Chemistry and Statistics

1
Chemistry and Statistics

• Dr. Jerry Lin
• February 1 6, 2007

2
Organization

• How to fail an FE Exam the top 10 list
• The fundamentals the building blocks
• Questions and answers what do you want to know?
• The practice no pain, no gain
• The reality check to determine your grade

3
10 best ways to fail an FE Exam .

• Expect the opportunity to take the exam for a
  second time
• Wait until the last minute to prepare intensively
• Get too ambitious to get all subjects well
  prepared
• Rely on multiple review manuals
• Give up on one or more subjects
• Consider this review class as a ticket to pass
• Guess by instinct for the unsolved problems
• Fall in love with one of the exam subjects
• Party in the night before the exam
• Think that I might just get lucky to pass

4

• Whats useful in the reference handbook?

5
(No Transcript)
6
(No Transcript)
7
(No Transcript)
8
(No Transcript)
9
Other useful stuff

• Concentration representations
• Oxidation state (and charge) calculation
• Balance a chemical equation
• Equilibrium (including acid-base) calculations
• Stoichiometric calculation
• Ideal Gas Law

10
Descriptive Statistics

• Sample size the number of data points
• Sorting place data in ascending/descending order
• Ranking give index to sorted data (minimum
  Maximum)
• 1st quartile the 25th data (observations), Q1
• 3rd quartile the 75th data (observations), Q3
• Mean the arithmetic average of data
• Median 2nd quartile, the 50th data, Q2
• Mode the most frequent observations
• Variance (Standard deviation)2, the spreadness
  of data

11
(No Transcript)
12
Probability Density Function
A mathematical representation (a function of an
independent random variable) showing the
likelihood (i.e., relative frequency) of data to
occur. The integration of a probability density
function give PROBABILITY.
13
Cumulative Distribution Function
14
(No Transcript)
15
(No Transcript)
16
(No Transcript)
17
(No Transcript)
18
Hypothesis Test

• Develop a hypothesis pair
• Define a confident level (1-?), i.e., select a
  significant level
• Collect necessary data (sampling)
• Select a statistics with known distribution for
  the hypothesis
• With the defined confidence level and calculated
  statistics, accept or reject the hypothesis.

Confidence level is usually selected to be 90
or 95 to ensure the quality of the hypothesis
test.
19
(No Transcript)
20
Example Hypothesis test

• A tire company advertises that one of their new
  tire products can last at least 100,000 miles. It
  is known that the manufacturing process has a
  standard deviation of tire life of 2500 miles. To
  test if the ad is valid, you take 15 tires and
  found that the average tire life is 100,250
  miles. What would you conclude?

21
(No Transcript)
22
Simple Linear Regression
Errors
23
(No Transcript)