STA 100 Lecture Notes Week 1 Professor Melcon
STA 100 Lecture Notes Week 1 Professor Melcon STAT 100
Popular in Applied Stat for Bio Sci
Popular in Math
This 4 page Class Notes was uploaded by Brittany Church on Monday September 26, 2016. The Class Notes belongs to STAT 100 at University of California - Davis taught by E. Melcon in Fall 2016. Since its upload, it has received 124 views. For similar materials see Applied Stat for Bio Sci in Math at University of California - Davis.
Reviews for STA 100 Lecture Notes Week 1 Professor Melcon
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 09/26/16
STAT 100 Dr. Melcon Fall 2016 This class uses Canvas. <Canvas.UCDavis.edu> Homework will be through canvas, due on Wednesdays. We also use Piazza as a forum/discussion site. We will also do programming through a program named R. Dr. Melcon’s Office Hours are Monday and Tuesday. Grade breakdown: Homework 15% Exams (2) 20% each exam Project 10% Final exam 35% Lecture 1 Statistic Science of analyzing data. There are 3 branches: 1) Summary/visualization - Trends 2) Inferential - Data for predictions of unknown quantities. 3) Probability Theory - study of probability / theory. Categorical Data: data that is naturally labels or categories. Example Disease status (yes or no) Gender Hair Color Military rank Types of categorical variables: 1) Nominal Categories - naturally do not have ranking or order 2) Ordinal Categories - naturally having a natural or not forced ranking Nominal Ordinal Hair Color Grades Eye Color Military Rank Numerical Data: data that is numbers. Types: 1) Discreet - natural gaps between the numbers or data. Example Number of __________ (arms, eyes, outbreaks) ***This can only be whole number values. 2) Continuous - where data can be ANY value between two numbers (or in an interval) Example Height (smallest infant compared to the tallest adult) Subject: General term. Can be a person, place, or thing that data is being measured from. * Population - all subjects of interest, usually VERY large. Not used often (constantly changing, too big, too many variables, impractical [money and time]. ** However, populations can be defined as a smaller group. Sample: A subject of the population or a subset of subjects. What makes a good sample? *Conclusions are only as good as your data. ● Similarity ● Representative: Similar characteristics of populations ● Relevant ● Random: random sample (a simple random sample) each subject has the same chance of being selected. **This reduces bias. To draw inference on the population, we need these for a good sample. Variable: something that changes from subject to subject. Random Variables: variable with a random component. Example Hair color, Class grade **Notation - Y : This represents all possible values of the thing of interest. ** y :is the i-th observed value of Y. Example Y = all possible heights A yi = 67 inches (Dr. Melcon’s height) How do we get data? I) Experiments: where you keep or want all subjects to have the same characteristics and manipulate groups. (**This is very rare and difficult to do) Example 20 plants at the same temperature, same water amount is given. However, 10 plants get a fertilizer, 10 plants get nothing (control). Pros Can sometimes conclude causation Cons: Expensive; Can’t use on humans because they differ by default. II) Observational Studies: Subjects are put into groups, possibly manipulated, but there are confounding variables. ** onfounding variable: anything that affects the outcome of our subject that we can’t control. Example Surveys: Subjects respond to questions and self record. Pros fast, cheap, wider reach, easier Cons: Bias, lies, not representative Lecture 2 Need to download the program R. Find at https://cloud.r-project.org/ Download the program for your computer’s requirements. Reducing Bias: Can make the following groups in addition to the treatment group: 1) Control Group: Observations are recorded, subjects are not manipulated 2) Placebo Group: Subjects are give non-reactive treatment 3) Nocebo Effect: Subjects have an adverse reaction a nonreactive treatment 4) Sham Group: An invasive procedure is done, but treatment is not Example Getting an IV, but no drug Blinding Studies: 1) Single Blinding: Either the subject or researcher does not know their group. 2) Double Blinding: both the subject and researcher do not know the groups. In statistics, we want to estimate unknown parameters. *A problem with drawing conclusions from a single sample is that there is sampling variation. Sampling Variation: Because every sample is different and random, the results are as well.
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'