Sampling Methods PDF

Simple Random Sampling Basic type of probability sample in which every individual in the population has an equal opportunity of selection Assign a number to each member of the population (Frame) Methods of simple random selection ⚫ Ideal Bowl ⚫ Random number table ⚫ Computer generated random sele...

Simple Random Sampling Basic type of probability sample in which every individual in the population has an equal opportunity of selection Assign a number to each member of the population (Frame) Methods of simple random selection ⚫ Ideal Bowl ⚫ Random number table ⚫ Computer generated random selection Sampling frame Serial number Name of sampling units 1 2 3 4 5 N ------------------------------------------------------------------------------------------------------------------------- Methods of selection of Simple Random Sample (A) Ideal Bowl Method Ideal Bowl Method In this method we prepare a frame and then we secure an ideal bowl as shown in the figure. This bowl should have a smooth inner surface spherical in shape (in order to avoid sticking of cards to its walls). Then we prepare a number of cards equal to N (population size) and these cards should be identical from every respect then on each card we put one of the serial numbers present in the frame. Ideal Bowl Method The cards are then folded in the same manner and put in the ideal bowl then they must be thoroughly mixed and a blindfolded person selects a number of cards from the bowl = n (sample size). Such selection can be done in 2 different ways; sampling without replacement or sampling without replacement. (B) Tables of random numbers They are lists of numbers where the digits 0-9 are arranged in a random order. By progressing along one or more adjacent rows of digits, it is possible to select the sample randomly. The starting point for sample selection from the tables should be determined randomly. Table of Random Numbers 38742 24201 25580 18631 01448 34768 89533 74163 92837 69008 28091 23715 67552 13487 06594 55983 45285 37836 74970 64602 01664 22496 81470 17206 68065 07271 43011 55337 Example: Simple Random Sample To evaluate the prevalence of tooth decay among 1200 children attending a school List of children attending the school Children numerated from 1 to 1200 Sample size = 100 children Random sampling of 100 numbers between 1 and 1200 Example: Simple Random Sample Data from members of the population that correspond to these numbers become members of the sample. Advantages of SRS • Most reliable of all sampling methods Selection procedures are simple Disadvantages of SRS 1. Difficulty in construction of frame particularly for large population. 2. The selection of sample even by the use of the table of random numbers is difficult particularly with large samples. 3. Sometimes all or most of the sample members belong to a certain population group while the other groups will not be properly represented (ex all sample members females) Disadvantages of SRS 4. Simple random samples cannot be used in populations with a high degree of heterogeneity (with a high degree of variability) since the degree of precision of sample results will be low and the simple random sample will not be the suitable sample to choose. 5. Most costly in case of household. Systematic Random Sample First element is selected randomly and then every nth element is selected Three steps involved ⚫ Divide total number of cases in population by the number of cases required for the sample This is the sampling interval, or the number of cases from one sampled case to another Systematic Random Sample ⚫ Randomly select a number within the sampling interval to identify first case to be sampled, counting from the first case on the list ⚫ After the first case is selected, select every nth case, where n represents the sampling interval Systematic Random Sampling N = 100, and n = 20  sampling interval = 100/20 = 5 List persons from 1 to 100 Randomly select a number between 1 and 5 (ex : 4)  1st person selected = the 4th on the list  2nd person =4 + 5 = the 9th etc ..... Systematic Random Sampling Systematic Sample More representative of the population than SRS (better coverage of the population) More easily selected than SRS Can be used for selection from telephone book Systematic Sample Better to be avoided if there is periodicity in the sampled population. Example…..if we need to select a day every 7th day out of a list of attendees of a certain clinic over a year. The systematic manner of selection may result in inclusion of only Saturdays in the sample. Periodicity Stratified Random Sample Sample that has members from each segment of the population. Divide the population into groups (strata) and select a random sample from each group. Strata could be age groups, genders or levels of education, for example Stratified Random Sample