Your SlideShare is downloading.
×

- 1. Introduction to Biostatistics and Types of Sampling Methods SUBMITTED BY DR. SUNITA OJHA ASSISTANT PROFESSOR SURESH GYAN VIHAR UNIVERSITY
- 2. • Statistics • Statistics is the study of the collection, organization, analysis, interpretation, and presentation of data. • Biostatistics • Application in biological experiment designing, data collection, analysis and interpretation. Biostatistics is applied in various fields such as medicine, pharmacy, agriculture, and fisheries. • Data • Data in statistics are based on individual observations. • Data can be counted or measured. It represents varying values of variable. • Variable • Any quantity or quality that shows variation from one individual to the other in same population is known as variable. • E.g. Plant height, height of adult male, weight of preschool children. Terminologies
- 3. • Qualitative variable • E.g. color of flower petal • Quantitative variable • E.g. Height of plant • Continuous variable • Continuous variables are those which can have any value within a certain range exhibited by the population. • Quantitative variables belong to this group. • E.g. weight, height, volume, time. • Discontinuous or Discrete variables • Variables with certain finite values without any intermediate values are discontinuous variables. • All qualitative variables and some quantitative variables are discrete • E.g. Organisms per unit area.
- 4. • Independent Variable • The variable which is manipulated by the investigator. • Dependent Variable • The variable which is observed or measured. • Control/Constant Variable • This is the variable which the investigator wants to keep constant in her experiment. Water (2L) Temp-100°C Time- 10 mins Water (2L) Temp-100°C Time- 30 mins Water (2L) Temp-100°C Time- 50 mins Independent Variable: Time Dependent Variable: Volume of water Control Variable: Temperature, atmospheric pressure Figure 1. Illustration of independent/ dependent/constant variables
- 5. • Sampling • Selection of a part of a population to represent a whole population is known as sampling • Sample • The part selected is known as sample • Population • The total no. of individual observation from which observations has to be made at particular time. • Finite: If a population consists of a fixed no. of values. E.g. Number of plants in a quadrat • Infinite: Population in which it is theoretically impossible to observe all the values (unlimited size). E.g. no. of phytoplankton in a pond. • Sampling • Complete coverage of population is too time consuming and expensive. Therefore a part of the population is selected that can represent the population. Population Sample
- 6. • A sample should possess the following essentials. • Selected samples should be homogeneous and should not have any difference when compared with population • More number of items are to be included for reliable results • The individual items in the sample should be independent of each other. • Advantages • Saves time. • Reduces cost. • Only method to study infinite population. • Gives more accurate result. • Shortcomings • Sampling is difficult for small population size. • Result can be faulty, misleading if sampling is not done properly. • Personal biasness results faulty results.
- 7. • Sampling Size • It is an important factor to be considered during sampling. • It should note be too small or too large. • The sampling size depends on the type of population. • Homogeneous population: Small sample size is required. • Heterogeneous population: Large sampling size is required. • For data classification large sized sample is required. • Methods of sampling • There are various methods of sampling which is listed in figure no. 3. The selection of sampling methods depends on the purpose of sampling and the nature of population. • Random Sampling Methods • Random sampling • In this method each item of the population has an equal chance of being included in sample. This method is also known as “unrestricted random sampling”.
- 8. Random Sampling Methods Random Sampling Stratified Sampling Systematic Sampling Cluster Sampling Non-Random Sampling Methods Judgmental Sampling Convenience Sampling Quota Sampling Lottery Method Random Numbers Sampling Methods Figure 2. Various sampling techniques
- 9. • Lottery Method • This is the most popular and simplest method of selecting a random sample from a finite population. • In this method, all items of population are numbered on separate slips of paper of identical size, shape and color. • These slips are folded and mixed up in a box and a blindfold selection is made. • For required sample size, the same number of slips are selected. • This method is in applicable for infinite population. • Random Numbers • This method is applied for large size population. • The random selection is conveniently done with the help of these tables of random numbers. • Tippets Random Number Tables: 1040 numbers of four digits each • Fisher and Yates Tables: 15000 numbers with two digits each • C.R. Rao, Mitra and Matahi Table of Random Numbers: 20000 digits grouped into 50000 sets of 4 digit number. • Snedecor and Cochran Random Number Tables: 1000 random numbers
- 10. • One can use the table of random numbers from any position either horizontally or vertically. Or one Can blindly start from any position of a table. • If we want to select 10 pods from 200 pods each pod should be assigned a number from 001- 200. • In the 5 digit Snedecor and Cochran Random Number Tables we can take three digits into consideration and other two are ignore. 05217 03164 19774 12696 05437 17805 09609 09284 17771 • Thus the sample selected will be • 052, 031, 197, 126, 054, 178, 096, 092, 177
- 11. • Stratified Sampling • Stratified sampling is done when population is heterogeneous with respect to variable or characteristic under study. • Here population is divided into relatively homogeneous groups or strata and a random sample is drawn from each group. • Systematic Sampling • For this we arrange the items in numerical, alphabetical, geographical or any other order. • Now the items are serially numbered. The first item is selected at random. • E.g. If we want to select a sample of 10 trees from 100 trees of a forest by taking every kth tree where ‘k’ refers to the sampling interval. k= N/n N= Population size N=Sample size • Here k=100/10=10. So evert 10th tree is taken as sample. i.e. 10th, 20th ,30th……
- 12. • Cluster Sampling • Widely used in geographical studies and when the units are spread over large geographical area. • This area can be divide into different clusters and data are collected from these clusters. • Non –Random Sampling Methods • Judgement, purposive or deliberate sampling • In this method choice of sample items depends exclusively on the judgement of the investigator. • E.g. 8 out of 20 persons use a particular type of toothpaste. So the investigator has selected those 8 person for his study. • Convenience sampling • Convenience sampling is a non-probability sampling technique where subjects are selected because of their convenient accessibility and proximity to the researcher • Quota sampling: • Here sample quotas are fixed according to any characteristic of the population like income, age, sex, religion.
- 13. References Khan, I. A., & Khanum, A. (1994). Fundamentals of biostatistics. Ukaaz. Sharma, A. K. (2005). Text book of biostatistics I. Discovery Publishing House. Daniel, W. W., & Cross, C. L. (2018). Biostatistics: a foundation for analysis in the health sciences. Wiley.