1. 1 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Chapter 17 Testing Hypotheses About Proportions

2. 2 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Objectives 60. Perform a one-proportion z-test, to include: writing appropriate hypotheses, checking the necessary assumptions, drawing an appropriate diagram, computing the P-value, making a decision, and interpreting the results in the context of the problem.

3. 3 Copyright © 2014, 2012, 2009 Pearson Education, Inc. 17.1 Hypotheses

4. 4 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Cracking Rate < 20%? Previously the cracking rate of casts was 20% After a new engineering process the cracking rate of 400 casts fell to 17%. Is this due to the new engineering or just random chance?

5. 5 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Hypotheses H 0 : H 0 usually states that there’s nothing different. H 0 : parameter = hypothesized value Note the parameter describes the population not the sample. H 0 is called the null hypothesis. H A : H A is a statement that something has changed, gotten bigger or smaller H A is called the alterative hypothesis.

6. 6 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Cracking Rate < 20%? Previously the cracking rate of casts was 20% After a new engineering process the cracking rate of 400 casts fell to 17%. Is this due to the new engineering or just random chance? Null Hypothesis: Nothing has changed H 0 : parameter = hypothesized value H 0 : p = 0.20 Alternative Hypothesis: H A : p < 0.20

7. 7 Copyright © 2014, 2012, 2009 Pearson Education, Inc. How Small to Convince Us? Had the new cracking rate been 1%, it would clearly indicate a change from 20%. Extremely unlikely that this could happen just by random chance Had the new cracking rate been 19.8%, we would be skeptical. Not so unlikely to be just random chance How about 17%? How likely is it that a random sample would have a cracking rate 17% or less?

8. 8 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Slide 1- 8 One-Proportion z-Test The conditions for the one-proportion z-test are the same as for the one proportion z-interval. (Independence, Randomization Condition, 10% Condition, and Success/Failure Condition) We test the hypothesis H 0 : p = p 0 using the statisticwhere When the conditions are met and the null hypothesis is true, this statistic follows the standard Normal model, so we can use that model to obtain a P-value (the probability of seeing our data if the null hypothesis is true). Notice that here we are finding the SD(p^) not the SE(p^)

9. 9 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Checking Conditions and Finding the Standard Error Checking Conditions: n = 400, p = 0.20 Success/Failure Condition np = (400)(0.20) = 80  10 nq = (400)(0.80) = 320  10 Independence plausible The Normal model applies. Find the standard deviation of the model. Note: Use p and not to find standard deviation.

10. 10 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Using the Normal Model p = 0.20, = 0.17, = 0.02 If the null hypothesis is true that the cracking rate is still equal to 20%, then the probability of observing a cracking rate of 17% in a random sample of 400 is 6.7%.

11. 11 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Using the TI-83+ Stat…Tests…1-PropZtest p0 is the hypothecated proportion.20 x: is the number of favorable observations we saw in the sample … If x isn’t given solve it using x = n*p^ = 400*.17 = 68 n: is the number in your sample 400 prop: choose the sign of the alternative hypothesis: < Choose calculate Result: z=-1.5 (this is the “test statistic” P-value =.0668 Slide 1- 11

12. 12 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Innocent until Proven Guilty Begin with the presumption of innocence (H 0 ). Collect evidence. Bank money in house Still wearing mask Getaway car found in his name Evidence beyond a reasonable doubt? Is 5% small enough chance? How about 1%? 6.7%?

13. 13 Copyright © 2014, 2012, 2009 Pearson Education, Inc. 17.2 P-Values

14. 14 Copyright © 2014, 2012, 2009 Pearson Education, Inc. The P-Value and Surprise The P-value is the probability of seeing data like these (or even more unlikely data) given the null hypothesis is true. Tells us how surprised we would be to get these data given H 0 is true. P-value small: Either H 0 is not true or something remarkable occurred P-value not small enough: Not a surprise. Data consistent with the model. Do not reject H 0.

15. 15 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Guilty or Not Enough Evidence? Defendant is either Guilty: P-value too small. The evidence is clear. Not Guilty: P-value not small enough. The evidence is not sufficient. Not the same as innocent. Maybe innocent or maybe guilty, but not enough evidence found. Two Choices Fail to reject H 0 if P-value large. Never accept H 0. Reject H 0 if P-value is small. Accept H A.

16. 16 Copyright © 2014, 2012, 2009 Pearson Education, Inc. When the P-Value is Not Small Don’t say: Accept H 0. We have proven H 0. Say: Fail to reject H 0 There is insufficient evidence to reject H 0. H 0 may or may not be true. Example: H 0 : All swans are white. If we sample 100 swans that are all white, there could still be a black swan.

17. 17 Copyright © 2014, 2012, 2009 Pearson Education, Inc. 17.3 The Reasoning of Hypothesis Testing

18. 18 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Step 1: State the Hypotheses H 0 : H 0 usually states that there’s nothing different. H 0 : parameter = hypothesized value Note the parameter describes the population not the sample. H 0 is called the null hypothesis. H A : H A is a statement that something has changed, gotten bigger or smaller H A is called the alterative hypothesis.

19. 19 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Hypotheses About the DMV The DMV claims 80% of all drivers pass the driving test. In a survey of 90 teens, only 61 passed. Is there evidence that teen pass rates are below 80%? H 0 : p = 0.80 H A : p < 0.80

20. 20 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Step 2: Model Decide on the model to test the null hypothesis and parameter. Check conditions – independence: 10% condition, Randomization – sample size: Success/Failure. If the conditions are not met, either quit or redesign the study. Normal models use z-scores. Other models may not use z-scores. Name the model, e.g. 1-proportion z-test.

21. 21 Copyright © 2014, 2012, 2009 Pearson Education, Inc. 1-Proportion z-Test Conditions Same as a 1-Proportion z-Interval Null Hypothesis H 0 : p = p 0 Test Statistics

22. 22 Copyright © 2014, 2012, 2009 Pearson Education, Inc. DMV Study: Checking Conditions Randomization Condition: The 90 teens were a random sample of all teens. 10% Condition: 90 is fewer than10% of the total number of all teens who take the driving test. Success/Failure Condition: np 0 = (90)(0.80) = 72  10 nq 0 = (90)(0.20) = 12  10 The conditions are satisfied. We can use the Normal model and perform a 1-Proportion z-Test.

23. 23 Copyright © 2014, 2012, 2009 Pearson Education, Inc. DMV Study: Mechanics Claim: 80% pass. 61 of 90 teens tested passed. Find P-value. n = 90, x = 61, p 0 = 0.80, Try this Using your TI-83+

24. 24 Copyright © 2014, 2012, 2009 Pearson Education, Inc. DMV Study: Conclusion Is the teen pass rate less than 80%? P-value = 0.002 What can be concluded? What does the P-value mean? P-value = 0.002 is very small → Reject H 0 The survey data provide strong evidence that the pass rate for teens is less than 80%. This should not be the end of the conversation. The next step would be to see if the pass rate is low enough to take further action.

25. 25 Copyright © 2014, 2012, 2009 Pearson Education, Inc. 17.4 Alternative Alternatives

26. 26 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Two-Sided Test For the new process the engineer may be interested in whether there has been a change in the cracking rate not just a decrease. H 0 : p = 0.20 H A : p ≠ 0.20 An alternative hypothesis where we are interested in deviation on either side is called a two-sided alternative. The P-value is the probability of deviating from either direction from the null hypothesis.

27. 27 Copyright © 2014, 2012, 2009 Pearson Education, Inc. One-Sided Test The engineer may be interested in whether there has been a decrease in the cracking rate. H 0 : p = 0.20 H A : p < 0.20 An alternative hypothesis where we are interested in deviation on only one side is called a one-sided alternative. The P-value for a one-sided alternative is always half the P-value for the two-sided alternative.

28. 28 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Hypothesis Writing Examples Write the null and alternative hypotheses and draw the appropriate curve. In the 1950s only about 40% of high school graduates went on to college. Has the percentage changed? 20% of the cars of a certain model have needed costly transmission work after being driven between 50,000 and 100,000 miles. The manufacturer hopes that the redesign of the transmission has solved the problem. We field test a new flavor of soft drink, planning to market it only if we are sure that at least 60% of the people like the flavor. Slide 1- 28

29. 29 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Hypothesis Writing Examples Write the null and alternative hypotheses and draw the appropriate curve. The drug Lipitor is meant to lower cholesterol. Is there evidence to support the claim that over 1.9% of the users experience flu like symptoms as a side effect? According to the US dept. of Health, 16.3% of Americans did not have health insurance coverage in 1998. A politician claims that this percentage has decreased since 1998. During the past 40 years, the monthly rate of return for a particular item has been 4.2 percent. A store analyst claims that it is different. Slide 1- 29

30. 30 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Not the Right Proportion of Male Babies? Under natural conditions, 51.7% of births are male. In Punjab India’s hospital 56.9% of the 550 births were male. Question: Is there evidence that the proportion of male births is different for this hospital? Think → Plan: We will have a two-tailed alternative. The parameter of interest is p. H 0 : p = 0.517 H A : p ≠ 0.517

31. 31 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Not the Right Proportion of Male Babies? Think → Model: Check the conditions Independence Assumption: The sex of one baby should not affect the sex of others. Randomization Conditions: The births were not random, so be cautious of the results. 10% Condition: 550 births is certainly less than the total number of all births.

32. 32 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Not the Right Proportion of Male Babies? Think → Model: Check the conditions (Continued). Success/Fail Condition: (550)(0.517)  10, (550)(0.583)  10 The Normal model does apply. We can use a one-proportion z-test.

33. 33 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Not the Right Proportion of Male Babies? Show → Mechanics:

34. 34 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Not the Right Proportion of Male Babies? Show → Mechanics: P-value = 2P(z > 2.44) = 2(0.0073) = 0.0146

35. 35 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Not the Right Proportion of Male Babies? Tell → Conclusion: Interpreting the P-value = 0.00146 – If the proportion of male babies were still 51.7%, then an observed proportion as different as 56.9% male babies would occur at random about 15 times in 1000. – This is so small a chance that I reject H 0. There is strong evidence that the proportion of boys is not equal to the baseline for the region. It appears larger.

36. 36 Copyright © 2014, 2012, 2009 Pearson Education, Inc. 17.5 P-Values and Decisions: What to Tell About a Hypothesis Test

37. 37 Copyright © 2014, 2012, 2009 Pearson Education, Inc. How Small a P-Value is Small Enough? How small is small enough is context specific. Test to see if a renowned musicologist can distinguish between Mozart and Hayden. P-value of 0.1 may be good enough. Just reaffirming known talent. A friend claims psychic abilities and can predict heads or tails. Very small P-value such as 0.01 needed. Breaking scientific theory.

38. 38 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Acceptable P-Value Depends on Result’s Importance Proportion of students with full time jobs has increased. Not that important. P-value = 0.05 will work. Testing the proportion of faulty rivets that hold together the wings of a commercial aircraft is now below the danger threshold. Life and death decision. Need a very small P-value Whether rejecting or failing to reject, always cite the P-value. An accompanying confidence interval helps also.

39. 39 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Home Field Advantage? Is there a home field advantage in baseball? The home team won 1277 of the 2429 (52.57%) games played in the season. Is there evidence to suggest that the home team wins more than 50%? Think → Plan: p = proportion of home team wins Hypotheses H 0 : p = 0.50 H A : p > 0.50

40. 40 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Home Field Advantage? Model → Independence Assumption: Questionable, but the 2011 season may be representative of all games past and future. 10% Condition: The 2011 season is less than 10% of all games played past and future. Success/Failure Condition: np = (2429)(0.5)  10 nq = (2429)(0.5)  10 The Normal Model Applies: Conduct a one-proportion z-test.

41. 41 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Home Field Advantage? Show → Mechanics: The P-value is about 0.0057.

42. 42 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Home Field Advantage? Tell → Conclusion: The P-value of 0.0057 says that if the true proportion of home teams wins is 0.5 then an observed value of 0.5257 would occur less than 6 times in 1000. This is so unlikely, so reject H 0. There is reasonable evidence that the true proportion of home team wins is greater than 50%. There appears to be a home field advantage.

43. 43 Copyright © 2014, 2012, 2009 Pearson Education, Inc. How Big a Home Field Advantage? Think → Model: Success Failure Condition – Home team wins: 1277  10 – Home team losses: 1152  10 Sampling distribution follows the Normal model. Find the one-proportion z-interval.

44. 44 Copyright © 2014, 2012, 2009 Pearson Education, Inc. How Big a Home Field Advantage? Show → Mechanics: For 95% confidence, z* = 1.96. A 95% confidence interval is: 0.5257 ± 0.0198 or (0.5059, 5455).

45. 45 Copyright © 2014, 2012, 2009 Pearson Education, Inc. How Big a Home Field Advantage? Tell → Conclusion: I am 95% confident that, in professional baseball, home teams win between 50.59% and 54.56% of the games. For a 162-game season, the low end gives the home team about 1/2 of an extra victory and the high end, about 4 extra wins.

46. 46 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Practice using the TI-83+ It is believed that the percent of convicted felons who have a history of juvenile delinquency is 70%. Is there evidence to support the claim the the actual percentage is more than the 70% if out of 200 convicted felons, we find that 154 have a history of juvenile delinquency? Alpha =.05 Stat…Tests…1-PropZtest p0 is the hypothecated proportion.70 x: is the number of favorable observations from the sampling 154 n: is the number in your sample 200 prop: choose the sign of the alternative hypothesis- > Choose calculate You should get a p value of.015 therefore, reject the null hypothesis which means that we reject the null hypothesis. The claim that more than 70% of the felons have a history of juvenile delinquency is supported by the data. Slide 1- 46

47. 47 Copyright © 2014, 2012, 2009 Pearson Education, Inc. More examples There are supposed to be 20% orange M&Ms in a bag. Suppose a bag of 122 has only 21 orange ones. Does this contradict the company’s 20% claim? Note that this is a 2-tailed test… so H 0 : p=0.20 and H A : p≠0.20. Check the model assumptions/conditions! Slide 1- 47

48. 48 Copyright © 2014, 2012, 2009 Pearson Education, Inc. Practice In the 1980s it was generally believed that autism affected about 6% of the nation’s children. Some people believe that the increase in the number of chemicals in the environment has led to an increase in the incidence of autism. A recent study examined 384 children and found that 46 of them showed signs of some form of autism. Is there strong evidence that the level of autism has increased? (Use an alpha of.05) Write the hypotheses, check the assumptions, draw the curve, find the pertinent statistics and critical values, find the p value, state your conclusion, etc. H0: p=0.06 HA: p>0.06 Test statistic = z = 4.93 P-value = 0.0000004 (i.e. a really small #) Conclusion: Since the P-value is less than alpha, we can reject the hypothesis that the true population proportion of kids with autism is 6%. The statistical evidence indicates that the true rate of autism has probably increased. Slide 1- 48

49. 49 Copyright © 2014, 2012, 2009 Pearson Education, Inc. 14. During the 2000 season, the home team won 138 of the 240 regular season games. Is this strong evidence of a home field advantage? (Let alpha=0.05) H0: p=0.50 HA: p>0.50 Test statistic = z = 2.32 P-value = 0.01 Conclusion: Since the P-value is less than alpha, we can reject the hypothesis that there is no home field advantage. Therefore, the statistical evidence suggests that there IS a home field advantage. Slide 1- 49

50. 50 Copyright © 2014, 2012, 2009 Pearson Education, Inc. 15. A garden center wants to store leftover packets of vegetable seeds for sale the following spring but the center is concerned that the seeds may not germinate at the same rate a year later. The manager finds a packet from the previous year and plants the seeds as a test. Although the package claims a rate of at least 92% only 171 of the 200 seeds sprout. Is this evidence that the seeds have lost their viability during the year in storage? (Let alpha=0.05) H0: p=0.92 HA: p<0.92 Test statistic = z = -3.39 P-value = 0.00035 Conclusion: Since the P-value is smaller than alpha, we reject the hypothesis that the true percentage of germinated seeds is 92%. The statistical evidence suggests that the seeds lose their effectiveness after a year. Slide 1- 50

51. 51 Copyright © 2014, 2012, 2009 Pearson Education, Inc. What Can Go Wrong? Don’t base your H 0 on what you see in the data. Changing the null hypothesis after looking at the data is just wrong. Don’t base your H A on the data. Both the null and alternative hypotheses must be stated before peeking at the data. Don’t make H 0 what you want to show to be true. H 0 represents the status quo. You can never accept the null hypothesis.

52. 52 Copyright © 2014, 2012, 2009 Pearson Education, Inc. What Can Go Wrong? Don’t forget to check the conditions. Randomization, 10% Condition, and Success/Failure Condition Don’t accept the null hypothesis. You can only say you don’t have evidence to reject H 0. If you fail to reject H 0 don’t expect a larger sample would reject H 0. Check the confidence interval. If its values would not matter to you, then a larger sample will unlikely be worthwhile.