1. 1 1 Slide © 2007 Thomson South-Western. All Rights Reserved Chapter 7 Sampling and Sampling Distributions Sampling Distribution of Sampling Distribution of Introduction to Sampling Distributions Introduction to Sampling Distributions Point Estimation Point Estimation Simple Random Sampling Simple Random Sampling Other Sampling Methods Other Sampling Methods Sampling Distribution of Sampling Distribution of

2. 2 2 Slide © 2007 Thomson South-Western. All Rights Reserved The purpose of statistical inference is to obtain The purpose of statistical inference is to obtain information about a population from information information about a population from information contained in a sample. contained in a sample. The purpose of statistical inference is to obtain The purpose of statistical inference is to obtain information about a population from information information about a population from information contained in a sample. contained in a sample. Statistical Inference A population is the set of all the elements of interest. A population is the set of all the elements of interest. A sample is a subset of the population. A sample is a subset of the population.

3. 3 3 Slide © 2007 Thomson South-Western. All Rights Reserved The sample results provide only estimates of the The sample results provide only estimates of the values of the population characteristics. values of the population characteristics. The sample results provide only estimates of the The sample results provide only estimates of the values of the population characteristics. values of the population characteristics. A parameter is a numerical characteristic of a A parameter is a numerical characteristic of a population. population. A parameter is a numerical characteristic of a A parameter is a numerical characteristic of a population. population. With proper sampling methods, the sample results With proper sampling methods, the sample results can provide “good” estimates of the population can provide “good” estimates of the population characteristics. characteristics. With proper sampling methods, the sample results With proper sampling methods, the sample results can provide “good” estimates of the population can provide “good” estimates of the population characteristics. characteristics. Statistical Inference

4. 4 4 Slide © 2007 Thomson South-Western. All Rights Reserved Simple Random Sampling: Finite Population n Finite populations are often defined by lists such as: Organization membership roster Organization membership roster Credit card account numbers Credit card account numbers Inventory product numbers Inventory product numbers n A simple random sample of size n from a finite population of size N is a sample selected such population of size N is a sample selected such that each possible sample of size n has the same that each possible sample of size n has the same probability of being selected. probability of being selected.

5. 5 5 Slide © 2007 Thomson South-Western. All Rights Reserved Simple Random Sampling: Finite Population In large sampling projects, computer-generated In large sampling projects, computer-generated random numbers are often used to automate the random numbers are often used to automate the sample selection process. sample selection process. Sampling without replacement is the procedure Sampling without replacement is the procedure used most often. used most often. Replacing each sampled element before selecting Replacing each sampled element before selecting subsequent elements is called sampling with subsequent elements is called sampling with replacement. replacement.

6. 6 6 Slide © 2007 Thomson South-Western. All Rights Reserved n Infinite populations are often defined by an ongoing process whereby the elements of the population consist of items generated as though the process would operate indefinitely. Simple Random Sampling: Infinite Population n A simple random sample from an infinite population is a sample selected such that the following conditions is a sample selected such that the following conditions are satisfied. are satisfied. Each element selected comes from the same Each element selected comes from the same population. population. Each element is selected independently. Each element is selected independently.

7. 7 7 Slide © 2007 Thomson South-Western. All Rights Reserved s is the point estimator of the population standard s is the point estimator of the population standard deviation . deviation . s is the point estimator of the population standard s is the point estimator of the population standard deviation . deviation . In point estimation we use the data from the sample In point estimation we use the data from the sample to compute a value of a sample statistic that serves to compute a value of a sample statistic that serves as an estimate of a population parameter. as an estimate of a population parameter. In point estimation we use the data from the sample In point estimation we use the data from the sample to compute a value of a sample statistic that serves to compute a value of a sample statistic that serves as an estimate of a population parameter. as an estimate of a population parameter. Point Estimation We refer to as the point estimator of the population We refer to as the point estimator of the population mean . mean . We refer to as the point estimator of the population We refer to as the point estimator of the population mean . mean . is the point estimator of the population proportion p. is the point estimator of the population proportion p.

8. 8 8 Slide © 2007 Thomson South-Western. All Rights Reserved Sampling Error Statistical methods can be used to make probability Statistical methods can be used to make probability statements about the size of the sampling error. statements about the size of the sampling error. Sampling error is the result of using a subset of the Sampling error is the result of using a subset of the population (the sample), and not the entire population (the sample), and not the entire population. population. The absolute value of the difference between an The absolute value of the difference between an unbiased point estimate and the corresponding unbiased point estimate and the corresponding population parameter is called the sampling error. population parameter is called the sampling error. When the expected value of a point estimator is equal When the expected value of a point estimator is equal to the population parameter, the point estimator is said to the population parameter, the point estimator is said to be unbiased. to be unbiased.

9. 9 9 Slide © 2007 Thomson South-Western. All Rights Reserved Sampling Error n The sampling errors are: for sample proportion for sample standard deviation for sample mean

10. 10 Slide © 2007 Thomson South-Western. All Rights Reserved Example: St. Andrew’s St. Andrew’s College receives St. Andrew’s College receives 900 applications annually from prospective students. The application form contains a variety of information including the individual’s scholastic aptitude test (SAT) score and whether or not the individual desires on-campus housing.

11. 11 Slide © 2007 Thomson South-Western. All Rights Reserved Example: St. Andrew’s The director of admissions The director of admissions would like to know the following information: the average SAT score for the average SAT score for the 900 applicants, and the 900 applicants, and the proportion of the proportion of applicants that want to live on campus.

12. 12 Slide © 2007 Thomson South-Western. All Rights Reserved Example: St. Andrew’s We will look at two alternatives for obtaining the desired information. n Conducting a census of the entire 900 applicants entire 900 applicants n Selecting a sample of 30 applicants, using Excel applicants, using Excel

13. 13 Slide © 2007 Thomson South-Western. All Rights Reserved Conducting a Census n If the relevant data for the entire 900 applicants were in the college’s database, the population parameters of interest could be calculated using the formulas presented in Chapter 3. n We will assume for the moment that conducting a census is practical in this example.

14. 14 Slide © 2007 Thomson South-Western. All Rights Reserved Conducting a Census n Population Mean SAT Score n Population Standard Deviation for SAT Score n Population Proportion Wanting On-Campus Housing

15. 15 Slide © 2007 Thomson South-Western. All Rights Reserved Simple Random Sampling The applicants were numbered, from 1 to 900, as The applicants were numbered, from 1 to 900, as their applications arrived. their applications arrived. She decides a sample of 30 applicants will be used. She decides a sample of 30 applicants will be used. Furthermore, the Director of Admissions must obtain Furthermore, the Director of Admissions must obtain estimates of the population parameters of interest for estimates of the population parameters of interest for a meeting taking place in a few hours. a meeting taking place in a few hours. Now suppose that the necessary data on the Now suppose that the necessary data on the current year’s applicants were not yet entered in the current year’s applicants were not yet entered in the college’s database. college’s database.

16. 16 Slide © 2007 Thomson South-Western. All Rights Reserved n Taking a Sample of 30 Applicants Excel’s RAND function generates Excel’s RAND function generates random numbers between 0 and 1 random numbers between 0 and 1 Excel’s RAND function generates Excel’s RAND function generates random numbers between 0 and 1 random numbers between 0 and 1 Simple Random Sampling: Using Excel Step 1: Assign a random number to each of the 900 applicants. applicants. Step 2: Select the 30 applicants corresponding to the 30 smallest random numbers. 30 smallest random numbers.

17. 17 Slide © 2007 Thomson South-Western. All Rights Reserved as Point Estimator of  as Point Estimator of  n as Point Estimator of p Point Estimation Note: Different random numbers would have identified a different sample which would have resulted in different point estimates. s as Point Estimator of  s as Point Estimator of 

18. 18 Slide © 2007 Thomson South-Western. All Rights Reserved PopulationParameterPointEstimatorPointEstimateParameterValue  = Population mean SAT score SAT score 990997  = Population std. deviation for deviation for SAT score SAT score 80 s = Sample std. s = Sample std. deviation for deviation for SAT score SAT score75.2 p = Population pro- portion wanting portion wanting campus housing campus housing.72.68 Summary of Point Estimates Obtained from a Simple Random Sample = Sample mean = Sample mean SAT score SAT score = Sample pro- = Sample pro- portion wanting portion wanting campus housing campus housing

19. 19 Slide © 2007 Thomson South-Western. All Rights Reserved n Process of Statistical Inference The value of is used to make inferences about the value of . The sample data provide a value for the sample mean. A simple random sample of n elements is selected from the population. Population with mean  = ? Sampling Distribution of

20. 20 Slide © 2007 Thomson South-Western. All Rights Reserved The sampling distribution of is the probability distribution of all possible values of the sample mean. Sampling Distribution of where:  = the population mean  = the population mean E ( ) =  Expected Value of

21. 21 Slide © 2007 Thomson South-Western. All Rights Reserved Sampling Distribution of Finite Population Infinite Population is referred to as the standard error of the mean. is referred to as the standard error of the mean. A finite population is treated as being A finite population is treated as being infinite if n / N <.05. infinite if n / N <.05. is the finite correction factor. is the finite correction factor. Standard Deviation of

22. 22 Slide © 2007 Thomson South-Western. All Rights Reserved Form of the Sampling Distribution of Central Limit Theorem: If we use a large ( n > 30) simple random sample, the sampling distribution of can be approximated by a normal distribution. If the simple random sample is small ( n < 30), the sampling distribution of can be considered normal only if we assume the population has a normal distribution.

23. 23 Slide © 2007 Thomson South-Western. All Rights Reserved Sampling Distribution of for SAT Scores SamplingDistributionof

24. 24 Slide © 2007 Thomson South-Western. All Rights Reserved What is the probability that a simple random sample What is the probability that a simple random sample of 30 applicants will provide an estimate of the population mean SAT score that is within +/  10 of the actual population mean  ? In other words, what is the probability that will be In other words, what is the probability that will be between 980 and 1000? Sampling Distribution of for SAT Scores

25. 25 Slide © 2007 Thomson South-Western. All Rights Reserved Relationship Between the Sample Size and the Sampling Distribution of and the Sampling Distribution of Suppose we select a simple random sample of 100 Suppose we select a simple random sample of 100 applicants instead of the 30 originally considered. applicants instead of the 30 originally considered. E ( ) =  regardless of the sample size. In our E ( ) =  regardless of the sample size. In our example, E ( ) remains at 990. example, E ( ) remains at 990. Whenever the sample size is increased, the standard Whenever the sample size is increased, the standard error of the mean is decreased. With the increase error of the mean is decreased. With the increase in the sample size to n = 100, the standard error of the in the sample size to n = 100, the standard error of the mean is decreased to: mean is decreased to:

26. 26 Slide © 2007 Thomson South-Western. All Rights Reserved Relationship Between the Sample Size and the Sampling Distribution of and the Sampling Distribution of With n = 30, With n = 100,

27. 27 Slide © 2007 Thomson South-Western. All Rights Reserved Recall that when n = 30, P (980 < < 1000) =.5034. Recall that when n = 30, P (980 < < 1000) =.5034. Relationship Between the Sample Size and the Sampling Distribution of and the Sampling Distribution of We follow the same steps to solve for P (980 < < 1000) We follow the same steps to solve for P (980 < < 1000) when n = 100 as we showed earlier when n = 30. when n = 100 as we showed earlier when n = 30. Now, with n = 100, P (980 < < 1000) =.7888. Now, with n = 100, P (980 < < 1000) =.7888.

28. 28 Slide © 2007 Thomson South-Western. All Rights Reserved A simple random sample of n elements is selected from the population. Population with proportion p = ? n Making Inferences about a Population Proportion The sample data provide a value for the sample proportion. The value of is used to make inferences about the value of p. Sampling Distribution of

29. 29 Slide © 2007 Thomson South-Western. All Rights Reserved Sampling Distribution of where: p = the population proportion The sampling distribution of is the probability distribution of all possible values of the sample proportion. Expected Value of

30. 30 Slide © 2007 Thomson South-Western. All Rights Reserved is referred to as the standard error of the is referred to as the standard error of the proportion. Sampling Distribution of Finite Population Infinite Population Standard Deviation of

31. 31 Slide © 2007 Thomson South-Western. All Rights Reserved The sampling distribution of can be approximated The sampling distribution of can be approximated by a normal distribution whenever the sample size by a normal distribution whenever the sample size is large. is large. The sampling distribution of can be approximated The sampling distribution of can be approximated by a normal distribution whenever the sample size by a normal distribution whenever the sample size is large. is large. The sample size is considered large whenever these The sample size is considered large whenever these conditions are satisfied: conditions are satisfied: The sample size is considered large whenever these The sample size is considered large whenever these conditions are satisfied: conditions are satisfied: np > 5 n (1 – p ) > 5 and Form of the Sampling Distribution of

32. 32 Slide © 2007 Thomson South-Western. All Rights Reserved For values of p near.50, sample sizes as small as 10 permit a normal approximation. For values of p near.50, sample sizes as small as 10 permit a normal approximation. With very small (approaching 0) or very large (approaching 1) values of p, much larger samples are needed. Form of the Sampling Distribution of

33. 33 Slide © 2007 Thomson South-Western. All Rights Reserved Recall that 72% of the Recall that 72% of the prospective students applying to St. Andrew’s College desire on-campus housing. n Example: St. Andrew’s College Sampling Distribution of What is the probability that What is the probability that a simple random sample of 30 applicants will provide an estimate of the population proportion of applicant desiring on-campus housing that is within plus or minus.05 of the actual population proportion?

34. 34 Slide © 2007 Thomson South-Western. All Rights Reserved Other Sampling Methods n Stratified Random Sampling n Cluster Sampling n Systematic Sampling n Convenience Sampling n Judgment Sampling