advantage of this sampling process is that it helps in capturing the characteristics of the main population within the sample (Barratt, Ferris and Lenton 2015). It is similar to the method of weighted average where the characteristics that are present in the population is taken in to consideration when the sample is being decided by the researcher. The major disadvantage that is present in this type of sampling process is that has to be dealt in subgroups without which the method will be ineffective in nature (Callegero et al. 2014).
Convenience sampling is a type of non-probability sampling process, which is based on the data that is collected from the participants who are willing in participating in the research study. An example of this sampling method is the polls that are taken over in Facebook. The advantage of this technique is that it is easy and helps in the creation of the hypothesis (Levy and Lemeshow 2013). The data that is collected can be done over a shorter period of time and is considered to the cheapest as well. The disadvantages of this technique are that it is susceptible to the biasness of selection and can be influenced as well beyond the control of the researcher. Another disadvantage is that it could lead to error in the process of sampling as well (Creswell et al. 2016).
Snowball sampling takes place when the participants in the research recruit other participants as well for the test. This technique is used in places where it is difficult to find the potential participants. It mainly consists of two steps, which includes in identifying the subjects who are potential within the population. The other step is to ask those people in recruiting their known participants. The advantage of this study is that it can be conducted in places where it is difficult to find participants. It also helps in discovering certain characteristics within the population that was unknown previously. The disadvantage is that it cannot help in determining the error that is present in sampling and make deductions regarding the sample that is obtained from the population (Grafstrom, Saarela and Ene 2014).
Quota sampling is the method through which representative data can be gathered from a particular group. It can be divided in to controlled and uncontrolled quota sampling where controlled sampling is based on proper restrictions, which limits the choices of the researcher. The uncontrolled sampling is where the researcher is free to choose any sample from the members of the group (Bernard, Wutich and Ryan 2016). The advantage of this sampling process is that it helps the researcher in saving up the cost and time. It is also not dependent on the presence of the frames that are required in the sampling procedure. The disadvantages of this technique is that there is no possibility in calculating the error that can take place in the process of sampling and the findings of the research can be risky as well due to the total population that has been surveyed. The researcher may also be biased during the process, which may lead to the wrong outcome and affect the quality of the work (Vehovar, Toepoel and Steinmetz 2016).
Therefore it can be recommended that the researcher can use up the techniques provided that they understand the limitations that are present with each technique. This will help them in analyzing the problems so that they can take better precautions and continue their process of research. The researcher has to understand the techniques that will suit in the research process so that the errors can be avoided. This will help them in carrying out the process of the research in a better way and help in reaching a finite conclusion.
Barratt, M.J., Ferris, J.A. and Lenton, S., 2015. Hidden populations, online purposive sampling, and external validity: Taking off the blindfold. Field Methods, 27(1), pp.3-21.
Bernard, H.R., Wutich, A. and Ryan, G.W., 2016. Analyzing qualitative data: Systematic approaches. SAGE publications.
Brick, J.M., 2015. Explorations in Non-Probability Sampling Using the Web. In Proceedings of Statistics Canada Symposium 2014.
Callegaro, M., Baker, R., Bethlehem, J., Göritz, A.S., Krosnick, J.A. and Lavrakas, P.J., 2014. Online panel research. Online panel research: A data quality perspective, pp.1-22.
Creswell, J., Plano Clark, V., Ivankova, N., Teddlie, C. and Tashakkori, A., 2016. Activity 22 Knowing about Probability Samples. 100 Activities for Teaching Research Methods.
Dever, J. and Valliant, R., 2014. Estimation with non-probability surveys and the question of external validity. In Proceedings of Statistics Canada Symposium.
Elliott, M.R. and Valliant, R., 2017. Inference for nonprobability samples. Statistical Science, 32(2), pp.249-264.
Etikan, I., Alkassim, R. and Abubakar, S., 2016. Comparision of Snowball Sampling and Sequential Sampling Technique. Biometrics & Biostatistics International Journal, 3(1), p.00055.
Etikan, I., Musa, S.A. and Alkassim, R.S., 2016. Comparison of convenience sampling and purposive sampling. American Journal of Theoretical and Applied Statistics, 5(1), pp.1-4.
Grafström, A., Saarela, S. and Ene, L.T., 2014. Efficient sampling strategies for forest inventories by spreading the sample in auxiliary space. Canadian Journal of Forest Research, 44(10), pp.1156-1164.
Hillygus, D.S., Jackson, N. and Young, M., 2014. Professional respondents in non-probability online panels. Online panel research: A data quality perspective, pp.219-237.
Levy, P.S. and Lemeshow, S., 2013. Sampling of populations: methods and applications. John Wiley & Sons.
Mercer, A.W., Kreuter, F., Keeter, S. and Stuart, E.A., 2017. Theory and Practice in Nonprobability Surveys: Parallels Between Causal Inference and Survey Inference. Public Opinion Quarterly, 81(S1), pp.250-271.
Muhib, F.B., Lin, L.S., Stueve, A., Miller, R.L., Ford, W.L., Johnson, W.D., Smith, P.J. and Community Intervention Trial for Youth Study Team, 2016. A venue-based method for sampling hard-to-reach populations. Public health reports.
Neuman, W.L. and Robson, K., 2014. Basics of social research. Pearson Canada.
Riley, W., Hays, R.D., Kaplan, R.M. and Cella, D., 2014. Sources of comparability between probability sample estimates and nonprobability web samples estimates. In Proceedings of the 2013 Federal committee on statistical methodology (FCSM) research conference.
Sarstedt, M., Bengart, P., Shaltoni, A.M. and Lehmann, S., 2017. The use of sampling methods in advertising research: A gap between theory and practice. International Journal of Advertising, pp.1-14.
Skinner, C. and Wakefield, J., 2017. Introduction to the design and analysis of complex survey data. Statistical Science, 32(2), pp.165-175.
Uprichard, E., 2013. Sampling: bridging probability and non-probability designs. International Journal of Social Research Methodology, 16(1), pp.1-11.
Vehovar, V., Toepoel, V. and Steinmetz, S., 2016. Non-probability Sampling. The SAGE Handbook of Survey Methodology, p.329.