Sampling

Click on the link below…

Sampling.ppt

1. Simple Random Sampling

Individuals are chosen in such a way that each has an equal chance of being selected, and each choice is independent of any other choice. If we wished to draw a sample of 50 individuals from a population of 600 students enrolled in a school, we could place the 600 names in a container, and, blindfolded, draw one name at a time until the sample of 50 was selected. However, this process is cumbersome (burdensome) and is rarely used. A more convenient way of selecting a random sample, so that the individuals can be equated, is by the use of a “table of random numbers.” When a table is used, it is necessary to assign consecutive numbers to each member of the population (from which the sample is to be selected). Then, entering the table at any page, row, or column, the researcher can select the sample from 001 to 999 (three digits); and from 0001 to 9999 (four digits). When a duplicated number or a number larger than the population size is encountered, it is skipped and the process continues until the desired sample size is selected.

*Ask me to demonstrate when you are in your expert group…!!!

2. Systematic Sampling

Consists of the selection of each nth term from a list. For example, if a sample of 200 were to be selected from a telephone directory with 200,000 listings, one would select the first name by selecting a randomly selected name from a randomly selected page. Then every thousandth name would be selected (200,000/200=1000th) until the sample of 200 names was complete. If the last page were reached before the desired sample size had been selected, the count would continue from the first page of the directory. Systematic sampling, sometimes called interval sampling, means that there is a gap, or interval, between each selection. This method is often used in industry, where an item is selected for testing from a production line (say, every fifteen minutes) to ensure that machines and equipment are working to specification. Alternatively, the manufacturer might decide to select every 20th item on a production line to test for defects and quality. This technique requires the first item to be selected at random as a starting point for testing and, thereafter, every 20th item is chosen. This technique could also be used when questioning people in a sample survey. A market researcher might select every 10th person who enters a particular store, after selecting a person at random as a starting point; or interview occupants of every 5th house in a street, after selecting a house at random as a starting point. It may be that a researcher wants to select a fixed size sample. In this case, it is first necessary to know the whole population size from which the sample is being selected. The appropriate sampling interval, I, is then calculated by dividing population size, N, by required sample size, n, as follows: I = N/n

3. Stratified Random Sampling

This is possible when the population is subdivided into smaller homogeneous groups, called strata (to get more accurate representation). For example, in an income study of wage earners in a community, a true sample would approximate the same relative number from each socioeconomic level of the whole community. If, in the community the proportion were 15% professional workers, 10% managers, 20% skilled workers, and 55% unskilled workers, the sample should include approximately the same proportions in order to be considered representative. Within each subgroup, a random selection should be used. Thus, for a sample of 100, the researcher would randomly select 15 professional workers from the subpopulation of all professional workers in the community, 10 managers from that subpopulation, and so on. This process gives the researcher a more representative sample than one selected from the entire community, which might be disproportionately weighted by a predominance of unskilled workers. In addition to socioeconomic status, such characteristics as age, sex, extent of formal education, racial origin, religious or political affiliation, or rural-urban residence might provide a basis for choosing a stratified sample (strata).

4. Area or Cluster Sampling

It is sometimes expensive to spread your sample across the population as a whole. For example, travel can become expensive if you are using interviewers to travel between people spread all over the country. To reduce costs you may choose a cluster sampling technique. This sampling technique is appropriate when, a) the population of interest is infinite, b) when a list of members of the population does not exist, or, c) when the geographic distribution of the individuals is widely scattered. Let’s say we want to select a sample of all public school elementary teachers in the five largest cities in Thailand (Bangkok, Nakhon Ratchasima, Nonthaburi, Chiang Mai, and Songkhla). A simple random sample would be impractical. From the 5 largest cities, three could be selected. From these three cities, all public elementary schools could be listed and a random sample of 30 (10 from each city) can be selected. From these 30 schools, we can then randomly select 5 teachers in each school. So, our true sample will be – 30 x 5 = 150 elementary school teachers.

Non-probability Sampling

Those that use whatever subjects are available, rather than following a specific subject selection process

May produce samples that do not accurately reflect the characteristics of a population

May lead to unjustifiable generalization and should not be used if random selection is practicable

Educational researchers, because of administrative limitations in randomly selecting and assigning individuals for a research, often use available classes as samples (e.g.: psychology professor uses students from Introduction to Psychology class as subjects – the results of this study can be generalized only to other similar groups of psychology students)

So, this kind of sampling may restrict generalizations (generalizable only to similar population)

Sample made up of volunteers may represent biased sample – volunteers may not necessarily be representative of a total population

1. Convenience sampling – selection based on the availability of subjects (whomever happens to be available at the time); also known as accidental sampling or haphazard sampling; sampling bias can occur

Volunteers

Pre-existing groups

2. Purposive sampling – selection based on the researcher’s experience and knowledge of the group being sampled; also known as judgment sampling; researcher believes a particular sample is appropriate for his/her study

Need for clear criteria for describing and defending the sample

Experience and prior knowledge of the researcher about a particular group (to be used as sample) are essential

The researcher can be “wrong”

3. Quota sampling – selection based on the exact characteristics and quotas of subjects in the sample when it is impossible to list all members of the population (e.g.: an interviewer might be told to go out and select 20 adult men and 20 adult women, 10 teenage girls and 10 teenage boys so that they could interview them about their television viewing)

Data gatherers are given exact characteristics and quotas of persons to be interviewed (e.g.: 35 working women with children under the age of 16, 20 working women with no children under the age of 16, etc.)

Data are obtained from easily accessible individuals

Thus, people who are less accessible (more difficult to contact, more reluctant to participate, and so forth) are underrepresented

Qualititative Sampling

Because of the unique characteristics of qualitative research, both random and non-random sampling techniques cannot be used. This calls for alternative sampling strategies for qualitative research.

In-depth inquiry

Immersion in the setting

Importance of context

Appreciation of participant’s perspectives

Description of a single setting

A qualitative researcher relies on his/her experience and insight to select participants. There are several types of sampling techniques possible for a qualitative study:

1. Intensity sampling = compare differences of two or more levels (“extremes”) of the research topic; e.g.: good vs. poor students (two extremes); effective vs. ineffective teachers (two extremes); small vs. medium vs. large size classes (three extremes), experienced vs. inexperienced teachers, etc.

2. Homogeneous Sampling = Subjects chosen by similarity based on a given characteristic

3. Criterion Sampling = Selection of all cases that meet a certain standard

4. Snowball Sampling = Selection of subjects who identify other subjects; one subject gives the researcher the name of another subject, who in turn provides the name of a third, and so on (e.g.: drug dealers, criminals, gay, isolated, etc.)

5. Random Purposive Sampling = From a purposive sample too large for a study, a random selection is performed

Sample Size?

Time

Money

Access

Complexity of data analysis