Textbook Notes (270,000)
CA (160,000)
UW (5,000)
STAT (30)
STAT211 (20)
Chapter 1-2

STAT211 Chapter Notes - Chapter 1-2: Confounding, Stratified Sampling, Wart


Department
Statistics
Course Code
STAT211
Professor
Joslin Goh
Chapter
1-2

This preview shows half of the first page. to view the full 3 pages of the document.
Jan 5, 7 Lecture 1,2 1
C1/2: Introduction and Gathering Data
Study of statistics can be broadly classified into:
1. Descriptive statistics
Describe characteristics of data collected in a clear and understandable way
oNumerical
oGraphical (histogram, boxplot)
2. Inferential statistics (inductive statistics)
Proceed from data characteristics to make generalizations, estimation, forecasts, or
other judgements based on the data
oEstimation
oConfidence intervals
oHypothesis testing
Statistics language
Variable any characteristic of an individual (does not have to be
human)
Response variable (dependent
variable)
an outcome of a study, variables of interest, what you get for
your request
Explanatory variable
(independent variable)
A measure in the study used to explain or influence the
response variable; your request
How do you collect data?
Observation Does not attempt to influence the subject
compare existing groups/situations
Observe and measure variables, but does not impose any conditions
Experiment Imposes treatment or condition on subject, then study the changes
caused by condition or treatment
Observation vs Experiment
A well designed experiment can give strong evidence of a cause-and-effect relationship
(explanatory variable causes changes in the response variable)
Not all research can be done in an experimental setting (time, money, ethical, other
considerations)
Prefer to do experiment (Observation can be as time consuming)
Example 1: Is this an experiment or observation? Over a four-month period, among 30 people with
bipolar disorder, patients who were given a high dose of omega-3 fats from fish oil improved more
than those given a placebo (fake omega 3)
Answer: experiment; patients have to take a pill (either fish oil or placebo)
Key Distinction: although it looks like an observation, you assign a condition (make certain groups do
something and other groups do something else)
Example 2: Is this an experiment or observation? Coffee stations in offices often ask customers to
leave money in a tray to pay for their coffee, but many do not to that. Researchers replaced the
picture of flowers on the wall behind the coffee station with a picture of staring eyes. They found that
the average contribution increase significantly above the well-established standard.
You're Reading a Preview

Unlock to view full version