Sampling Techniques Lecture Notes PDF

Sampling Techniques  Sample: a part of a population, generally selected so as to be representative of the population in the variable(s) under study. Sampling is a mandatory in researches as comprehensive studied are generally impractical. Advantages of sampling over a comprehensive su...

Sampling Techniques  Sample: a part of a population, generally selected so as to be representative of the population in the variable(s) under study. Sampling is a mandatory in researches as comprehensive studied are generally impractical. Advantages of sampling over a comprehensive survey: 1. Lower cost 2. Greater speed 3. Greater information 4. Sometimes the sampling is imperative when the comprehensive survey is impossible; e.g. collection of data about fish, animals, nomadic population; data collection in testing the quality of certain industrial products e.g.matches lamps. Another example is the examination of blood of patients.  Types: I. Non-Probability OR Non-Random samples II. Probability OR Random samples I. Non-probability or Non-random Types a. Purposive samples: They are chosen according to one's own judgment: i.e. not chosen in a random way and thus one can not generalize their results.  b. Pre-test or pilot study: It is used in pre testing the studies so as to avoid missing of important parameters and to exclude any unnecessary variables from the start, and thus saving time, money and personnel. c. Quota sample: It is used in especially in the U.S.A by Gallup Institute prior to voting. The investigator is asked to obtain the required information from a number of specific individuals (classified in different groups). This method of sampling is of no use in community medicine or clinical practice.  d. Convenience sample is one of the main types of non-probability sampling methods. A convenience sample is made up of people who are easy to reach.  Disadvantages of non-probability samples: 1. Probability of selection cannot be determined 2. The standard error of the sample mean cannot be estimated 3. The sampling results cannot be generalized II. Probability or Random Characteristics: 1. The probability of selection of an individual from the population can be determined. 2. The standard error of the sample result can be computed. 3. Generalization of the sample results over the total population can be made.  Types of probability samples: 1. Simple random sample 2. Systematic random sample 3. Stratified random sample 4.Multi-stage random sample 5. Cluster sample 1. Simple random sample Methods of selection: 1.Method of ideal bowl: Serial no. Name 1 2 3 - We make a frame as shown in this figure. - Use a number of paper cards equal to the population size. - Each of these papers will bear a serial number taken from the frame. - All these papers must be similar in every respect in order to avoid bias. - All these papers must be folded in the same manner and put in the ideal bowl, then thoroughly mixed before selection. 2. Tossing a coin. 3. Taking random balls: each of which represents a member of the population, out of a container after proper mixing. 4. Computer: by generating random numbers.  Advantages Simple random sample: 1. It is considered the basic type of probability samples. 2. Every member of the population has an equal chance or probability of being selected.  Disadvantages Simple random sample: 1. Construction of the frame may be difficult, especially when the population size is large. 2. The sample members may be concentrated in a certain sector of the population e.g. all members may be females. 3. It is not suitable for population of great variability as the standard deviation is high and the degree of precision is low  2. Systematic random sample Methods of selection: If we have a population of 100 individuals and we want a sample of 10, we can select any number between 0 and 10 at random, say 4. Then we go on taking individuals of numbers 4, 14, 24, 34, 44, 54, 64, 74, 84, 94. This will constitute the sample.  Advantages: 1. Ease of selection. 2. It is well distributed over various populations. 3. The degree of precision of its results is better than simple random sample.  Disadvantages: 1. Difficulty in constructing the frame if the population is large. 2. Sometimes, we cannot obtain the required sample size, which may be smaller than the required e.g. if the population size = 11 and the sample size = 4: if the first chosen number is number 2, we can obtain the required sample size [2,5,8,11]. But, if the first number is number 4, the sample size will be 3 instead of 4 [4, 7, 10].  3. Stratified random sample This is the best type of probability sample, particularly if there is a great variability among the population members i.e. if the value of the SD is high. Where population divided into a number of distinct categories, the frame can be organized into separate "strata." Each stratum is then sampled as an independent sub-population, out of which individual elements can be randomly selected.  Advantages: 1. Every unit in a stratum has same chance of being selected. 2. Using same sampling fraction for all strata ensures proportionate representation in the sample. 3. Adequate representation of minority subgroups of interest can be ensured by stratification. The general principle is to divide the population into a number of strata fulfilling two conditions: a) Each stratum should be homogeneous as possible. b) Each stratum should be exclusively sharply defined. According to According to Socioeconomi educational occupation c standard level - Illiterate - Professional - High - Read and - Semi- - Middle write professional - Low - Basic - Skilled education workers - Secondary - Semi-skilled - Higher workers education - Unskilled workers  Disadvantage: 1. sampling frame of entire population has to be prepared separately for each stratum 2. Second, when examining multiple criteria, stratifying variables may be related to some, but not to others, further complicating the design, and potentially reducing the utility of the strata. 3. In some cases (such as designs with a large number of strata, or those with a specified minimum sample size per group), stratified sampling can potentially require a larger sample than would other methods.  4.Multistage random sample We use this type of sample if there is a very large population. Method First stage: A list for all the governorates is made. Second stage: select a sample of two governorates by simple random method. Third stage: select random number of districts chosen in all states. Forth stage: select random number of talukas, villages. Fifth stage: units will be houses. — All ultimate units (houses, for instance) selected at last step are surveyed.  Advantages: 1. Construction of the frame is always very easy unlike the previous three types of samples. 2. It is the most suitable sample type to be used for selecting a sample representing the whole country e.g. for choosing a representative sample from the Egyptian population. 3. Cost and speed that the survey can be done in 4. Convenience of finding the survey sample 5. Normally more accurate than cluster sampling for the same size sample  Disadvantages: 1. Not as accurate as Simple Random Sample if the sample is the same size.  5- Cluster sample When all population in any stage of multistage random sample were chosen en toto (including all ages ، sexes, socioeconomic classes) the sample then is called cluster sample. Cluster sampling is an example of 'two-stage sampling'. First stage a sample of areas is chosen; Second stage a sample of respondents within those areas is selected. Population divided into clusters of homogeneous units, usually based on geographical contiguity. Sampling units are groups rather than individuals. A sample of such clusters is then selected. All units from the selected clusters are studied.  Advantages: 1. Cuts down on the cost of preparing a sampling frame. 2. Can reduce travel and other administrative costs. Disadvantages: 1. Sampling error is higher for a simple random sample of same size. — Often used to evaluate vaccination coverage in EPI