PXGS6102SurveySampling

SAMPLING

Sampling means selecting a given number of subject from a defined population as representative of that population.

The universe or target population means all members of a real or hypothetical set of people, events or object to which we wish to generalise the results of our research. The advantage of drawing a small sample from a large target population is that it saves time and expense of studying the entire population. If the sampling is done properly, the researcher can reach conclusions about an entire target population that are likely to be correct within a small margin of error by studying a relatively small sample.

UNIVERSE

(TARGET POPULATION) all Form 1 students in Malaysia

ACCESSIBLE all Form 1 students in Daerah

POPULATION Petaling

SAMPLE randomly select 2 schools

(10 classes x 45 students = 450)

It is seldom possible for a researcher to draw a representative sample from a target population such as all Form 1 students in Malaysian schools. To select a sample, the researcher has to draw his/her sample from an accessible population such as all Form 1 students in Daerah Petaling. The intention of the researcher is to generalise to the accessible population of Form 1 students the findings obtained from the sample. However, this can only be done if the sample drawn is random. A random sample is one in which all members of the population had an equal chance of being selected. To further strenghten generalisability of findings, the researcher should ensure population validity. The researcher must be able to demonstrate that the accessible population is closely comparable to the target population on a few variables that appear most relevant to the study, he/she has established population validity. Eg. you have to show that the Form 1 students in Petaling are closely comparable to Form 1 students in the whole of Malaysia in terms of academic performance in mathematics, reading ability, socio-economic status, etc.

Sampling Techniques

The following are some sampling techniques researchers use to enable them to generalise their findings to the larger population, which is the intention of most research>

a) Simple Random Sampling

All individuals in the defined population have an equal and independent channce of being selected as a member of the the sample. By 'independent' is meant that the selection of one individual does nt affect in any way the selection of any other individual. i.e each individual, event or object has an equal probability of being selected. Suppose for example there are 1000 Form 1 students in Daerah Petaling and we want to select a simple random sample of 100 students, when we select the first case, each student has one chance in 1000 of being selected. Once the student is selected, the next student to be selected has a 1 in 999 chance of being selected. Thus, as each case is selected, the probability of being selected next changes slightly because the population from which we are selecting has become one case smaller.

Using a Table of Random Numbers to select a sample. Obtain a list of all Form 1 students in Daerah Petaling and assign a number to each student. Then get a table of random number which consist of a long series of 5-digit number generalted randomly by a computer. Using the table, you randomly select a row or column as a starting point, the select all the number that follow in that row or column. If more numbers are needed, proceed to the next row or column until enough number have been selected to make-up the desired sample size.

b) Systematic Sampling

If it can be ensured that the list of students from the accessible population is randomly listed than systematic sampling can be used. First, you divide the accessible population (1000) by the sample desired (100) which will give you 10. Next, select a figure less or smaller than the number arrived by the division, i.e. less than 10. If you choose 8, then you select every eighth name from the list of population. This method differs from random sampling because each memebr of the population is not chosen independently.

c) Stratified Sampling

In certain studies, the researcher wants to ensure that certain sub-groups of individuals are included in the sample and for this stratified sampling is preferred. For example, if you intend to study differences in spatial ability among Form 1 students according to academic performance (high, middle & low), random sampling may not ensur that you have sufficient number of students for the 3 ability levels. What you should do is to divide the accesible population (Daerah Petaling) into the 3 levels of academic performance and then from each group randomly select for the sample. The proportion of students selected from each group usually is the same as the proportion of that group in the accessible population and the target population. eg. if only 25% are high performing students, then you sample should also have 25% high performing students.

d) Cluster Sampling

In cluster sampling, the unit of sampling is not the individual but rather a naturally group of individuals. Cluster sampling is used when it is more feasible or convenient to select groups of individuals than it is to select individuals from a defined population. Eg. in a particular district there are 1000 Form 1 students and they are in 25 classrooms of 45 students each. In cluster sampling, you draw a random sample of 5 classrooms from the list of 25 classrooms. Then you study every student each of the 5 classrooms. The main advantage of cluster sampling is that it saves time and money, It's disadvantage is that it is less accurate than simple random sampling because there is more sampling error. .