ChE253K Spring09 Lecture06.Rev01

# ChE253K Spring09 Lecture06.Rev01 - Class Business Homework...

This preview shows pages 1–12. Sign up to view the full content.

1 ChE 253K Spr09 Lecture 06 Class Business Homework 02 – Team Project Q’s

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
2 ChE 253K Spr09 Lecture 06 What To Do When The Data Ain’t Pretty Lecture 05 -- Descriptive Statistics
3 ChE 253K Spr09 Lecture 06 Outline Of This Lecture Data distribution summary statistics Normal, skewed and abnormal distributions Mean, median, std dev, inter-Q range Outliers data decontamination Outlier ID tests: actual errors or IQR Recalculate stats, but report outliers Modified box plots handle outliers elegantly

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
4 ChE 253K Spr09 Lecture 06 FFF: Data Selects Statistics
5 ChE 253K Spr09 Lecture 06 Three Centers: Four Spreads Mean Σ x(i)/N Median Middle Odd: Mid-Point Even: Mid-Mean Mode Most (multiple possible) Range Inter-Q Range Std Deviation Mean Abs Deviation N x - x(i) N 1 i = min(x(i)) max(x(i)) - Q(25%) Q(75%) - ( 29 ( 29 1 N x - x(i) 2 N 1 i - =

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
6 ChE 253K Spr09 Lecture 06 Centers & Spreads: Emissions Data 0 5 10 15 20 25 30 <9 11 15 19 23 27 31 Interval Midpoint Frequency 0 5 10 15 20 25 30 <9 11 15 19 21 27 31 Interval Midpoint
7 ChE 253K Spr09 Lecture 06 “Normal” Data Distribution A ~ symmetric “bell” Tall & narrow s small Short & broad s large Normal distributions could indicate: One true mean Measurements w/ many small, random errors Summary Stats: Center: Mean Spread: Std Deviation Sources: Levine, Statistics for Managers

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
8 ChE 253K Spr09 Lecture 06 Skewed Data Distribution A “skewed” bell Skewed distributions could indicate: Bias toward smaller or larger measurements A process with a lower, but not an upper limit Summary Stats: Median, mode or mean IQR, MAD or std dev Sources: Witte, Statistics
9 ChE 253K Spr09 Lecture 06 Other Data Distributions No-bell curve What some shapes could indicate: Bimodal: a process with two results, or two processes Flat: a process with many equally likely results Scattered: a process or measurement out of control Summary Stats: Plot Sources: Web; Witte, Statistics

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
10 ChE 253K Spr09 Lecture 06 Exercise: Distrib’n Summary Stats Data Distribution Summary Stats Symmetric or “Normal” Center: Mean Spread: Std Deviation Skewed or “Leaning” Center: Median or Mode Spread: Inter-Q Range Assorted Others Just plot it Outliers Significant Center: Spread:
ChE 253K Spr09 Lecture 06 Systematic Errors Statistical Tests Inter-Q Range Grubbs Test Report Outliers ! Sources: www.vias.org

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

## ChE253K Spring09 Lecture06.Rev01 - Class Business Homework...

This preview shows document pages 1 - 12. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online