COURSE INFO

TERM                         Fall 2023                                                    TIME     S 1.00 – 3.00 PM

Dr. Mohamed Hadji    mohamed.hadji@univ-saida.dz

mohamedhadji1983@gmail.com

Lesson 12   SAMPLING  

 

 

12.1. What is a sample?

Most people, we think, base their conclusions about a group of people (students, Republicans, football players, actors, and so on) on the experiences they have with a small number, or sample, of individual members. Sometimes such conclusions are an accurate representation of how the larger group of people acts or what they believe, but often they are not. It all depends on how representative (i.e., how similar) the sample is of the larger group. One of the most important steps in the research process is the selection of the sample of individuals who will participate (be observed or questioned). Sampling refers to the process of selecting these individuals.

 

12.2. Sample and populations

A sample in a research study is the group on which information is obtained. The larger group to which one hopes to apply the results is called the population. All 700 (or whatever the total number of) students at State University who are majoring in mathematics, for example, constitute a population; 50 of those students constitute a sample. Students who own automobiles make up another population, as do students who live in the campus dormitories. Notice that a group may be both a sample in one context and a population in another context. All State University students who own automobiles constitute the population of automobile owners at State, yet they also constitute a sample of all automobile owners at state universities across the United States.

When it is possible, researchers would prefer to study the entire population of interest. Usually, however, this is difficult to do. Most populations of interest are large, diverse, and scattered over a large geographic area. Finding, let alone contacting, all the members can be time-consuming and expensive. For that reason, of necessity, researchers often select a sample to study. Some examples of samples selected from populations follow:

A researcher is interested in studying the effects of diet on the attention span of third-grade students in a large city. There are 1,500 third-graders attending the elementary schools in the city. The researcher selects 150 of these third-graders, 30 each in five different schools, as a sample for study.

An administrator in a large urban high school is interested in student opinions on a new counseling program in the district. There are 6 high schools and some 14,000 students in the district. From a list of all students enrolled in the district schools, the administrator selects a sample of 1,400 students (350 from each of the four grades, 9–12) to whom he plans to mail a questionnaire asking their opinion of the program.

The principal of an elementary school wants to investigate the effectiveness of a new U.S. history textbook used by some of the teachers in the district. Out of 22 teachers who are using the text, she selects a sample of 6. She plans to compare the achievement of the students in these teachers’ classes with those of another 6 teachers who are not using the text.

12.2. Types of sampling

There are two main types of sampling methods: random sampling method and non- random sampling methods. Each one is divided into sub methods.

12.2.1. Random sampling methods

There are three main kinds of random sampling methods when a researcher intend to collect data from participants.

A. Simple random sampling (SRS) is the most commonly used method of selecting a probability sample. This method aligns with the definition of randomization, where each element in the population has an equal and independent chance of selection.

To illustrate, we consider the example of a class with 80 students. The first step is to assign a number from 1 to 80 to each student. Suppose you want to select a sample of 20 using the simple random sampling technique. You can use methods like the fishbowl draw, a table for random numbers, or a computer program to randomly select the 20 students. These 20 students then form the basis of your inquiry.

B. Stratified random sampling, as discussed, recognizes that the accuracy of an estimate depends on the variability or heterogeneity of the study population concerning characteristics strongly correlated with the research objective (Principle 3). In stratified random sampling, the researcher aims to stratify the population so that each stratum is homogeneous regarding the chosen characteristic.

It is crucial that the chosen characteristics for stratification are easily identifiable in the study population. For instance, it is simpler to stratify a population based on gender than on age, income, or attitude. Additionally, the chosen characteristic should be related to the main variable being explored. After dividing the sampling population into non-overlapping groups (strata), you then apply simple random sampling within each stratum to select the required number of elements.

There are two types of stratified sampling:

1.       Proportionate Stratified Sampling: The number of elements selected from each stratum is proportional to its size in the total population.

2.       Disproportionate Stratified Sampling: The size of the stratum is not considered, and the required number of elements is selected without proportionality.

C. Cluster sampling is a technique employed when it becomes challenging and costly to individually identify each element in a large population, such as in the case of a city, state, or country. Unlike simple random and stratified sampling, which rely on the researcher's ability to identify each element, cluster sampling involves dividing the sampling population into groups, referred to as clusters. These clusters are formed based on visible or easily identifiable characteristics.

 

In cluster sampling, the researcher selects entire clusters rather than individual elements, and then within each chosen cluster, elements are randomly selected using the simple random sampling (SRS) technique. Clusters can be created based on geographical proximity or a common characteristic that correlates with the main variable of the study, similar to stratified sampling.

The formation of clusters facilitates a more practical approach, especially when dealing with large populations. The researcher can choose to cluster based on geographical regions or characteristics relevant to the study. Depending on the level of clustering, sampling may be conducted at different stages, leading to single, double, or multiple stages of clustering. These various levels or stages of clustering will be further elaborated on later.

12.2. 2. Nonrandom sapling methods

A. Quota Sampling

Quota sampling is primarily driven by the researcher's convenience and accessibility to the sample population. This method involves selecting individuals based on a visible characteristic, such as gender or race that is of interest to the study. The researcher identifies a convenient location and approaches individuals exhibiting the relevant characteristic, asking them to participate until the predetermined quota is reached.

For instance, if the goal is to survey 20 male students to determine the average age in a class, the researcher positions themselves at the classroom entrance and queries the age of each male student entering until the quota is met. Similarly, if investigating the attitudes of Aboriginal and Torres Strait Islander students, the researcher selects a convenient location and collects information from such students whenever they are encountered.

Advantages of quota sampling include cost-effectiveness, as it does not require extensive information about the sampling population, and it ensures the inclusion of the desired demographic. However, its limitations include the inability to generalize findings to the entire population due to the non-probabilistic nature of the sample. Additionally, the accessibility of individuals may lead to a sample that is not fully representative of the entire population, as those easily reachable may possess unique characteristics. To enhance representativeness, researchers can diversify the locations from which they draw the sample to capture a broader range of characteristics.

B. Accidental Sampling

Accidental sampling is based on convenience in accessing the sampling population. Unlike quota sampling, which aims to include individuals with an obvious or visible characteristic, accidental sampling does not make such an attempt. Data collection stops when the researcher reaches the predetermined number of required respondents for the sample. This sampling method is commonly employed in market research and journalism, sharing similar advantages and disadvantages with quota sampling. However, since accidental sampling is not guided by obvious characteristics, some contacted individuals may lack the necessary information.

C Judgmental or Purposive Sampling

Purposive sampling relies on the researcher's judgment to identify individuals who can provide the best information for achieving the study's objectives. Researchers selectively approach individuals whom they believe are likely to possess the required information and are willing to share it. This sampling strategy is particularly valuable when constructing a historical reality, describing a phenomenon, or exploring areas with limited existing knowledge. While more prevalent in qualitative research, it can also be adapted for quantitative research, where a predetermined number of individuals are chosen based on their perceived ability to contribute relevant information.

D. Expert Sampling

Expert sampling shares similarities with judgmental sampling, with the key distinction being that, in expert sampling, respondents must be known experts in the field of interest. This approach is employed in both qualitative and quantitative research studies. In qualitative research, the number of experts contacted depends on reaching data saturation, ensuring comprehensive coverage. In quantitative research, researchers determine the number of experts to be contacted without consideration of data saturation. The process involves identifying individuals with demonstrated expertise, obtaining their consent, and then collecting information either individually or collectively, such as in a group setting.


آخر تعديل: السبت، 10 فبراير 2024، 8:39 PM