- Page 1 and 2: Principles of Hypothesis Testing fo
- Page 3 and 4: Objectives • Formulate questions
- Page 5 and 6: Estimation and Hypotheses ‣Infere
- Page 7 and 8: You Use Hypothesis Testing • Desi
- Page 9 and 10: Analysis Follows Design Questions
- Page 11 and 12: Estimation and Hypotheses Inference
- Page 13 and 14: Pictures, Not Numbers • Scatter p
- Page 15 and 16: Like the Washington Post Weather, T
- Page 17 and 18: Distributions • Parametric tests
- Page 19: Binary Distribution • Binomial di
- Page 23 and 24: Null Hypothesis • For superiority
- Page 25 and 26: Example Hypotheses • H 0 : μ 1 =
- Page 27 and 28: Use a 2-Sided Test • Almost alway
- Page 29 and 30: Outline Estimation and Hypotheses
- Page 31 and 32: Information at Hand • 1 or 2 samp
- Page 33 and 34: One Sample: Cholesterol l Sample Da
- Page 35 and 36: Situation • May be you are readin
- Page 37 and 38: Cholesterol Sample Data • Populat
- Page 39 and 40: Test Statistic • Basic test stati
- Page 41 and 42: Unknown Truth and the Data Data Tru
- Page 43 and 44: Type II Error (or, 1- Power) • β
- Page 45 and 46: Cholesterol Sample Data • N = 25
- Page 47 and 48: Experiment Develop hypotheses Colle
- Page 49 and 50: Z or Standard Normal Distributionib
- Page 51 and 52: How to test? ‣Rejection interval
- Page 53 and 54: Cholesterol Rejection Interval Usin
- Page 55 and 56: Side Note on t vs. Z • If s = σ
- Page 57 and 58: Z-test: Do Not Reject H 0 Z X −
- Page 59 and 60: Determining Statistical Significanc
- Page 61 and 62: Cholesterol: t-statistic X − μ 2
- Page 63 and 64: How to test? Rejection interval j
- Page 65 and 66: Cholesterol Example • P-value for
- Page 67 and 68: P-value Interpretation Reminders
- Page 69 and 70: Outline Estimation and Hypotheses H
- Page 71 and 72:
Cholesterol Confidence Interval Usi
- Page 73 and 74:
Cholesterol Confidence Interval Usi
- Page 75 and 76:
Hypothesis Testing and dConfidence
- Page 77 and 78:
Interpret a 95% Confidence Interval
- Page 79 and 80:
Take Home: Hypothesis Testing • M
- Page 81 and 82:
Take Home: CI • Meaning/interpret
- Page 83 and 84:
Outline Estimation and Hypotheses H
- Page 85 and 86:
Linear regression • Model for sim
- Page 87 and 88:
More Than One Covariate • Y i =
- Page 89 and 90:
Repeated Measures (3 or more time p
- Page 91 and 92:
Do Not Use Correlation. Use Regress
- Page 93 and 94:
Outline Estimation and Hypotheses H
- Page 95 and 96:
Omics • False negative (Type II e
- Page 97 and 98:
What do you need to think about?
- Page 99 and 100:
Little Diagnostic Testing Lingo •
- Page 101 and 102:
Example: Western vs. ELISA • 1 mi
- Page 103 and 104:
1% Prevalence • 10980 total test
- Page 105 and 106:
10% Prevalence • 99% PPV • 99.9
- Page 107 and 108:
Prevalence Matters • PPV and NPV
- Page 109 and 110:
What do you need to think about?
- Page 111 and 112:
Avoid Common Mistakes: Hypothesis T
- Page 113 and 114:
Avoid Common Mistakes: Hypothesis T
- Page 115 and 116:
Comparing A to B • Appropriate
- Page 117 and 118:
3 Studies. 3 Answers, Maybe • Stu
- Page 119 and 120:
Misconceptions • A small p-value
- Page 121 and 122:
Questions?
- Page 123 and 124:
Do Not Reject H 0 σ 46 220=X > μ
- Page 125 and 126:
2 Samples: Same Variance + Sample S
- Page 127 and 128:
One Sample Binary Outcomes • Exac
- Page 129 and 130:
Normal/Large Sample Data? No Binomi