Transcript Introduction to ANALYSIS

Introduction to ANALYSIS
Why a person should not fear
Statistics
Even babies can love statistics…..
And even professors love
babies…
Statistics is a way to get INFORMATION
from data.
It may look complicated….
London Heathrow Airport
http://finance.yahoo.com/q?s=^DJI
Some other statistics….
http://www.measuringworth.com/uscompare/
http://www.x-rates.com/d/CAD/CNY/graph120.html
Let’s take a look at what statistics are and
what they do.
http://en.wikipedia.org/wiki/Statistics
History
Karl Pearson
1857-1936
Charles Darwin
1809-1882
Francis Galton
1822-1911
Charles Spearman
1863-1945
Statistics:
Getting information from data.
Who (what) are we studying?
Population
1. Census
2. Sample
Statistics:
Getting information from data.
Who (what) are we studying?
Population
1. Census
2. Sample
a. Describe
b. Infer
Statistics:
Getting information from data.
Who (what) are we studying?
Population
Sample
Two types of analysis
1. Descriptive
2. Inferential
Probability
Statistics:
Getting information from data.
Inferential
The Probability that what we find in the
Sample will apply to the whole Population.
Statistics:
Getting information from data.
Inferential
Whenever a sample is utilized, there will
always be a “sampling error.”
Statistics:
Getting information from data.
Inferential
Whenever a sample is utilized, there will
always be a “sampling error.”
That “sampling error” will increase as the
variability of the measure increases.
That “sampling error” will decrease as the
size of the sample increases.
Statistics:
Getting information from data.
Inferential
The “sampling error.”
SE   x / n
Most inferential statistics are simply the number of
sampling errors a finding is from what was expected.
Finding  Expectation
Test 
SamplingError
THIS
is the universal statistic….
Finding  Expectation
Test 
SamplingError
The universal statistic….
For example, the t-test is:
t
x  0
0
n
Or:
Finding  Expectation
Test 
SamplingError
Ferguson’s Epitaph for a Statistician
With no applause from saint or devil
He was significant at the .05 level.
He squeezed life’s data, but when done
He failed to reach the .01.
Fergusons, G. A. (1981). Statistical Analysis in Psychology and Education
(5th Ed.). McGraw Hill: New York.