16
Pg. 268 - 285

Pg. 268 - 285. Exploratory data analysis describes what data say by using graphs and numerical summaries. What if we want to ask a large group of individuals

Embed Size (px)

Citation preview

Page 1: Pg. 268 - 285. Exploratory data analysis describes what data say by using graphs and numerical summaries. What if we want to ask a large group of individuals

Pg. 268 - 285

Page 2: Pg. 268 - 285. Exploratory data analysis describes what data say by using graphs and numerical summaries. What if we want to ask a large group of individuals

Exploratory data analysis describes what data say by using graphs and numerical summaries.

What if we want to ask a large group of individuals questions?

Page 3: Pg. 268 - 285. Exploratory data analysis describes what data say by using graphs and numerical summaries. What if we want to ask a large group of individuals

To get valid answers, we need to produce our data carefully. Often we use samples to represent a

larger population. Section 5.1 deals with the many ways to choose samples.

Page 4: Pg. 268 - 285. Exploratory data analysis describes what data say by using graphs and numerical summaries. What if we want to ask a large group of individuals

To get valid answers, we need to produce our data carefully. Once you have chosen a sample, you

have a few ways to gather data.

Page 5: Pg. 268 - 285. Exploratory data analysis describes what data say by using graphs and numerical summaries. What if we want to ask a large group of individuals

An observational study observes individuals and variables of interest, but does not try to influence them.

An experiment deliberately imposes some treatment on individuals in order to observe their responses.

An observational study observes individuals and variables of interest, but does not try to influence them.

An experiment deliberately imposes some treatment on individuals in order to observe their responses.

Page 6: Pg. 268 - 285. Exploratory data analysis describes what data say by using graphs and numerical summaries. What if we want to ask a large group of individuals

Read Example 5.1 on page 270 of your text.

Page 7: Pg. 268 - 285. Exploratory data analysis describes what data say by using graphs and numerical summaries. What if we want to ask a large group of individuals

Explanatory variable – attempts to explain the observed outcome (p. 121)

Lurking variable – a variable not among the explanatory variables, but still may influence the interpretation of relationships among those variables. (p. 226)

Response variable - Measures an outcome of a study. (p. 121)

Page 8: Pg. 268 - 285. Exploratory data analysis describes what data say by using graphs and numerical summaries. What if we want to ask a large group of individuals

Section 5.2 is about designing valid experiments that can be used to determine causation.

Section 5.3 is about using simulation to produce data.

Page 9: Pg. 268 - 285. Exploratory data analysis describes what data say by using graphs and numerical summaries. What if we want to ask a large group of individuals

Population: The entire group of individuals that we want information about.

Sample: The part of the population we actually look at to gather information.

Population: The entire group of individuals that we want information about.

Sample: The part of the population we actually look at to gather information.

Page 10: Pg. 268 - 285. Exploratory data analysis describes what data say by using graphs and numerical summaries. What if we want to ask a large group of individuals

Census: Attempts to contact every individual in the entire population.

Census: Attempts to contact every individual in the entire population.

Page 11: Pg. 268 - 285. Exploratory data analysis describes what data say by using graphs and numerical summaries. What if we want to ask a large group of individuals

Voluntary Response Samples▪ Ex: call-in polls, text your vote, etc.

People choose themselves by responding.

People with strong opinions (especially negative opinions) are more likely to respond. ▪ This leads to bias in your sample.

Page 12: Pg. 268 - 285. Exploratory data analysis describes what data say by using graphs and numerical summaries. What if we want to ask a large group of individuals

Convenience Sampling▪ Ex: Sitting outside a mall or grocery store

Choosing the individuals that are easiest to reach▪ Another source of bias: convenience

sampling is almost guaranteed not to represent the entire population.

Page 13: Pg. 268 - 285. Exploratory data analysis describes what data say by using graphs and numerical summaries. What if we want to ask a large group of individuals

Bias: The design of a study is biased if it systematically favors certain outcomes.

Bias: The design of a study is biased if it systematically favors certain outcomes.

Page 14: Pg. 268 - 285. Exploratory data analysis describes what data say by using graphs and numerical summaries. What if we want to ask a large group of individuals

Bias: The design of a study is biased if it systematically favors certain outcomes.

Bias: The design of a study is biased if it systematically favors certain outcomes.

THE SOLUTION!

Let chance choose the

sample.

This is the essential principle of statistical

sampling.

Page 15: Pg. 268 - 285. Exploratory data analysis describes what data say by using graphs and numerical summaries. What if we want to ask a large group of individuals

Simplest way: Put the whole population in a hat and draw out a handful of individuals for your sample.

Page 16: Pg. 268 - 285. Exploratory data analysis describes what data say by using graphs and numerical summaries. What if we want to ask a large group of individuals

Practice problems: #1-4 (p. 273)