Unformatted text preview: 11 Uses of Statistics Two primary uses for statistics: Descriptive statistics the collection, organization, presentation and summary of data. Inferential statistics generalizing from a sample to a population, estimating unknown parameters, drawing conclusions, making decisions. 12 Uses of Statistics Overview of Statistics Statistics
Visual Displays Numerical Summaries
13 Making Inferences from Samples Estimating Parameters Testing Hypotheses Statistical Challenges Working with Imperfect Data State any assumptions and limitations and use generally accepted statistical tests to detect unusual data points or to deal with missing data. Dealing with Practical Constraints You will face constraints on the type and quantity of data you can collect. 14 Statistical Challenges Upholding Ethical Standards Know and follow accepted procedures, maintain data integrity, carry out accurate calculations, report procedures, protect confidentiality, cite sources and financial support. Using Consultants Hire consultants at the beginning of the project, when your team lacks certain skills or when an unbiased or informed view is needed.
15 Statistical Pitfalls Pitfall 1: Making Conclusions about a Large Population from a Small Sample Be careful about making generalizations from small samples (e.g., a group of 10 patients). Pitfall 2: Making Conclusions from Nonrandom Samples Be careful about making generalizations from retrospective studies of special groups (e.g., heart attack patients).
16 Statistical Pitfalls Pitfall 3: Attaching Importance to Rare Observations from Large Samples Be careful about drawing strong inferences from events that are not surprising when looking at the entire population (e.g., winning the lottery). Pitfall 4: Using Poor Survey Methods Be careful about using poor sampling methods or vaguely worded questions (e.g., anonymous survey or quiz).
17 Statistical Pitfalls Pitfall 5: Assuming a Causal Link Based on Observations Be careful about drawing conclusions when no cause-and-effect link exists (e.g., most shark attacks occur between 12p.m. and 2p.m.). Pitfall 6: Making Generalizations about Individuals from Observations about Groups Avoid reading too much into statistical generalizations (e.g., men are taller than women).
18 Statistical Pitfalls Pitfall 7: Unconscious Bias Be careful about unconsciously or subtly allowing bias to color handling of data (e.g., heart disease in men vs. women). Pitfall 8: Attaching Practical Importance to Every Statistically Significant Study Result Statistically significant effects may lack practical importance (e.g., Austrian military recruits born in the spring average 0.6 cm taller than those born in the fall).
19 Statistics: An Evolving Field Statistics is a relatively young field, having been developed mostly during the 20th century. Its mathematical frontiers continue to expand with the aid of computers. Major recent developments include - Exploratory data analysis (EDA) - Computer-intensive statistics - Design of experiments - Robust product design - Advanced Bayesian methods - and more
110 Applied Statistics in Business and Economics
End of Chapter 1 111 ...
View Full Document
- Fall '07
- Statistics, Statistical hypothesis testing, Statistical significance, data integrity, Statistical Pitfalls