SAMPLING AND NORMAL DISTRIBUTION.pptx pgs
- 2. LEARNING OBJECTIVES •What is sampling? •Why we need sampling? •What is the characteristics of a good sample? •Types of sampling methods
- 3. POPULATION A group of items having one or more common characteristics. SAMPLE A representative part of a population
- 4. SAMPLING •Process of choosing of a sample from a population.
- 5. Target population: The population to be studied/ to which the investigator wants to generalize his results. Sampling unit: Smallest unit from which sample can be selected Sampling frame List of all the sampling units from which sample is drawn Sampling scheme Method of selecting sampling units from sampling frame
- 8. CHARACTERISTICS OF GOOD SAMPLE •Truly representative •Should have all the characteristics in population •Have small sampling error •Economically viable •Optimum size •Results can be generalised to population
- 9. WHY SAMPLING •Feasibility •Limited Resources •Duration Of Study Can Be Reduced •Increased Accuracy
- 10. SAMPLING Advantages •Reduces demand on resources –finance, personnel and materials •Results are obtained quickly with better accuracy Disadvantages •Sampling error •Create a feeling of discrimination •Every unit in the population not legally recorded
- 11. TYPES OF SAMPLING •PROBABILITY SAMPLING/RANDOM •NON PROBABILITY SAMPLING/ NON RANDOM
- 12. SAMPLING METHODS
- 13. PROBABILITY SAMPLING •SIMPLE RANDOM SAMPLING •SYSTEMATIC SAMPLING •STRATIFIED SAMPLING •CLUSTER SAMPLING •MULTISTAGE SAMPLING •MULTIPHASE SAMPLING
- 14. SIMPLE RANDOM SAMPLING •Every sampling unit have an equal chance of being selected. • Suitable for Small, Homogeneous & Readily available • Done by lottery method/fish bowl Table of random numbers Computer generated random number
- 16. TABLE OF RANDOM NUMBERS 6 8 4 2 5 7 9 5 4 1 2 5 6 3 2 1 4 0 5 8 2 0 3 2 1 5 4 7 8 5 9 6 2 0 2 4 3 6 2 3 3 3 2 5 4 7 8 9 1 2 0 3 2 5 9 8 5 2 6 3 0 1 7 4 2 4 5 0 3 6 8 6
- 17. SYSTEMATIC RANDOM SAMPLING PRINCIPLE •Selecting one unit at random and additional units at regular sampling intervals
- 18. SYSTEMATIC SAMPLING EVERY KTH UNIT IS SELECTED K=SAMPLE INTERVAL K=TOTAL POPULATION /SAMPLE SIZE REQUIRED
- 19. STRATIFIED RANDOM SAMPLING •The population under study first divided into homogenous groups-strata •A sampling frame is created for each strata •The sample is drawn from each stratum at random in proportion to its size
- 20. STRATIFIED SAMPLING Draw a sample from each stratum
- 21. CLUSTER SAMPLING • When population is heterogeneous • Clusters are formed by grouping units on the basis of their geographic locations • A cluster sample obtained by selecting clusters on the basis of simple random sampling • From the selected clusters, every unit is included
- 22. CLUSTER SAMPLING Section 4 Section 5 Section 3 Section 2 Section 1
- 25. NON PROBABILITY SAMPLING •Convenient sampling •Judgment sampling •Quota sampling •Snow ball sampling
- 26. CONVENIENT SAMPLING Principle •Sample is selected according to choice and accessibility irrespective of the representativeness Use •For pilot studies to get basic idea about the study variable Disadvantage •Biases are maximum •Results are unsatisfactory in terms of drawing conclusion
- 27. JUDGMENT SAMPLING Principle: • Choice of study units depends exclusively on the judgment of the investigator Use: • To study effect of some unknown factors whose characteristics are known • In solving urgent public health problems Disadvantage: • Not scientifically proven appropriate method • Does not follow basic laws of sampling
- 28. QUOTA SAMPLING Population is divided into quotas according to some specific characteristics Eg; sample polling before elections •Number of study units selected within each quota depend on personal judgment of the investigator Used •In public opinion studies Disadvantage •Does not follow laws of sampling
- 29. SNOW BALL SAMPLING •Non-probability based sampling technique that can be used to gain access to such populations which is hidden or hard-to-reach (sensitive issues) •To create a snowball sample, there are two steps: (A) trying to identify one or more units in the desired population; and (b)using these units to find further units and so on until the sample size is met.
- 33. CHARACTERISTICS BELL SHAPED CURVE SYMMETRICAL AROUND MEAN OF DISTRIBUTION. MEAN = MEDIAN = MODE
- 34. SKEWED DISTRIBUTION Having extreme values on one or the other side of a distribution.
- 35. SKEWNESS •RIGHT / POSITIVE SKEWING: MEAN>MEDIAN>MODE (TAIL TOWARDS RIGHT), MEAN TOWARDS RIGHT END •LEFT / NEGATIVE SKEWING: MEAN<MEDIAN<MODE (TAIL TOWARDS LEFT), MEAN TOWARDS LEFT END
- 36. • MEASURE OF PEAK • MESOKURTOTIC – NORMAL DISTRIBUTION • LEPTOKURTOTIC - +VE KURTOSIS - NARROW PEAK - VALUES PILE UP NEAR MEAN • PLATYKURTOTIC - -VE KURTOSIS - FLAT PEAK - VALUES EVENLY DISTRIBUTED AROUND MEAN KURTOSIS
- 37. KURTOSIS
- 38. THANKYOU
Editor's Notes
- #7: Picture of sampling breakdown
- #17: For example, to select a systematic sample of 45 students from the list of 45,000 students at a university, the sampling interval k is 1000. Suppose the random integer we choose is 597. Then the students numbered 597, 1597, 2597, ..., 44,957 would be in the sample.
- #19: Theywereinterestedincomparingtheeducationalandworkforceexpe- riences of male and female graduates, so they stratified their sampling frame by gender and took separate random samples of male and female graduates. Because there were many more male than female graduates, they sampled a higher fraction of female graduates than male graduates in order to obtain comparable precisions for the two groups.
- #28: quota sampling, the population is divided into different subpopulations just as in stratified random sampling,but with one important difference: Probability sam- pling is not used to choose individuals in the subpopulation for thesampleThebigdrawbackofquotasamplingisthatwedonotknowiftheunitschosenfor the sample exhibit selection bias. If selection of units is totally up to the interviewer, she or he is likely to choose the most accessible members of the populationimpractical, overly costly, or considered unnecessary, or when the persons designing the sample just do not know any better