Introductory Statistics Introductory Statistics
The Five Processes in Statistics (Daniel) Design the Study (Can) Collect Data (Discern) Describe the Data (More) Make Inference (Truth) Take Action
Questions that you may address Who are safer drivers at BYU-I, male/female, Utahns or non-Utahns? Who is likely to win the presidential race? Is a vaccination effective in treating children? Is this drug rehabilitation program effective?
Population and Sample What would be population of interest? Who are safer drivers at BYU-I, male/female, Utahns or non-Utahns? Who is likely to win the presidential race? Is a vaccination effective in treating children? Is this drug rehabilitation program effective? Feasible to get data from everyone?
Descriptive Statistics Two Types of Descriptive Statistics Graphical Numerical (e.g. Average, Median, Percentage, correlation)
Inferential Statistics - then Take Action Inference vs Deduction Deduction – Most Math Inference - Statistics Take Sample of U.S. to find that 49% of sample would vote for Obama. We would infer this percent (with Margin of Error) to the whole population of the U.S. Large Small Small - Sample Large - Population
Quantitative vs Categorical Quantitative (Numerical) Ex. Height, Body Temperature, Cost of new medical equipment, Number of Siblings, Typical Questionnaire ratings Categorical (Non-numerical) Ex. Gender, Class in College, Race
Types of Data Collecting Census Taken every Ten Year Observational Study – Collects Data but does not attempt to manipulate or influence the outcomes Examples: Household and Governmental Surveys, Exit Polls after elections Designed Experiment – applies a treatment to individuals and attempts to isolate the effect on the outcome Examples: Drug, Vaccination Treatment
Sampling from the Population Simple Random Sample – A Probability Sample Consists of individuals from the population chosen in such a way that every set of n individuals has an equal chance of being selected (names out of a hat) – class example We do Random Sampling to Minimize Bias Other Valid Sampling Methods Stratified Sampling – Divide Population into groups and do a SRS for each group Cluster Sampling – Divide the Population into groups, take a SRS of the groups and sample everyone in the selected groups Systematic Sampling - Select every kth individual and then choose the first person with a random start. Invalid Sampling Methods – Introduce Bias Voluntary Response Sampling Convenience Sampling
Vocabulary for Experiments Subjects – Person or object in study Response Variable – Variable of Interest Treatment Groups – Variable(s) that describe or explain the changes in the response variable Principles of Experimental Design Control – Control any possible lurking variables Randomize – Randomize Subjects to Specific Treatments Replicate – Have as many Subjects for each Treatment Example - Jonas Salk – Vaccination for Polio Double Blind Study Placebo Group