Sampling is one of the most important factor in educational research, sampling is a way of sele cting a fraction of people in a population of research in order to find the needed information which can generalised and extend their results further than the respondents, The researcher may not manage to collect the information for every individual of interest as it is costly, time consuming and stress nous, as(Mcmillin,1996) stated that when conducting an educational research, it is usually impractical and unnecessary to measure all the elements in the population of interest. The are two main types of sampling procedures suitable for conducting educational research, which is probability also known as random sampling and non-probability or non-random sampling. This paper will further discuss the two categories of sampling procedures that are commonly used in educational research and highlight the strength and weakness of using the discussed sampling procedures.
Probability sampling is the type of research in which the researcher chooses the respondents randomly. In this type of research each respondent has an equivalent possibility of taking part in the research. probability sampling is a way of choosing the respondents randomly as (Mcmillin,1996) highlighted that “this way of random selection implies at least every member of the population or the subgroups has an equal chance of being selected” Each of the population has equal chance of being picked up into the sample, moreover one member does not effect in selection of the other and there is no relation with one another, hence it is free from subjective factor or personal error or bias and prejudices or imagination of the researcher. The researcher makes sure that the sample formed by this method, may be representative of the population. while (Cohen etal.2007) noted that probability sampling is a best procedure for selecting sample he further stated that “they are many types of probability sampling procedure used in educational research, but only four are commonly used, these are simple random, systematic, stratified and cluster random sampling procedures.
A simple random sampling procedure is known to be a classic form of probability. (Mcmillin,1996;p87) defined it as “ In simple random sampling every member of the population has an equal and independent chance of being selected for the sample”.The researcher may write on the pieces of papers with yes and no , then them in a bowl then each member picks up a paper the ones with yes becomes the respondent.( Cohen etal.2007) stated that where this implies and that of ensuring representatives that the population is greatly adverse or heterogeneous. A simple random sampling procedure has it strengths and its weakness. It strength include, it does not have mistakes in classification for its targets the rightful population, (Kombo&Tromp, 2006) identified its strength as “the samples yield research data that can be generalised to a larger population”. It is not biased and prejudice. It allows the researcher to use rational statistics to the information. (Singh, 2006) highlighted some strengths as “It gives each element in the population an equal probability of getting into the sample; and all choices are independent of one another”, hence It gives each possible sample combination an equal probability of being chosen. Its weakness include that it can be very difficult to get a fulfilled information from the categorized wide population, some members may move out and non-response error is high, (Kombo&Tromp, 2006; p79) observed that “it is not the most statistically efficient method of sampling. The researcher may, just because of luck draw, not get good representation of subgroups in population”. The samples can be over presented and underrepresented. (Kothari,2004) noted that cases selected too widely dispersed or even impossible to contact and thus adherence to the whole sample may not be possible.in the selection of a given population this translate that the respondents may be too scattered apart hence making the collection too difficult. The respondents may be of the same heterogeneous group or age hence difficult to generalize.

The other probability sampling is the systematic. This is the use of every nth number from the listed number or given numbers of the population. (Singh, 2006) defined systematic sampling as “an improvement over the simple random sampling. This method requires the complete information about the population. There should be a list of information’s of all the individuals
Of the population in any systematic way. Now the researcher decide the size of the sample then selects each /nth individual from the list and thus get the desired size of sample”. Thus for this procedure to be effective, the sampling population are arranged in any systematic way and it involves choosing members from a population list in a systematic manner. For example, if from a population of, say, 1,000, a sample of 150 is needed, then every tenth person can be selected. Strength of systematic sampling is, that it allows a wider range of population to be covered of which there is a greater precision. (Cohen etal.2007) noted that systematic random sampling is easier and faster to draw than simple random sampling and it can be executed without any mistakes”, however, every member of the population has a same probability of taking part and not so biased. (Mcmillin, 1996; p88) noted that “There is a possible weakness in systematic sampling if the list of cases in the population is arranged in a systematic pattern”. For example if a list of grade seven pupils is school are arranged by classroom and pupils are arranged in their ability groups. If every nth pupil that is chosen matches with pattern, the sample would represent only a certain level of ability and would not be representative of the population. And the (Kombo&Tromp, 2006) stated that “the response may be too low since the respondent’s availability is unpredictable”. (Cohen etal.2007; p87) noted that, “This is not free from error, since there is subjectivity due to different ways of systematic list by different individuals”.

Stratified sampling procedure is the other probability technique, this is the one modified way of the other sampling procedures.( Mcmillin,1996;p79) explained that “stratified sampling involves dividing your population into homogenous subgroups and then taking simple random sample in each group”, it is a modification of a simple sampling procedure for the first thing to be done is regrouping the members in group for example according to the sex, age and level, then a simple random sampling procedure is applied in selecting the respondents. The strength of this technique is that all groups will be represented that is the sex, age ,level and common characterizes, it is easy to apply and has a good representation of the population, (Kombo&Tromp,2006;p88) identified that “It is an objective method of sampling and Observations can be used for inferential purpose”. Because to it nature it minimizes the cost when conducting it. (Mcmillin, 1996) stated that “the replacement of a unit can be done conveniently, if the originally selected respondents has not availability. Stratified random sampling procedure has its own weaknesses such as it is a complicated method of sampling especially on larger populations. (Cohen etal.2007;p88) recognized that “It involves errors when we consider the primary and secondary stages and It is again a subjective phenomenon”, it can be biased due to the inappropriate grouping of the members and not having the equal numbers, for it becomes difficult to generalized the results.

The last one in the probability is the cluster sampling procedure it is ridiculous to sample individual members large population of interest. Cluster sampling involves a wider and larger target population. This involves random selection of naturally occurring groups or areas and then selecting individual’s elements from the chosen groups or areas. (Cohen etal.2007). For instance listing all the primary schools in Lusaka district then randomly picking 40% of listed primary schools, the researcher will use the selected schools for research instead of all conducting the research in the whole district. The strength of cluster sampling is that it saves time and cost of conducting a research, the results can be generalized. (;Cohen etal.2007p80) recognized that “It is practicable and highly applicable in education and Observations can be used for inferential purpose” it’s weakness is that although it saves and money the results may not be accurate, hence may only benefit the whole area, for sometimes information cannot be generalized for example the research findings of kabulonga schools cannot be of the same solution of kayama school but they are all one district.( Mcmillin,1996;p81) identified that “there is a risk of missing on important sub group”, it may have inadequate representation of the target group. Cluster sampling procedure may be biased because it may not have the equal number of respondents and it may have so many mistakes.

The other type of the sampling is the non-probability, this type of sampling is not random but the researcher chooses who to take part or involve in the research and not all members of a population of interest have an equal chance of taking part in a research.( Cohen etal.2007;p133) indicated that “The selectivity which is built into a nonprobability sample derives from the researcher targeting a particular group, in the full knowledge that it does not represent the wider population; it simply represents itself”. The researcher looks for an individual that can provide the information regardless where they are coming from. In educational research the researcher would use purposive sampling, convenience sampling, quota sampling and snowball sampling procedure.

Snowball sampling procedure is when the researcher identifies an individual who can provide the required information or an individual who can help to identify a rightful respondants; hence it may start with very few respondents as they will increase in due course. (Kombo&Tromp, 2006; p83) defined that “snowball or chain sampling begins by asking a number of people, can you identify specific kinds of cases”. The strength of snowball sampling is that it saves time and money. It allows the researcher gather information to the right people, the results are accurate and can be generalized, and the research is entirely in the researcher’s hand. (Cohen etal.2007; p113) noted that “This method is useful for sampling a population where access is difficult, maybe because it is a sensitive topic (e.g. teenage solvent abusers) or where communication networks are undeveloped”, The weakness of this sampling is that the researcher may choose a wrong respondent or the respondent may invite a friend who may not be a right respondent, it is biased and the researcher may select unreliable respondent which will affect the result.

Purposive sampling is another non probability procedure. This is so close to snowball sampling but this one the researcher identifies people that has the characteristics needed for the study, for example if the is about “poor performance in reading” the teacher of English should be the priority.(Mcmillin.1996) identified that “In purposive sampling (sometimes referred to as sampling judgmental sampling) the researcher selects par- Particularly informative or useful elements from the population that will be representative or in- formative about the topic”. Purposive sampling has its own strength such as it is cheap and easier to execute, the researcher is able to include what he has perceived. In many cases purposive sampling is used in order to access ‘knowledgeable people’, i.e. those who have in-depth knowledge about particular issues, maybe by virtue of their professional role, power, access to networks, expertise or experience (Cohen etal.2007).the researcher understands the research it hence selecting right people and gets rightful information and be generalization and problem solving. The researcher is able to Use of the best available knowledge concerning the sample subjects and Better control of significant variables (Singh, 2006). It enables the researcher to use judgment and choose people that are presented or are available that best meet research objectives and the target groups, which make it for the researcher to successfully correct accurate information. The weaknesses of the purposive sampling include be biased because the respondents are handpicked and the respondents may not be interested or cannot be reliable hence can affect the results. (Singh, 2006) noted that “Inability to utilize the inferential parametric statistics and Inability to make generalization concerning total population”. In purposive sampling consistency of the standard information are not reliable especially if the researcher does not have the Knowledge of population.

Convenience sampling is also a non-probability .the researcher chooses respondents to those who are nearest to them. (Mcmillin,1996) illustrated that “A convenience sample is a group of subjects selected because of availability, for example, a university class of a professor conducting some research” whilst (Kombo&Tromp,2006) stated that “in this method is based on using people who are a captive audience, people the researcher meet haphazardly or accidently”. The researcher collects information from different types of people those who maybe just passing by the use of procedure, further more (Kothari, 2004) states clearly that Although we need to be very way of convenience samples, often this is the only type of sampling possible, and the primary purpose of the research may not be to generalize but to better understand relationships that may exist. The strength of this type of non-probability is that it targets the people who are the nearest hence reducing on transport and logistics, it is cheap and it easily done for it is not complicated to find the respondents. The weaknesses of this sampling procedure are that, it is not a representative of the population, it is not free from error and parametric statistics cannot be (Singh, 2006). The researcher may not collector accurate information as the respondent may not have detailed information or ignorant about the research topic. The last one in educational researcher is the quote sampling procedure. (Kothari.2004) identified it as “Quota sampling is used when the researcher is unable to take a probability sample but still wants a sample that is representative of the entire population”. This dividing the respondents in relevant strata such as age, gender of geographical. The strength of this sampling is that It is an improvement over the judgment sampling (Singh, 2006).it is easy and not complicated. It is most frequently used in social surveys Quota sampling has been described as the non-probability equivalent of stratified sampling. Like a stratified sample, a quota sample strives to represent significant characteristics (strata) of the wider population; unlike stratified sampling it sets out to represent these in the proportions in which they can be found in the wider population (Singh,2006). However the weakness of quota sampling includes it is not a representative sample, it is not free from error and it has the influence of regional geographical and social factors.

In conclusion probability sampling is also known as random sampling in which the all the member of the population of interest is given equal opportunities to take part in the research Random sampling techniques are based on the theory of probability and usually produce “good” samples [because it] is representative of the population that is being studied(Mcmillin,1996), there four types of probability sampling commonly used in educational research these are simple random, systematic, stratified and cluster sampling procedure. On probability known as nonrandom, Non-random samples are said to be bias samples because they are almost always systematically different from the population on certain characteristics (Kothari, 2004), the respondents are had picked. The four sampling procedure include purposive, snowball, convenience and quota sampling. These procedures have its strength and weakness, as most of them are easy, and cheap. These are very important to educational research to understand how sampling may affect research it is essential to know the characteristics of different sampling procedures. This knowledge will help the researcher interpret the sample that is used. It will be helpful to know the strengths and weaknesses of each sampling procedure, hence choosing right one when conducting the research.

References

Cohen. Mainly. And Morrison. (2007). Research Methods in Education6th Ed. New York. Rutledge.

Komba and Tromp. (2006).proposal and thesis writing; an introduction. Nairobi. Pauline’s.

Kothari.c.R. (2004). Research Methodology and Techniques 2ed.New Delhi. New age international Inc.

Singh.Y.K. (2006). Fundamentals of Research Methodology and Statistics. New Delhi.

Mcmillin, J.H. (1996).Edcational Research; fundamental for the Consumers 2ed. New York. Harper Collins.