CHAPTER 0: Getting Started

1
CHAPTER 0Getting Started
Basic Practice of Statistics 7th Edition

• Lecture PowerPoint Slides

2
In chapter 0 we cover

• Where the data come from matters
• Always look at the data
• Variation is everywhere
• What lies ahead

3
Getting started

• Does Soundscan know whats hot in popular music?
• Is hormone therapy appropriate after menopause?
• What are the extent, cause, and possible
  solutions to global climate change?
• Statistics is the science of learning from data.

4
Data are numbers, but

• Congratulations, your baby weighs 8.5!
• 8.5 what? Kilograms? Thats a BIG baby!
• Data are numbers with a context.
• Likewise, we will use our human judgment to guide
  our use of statistical tools, such as graphs and
  calculations.

5
Where the data come from matters

• Observational studies versus experiments
• Early hormone replacement studies
  observational. Challenge lurking variables,
  such as education and affluence of patients
• Later, definitive studies experiments, decided
  no reduction in heart attack risk due to therapy
• Just one example of complex relationship between
  variables being oversimplified

6
Always look at the data

• Example graph
• A few carefully chosen graphs are often more
  instructive than great piles of numbers.

7
Just ALWAYS do experiments?

• Cant.
• Isnt always necessary well-designed
  observational studies, such as sample surveys,
  can yield high-quality information
• Well-designed ? large number of respondents,
  necessarily
• Example Ann Landers reader survey

8
Variation is everywhere

• Gas prices, June 1990 to August 2013

9
Because variation is everywhere,

• Conclusions are uncertain.
• Example efficiency of HPV vaccine
• was 98 (95 percent confidence interval 86 to
  100).
• Truth? Its in there or not
• Variability is inescapable.

10
What lies ahead

• Purpose give a working knowledge of the ideas
  and tools of practical statistics.
• Three main areas
• Data analysis (Chapters 16)
• Data production (Chapters 8 and 9)
• Statistical inference (Chapters 1318, 2027)

11
Doing statistics is

• More than manipulating numbers!
• You must
• State a problem in its real-world context,
• Plan your specific statistical work in detail,
• Solve the problem by making the necessary graphs
  and calculations, and
• Conclude by explaining what your findings say
  about the real-world setting.