Mnu Biostatistics Lecture 5 PDF

2024/2025 Mathematics and Biostatistics (MS 101) Level 1 Lecture (5) Assoc. Prof. Ahmed R. Elsheakh Director of the PharmD Clinical Pharmacy Program Faculty of Pharmacy - Mansoura National University E-mail: [email protected]...

2024/2025 Mathematics and Biostatistics (MS 101) Level 1 Lecture (5) Assoc. Prof. Ahmed R. Elsheakh Director of the PharmD Clinical Pharmacy Program Faculty of Pharmacy - Mansoura National University E-mail: [email protected] 1  Part I Biostatistics 2 Inferential Statistics 3  Types of Statistics Descriptive Inferential 4  Descriptive vs. Inferential Statistics Items Descriptive Statistics Inferential Statistics Uses Organizing, analyzing, Comparing data, and making and presenting data in a hypotheses, decisions or predictions meaningful manner about a population Using Charts, graphs, tables, Probability etc Characteristics Gives information about  Makes inferences about the raw data population using samples' data  Attempts to reach the conclusion about the population 5  Steps for statistical analysis when conducting research 1. Write your hypotheses and plan your research design 2. Collect data from a sample 3. Summarize your data with descriptive statistics 4. Test hypotheses or make estimates with inferential statistics 5. Interpret your results 6  Contents Introduction to inferential statistics Hypothesis Statistical Tests Probability of chance (P-value) Types of errors Confidence interval (CI) 7  Th Role of Inferential Statistics Make inferences or judgments about Determine the probability that a a population based on the data from a conclusion based on data collected sample drawn from the population. from a sample is true. 8 1. Hypothesis 9  An assumption or hypothesis is made that: "Every person is innocent until proved guilty" A person who has been indicted for committing a crime and is being tried in a court.  Based on the available evidence, the judge will make one of two possible decisions: The person is either guilty or not guilty.  At the outset of the trial, the person is presumed not guilty.  The prosecutor’s effort is to prove that the person has committed the crime and, hence, is guilty. 10  Hypothesis  A statement of what the researcher(s) predict the outcome of the study will be.  Stated at the start of the study. Hypothesis test (statistical test) A procedure used in statistics to evaluate the evidence against a specific hypothesis or claim about a population. The goal is to either reject or accept the hypothesis. 11  Null hypothesis (H0) Alternative hypothesis (H1 or Ha) pronounced H-naught No difference Real difference By chance NOT by chance  Represents the default assumption. Proposes a specific difference or  Suggesting that there is no relationship between variables, significant difference or relationship indicating that the observed results between variables or that any are not due to chance. observed differences are due to chance. 12 H0: The person is not guilty. Ha: The person is guilty. 13 Ex #2: a new drug is tested for reducing blood pressure in hypertensive patients. H0: The new drug has no effect on reducing blood pressure in the patients. H1 or Ha: The new drug has a significant effect on reducing blood pressure in the patients. 14 15 For testing if the difference is real or not, we should study the following: Probability of chance (p-value). Types of errors: Type I and Type II errors. Confidence interval (CI). 16 17  Exercises 1. What is the purpose of a hypothesis in biostatistics? a) To summarize the results of a study b) To provide a definitive conclusion about a research question c) To propose an explanation or statement that can be tested through statistical analysis d) To determine the sample size for a study 18  Exercises 2. Which component of a hypothesis represents the default assumption? a) Null hypothesis b) Alternative hypothesis c) Research hypothesis d) Experimental hypothesis 19  Exercises 3. The alternative hypothesis in a drug experiment states: a) There is no significant difference in the treatment groups b) The drug has no effect on patient outcomes c) The drug has a significant effect on the outcome being measured d) The drug is not effective compared to other treatments 20 2. Probability of chance (p-value) 21  Probability of chance (p-value) The p-value gives the probability of any difference in results having happened by chance. Detect how likely the null hypothesis is true. It ranges from 0-1. 22  Examples P=0.5 means the probability of the difference happened by chance is 0.5 in 1, or 50:50. P=0.05 means the probability of the difference happened by chance is 0.05 in 1, or 1:20 23  P< 0.05 is significant (*) P< 0.01 is highly significant (**) P< 0.001 is very highly significant (***) 24  The smaller the P-value, the greater the evidence against the null hypothesis. The critical value is value of test statistic that separates values that are likely to occur from those that are unlikely to occur. 25 26 Interpret the data according to the following table. 27 Interpret the data according to the following table. 28 Interpret the data according to the following table. 29 30 31  References Rose, M. R. (2024). Biostatistics and Evidence-Based Medicine. In Harriet Lane Handbook (23th ed. Chapter 29, pp. 746-757.e1). Elsevier. Weiss, N. A. (2014). Introductory statistics (7th ed.). Pearson. Rosner, B. (2015). Fundamentals of biostatistics (8th ed.). Brooks/Cole, Cengage Learning. Kirkwood, B. R., & Sterne, J. A. C. (2003). Essential medical statistics (8th ed.). Wiley-Blackwell. 32 Helpful Textbook on Biostatistics for further reading 33 34