Blog

Book Review – Outliers by Malcolm Gladwell

Outliers by Malcolm Gladwell

Outliers by Malcom Gladwell is a very interesting book!  Gladwell has done a fantastic job at providing short stories and case studies to support each of his main points.

Gladwell points out what we already knew, but did not put together.  It’s like having a jigsaw puzzle box sitting on your coffee table for years, but never diving in to put the puzzle together.  If you did, you would discover that the greatest and wealthiest businessmen were born only a few years apart in the mid nineteenth century and software revolutionists were born only a couple of years apart in the 1950s.  Offspring of Jewish immigrants make the best lawyers and cultures that value respect do not make great pilots.  Gladwell makes these connections for us.

Outliers is about the opportunities that make successful people.  A set of circumstances and a certain upbringing provides the foundation for a person’s success.  A genius could end up living on a farm in rural Missouri with only a high school diploma while poor immigrants become successful businessmen.  Why?  It’s about opportunity and chance.  For example, poor Jamaican girls are born at the right time by chance then receive amazing opportunities that lead to a college degree.­­

I highly recommend Outliers by Malcom Gladwell.  It’s an easy read and inspirational to anyone who may think that success is bred from a certain pedigree.  If you’re not given the opportunities, go make some!

Conditional Formatting in Excel – Highest Number in Each Row

Conditionally formatting each row individually is an issue that I struggled with for some time and finally found an answer.  I have a table that lists 28 different activities by day of the week.  On the report I need to highlight the day with the highest count per activity.

 

 

The solution is to essentially conditionally format each row to highlight the highest number.  But who wants to take time formatting 28 rows?  Plus, there are several other cities to analyze.  So it’s really 28 rows times X number of cities.  Spending a lot of time formatting a report does not make sense to me.  There has to be an easier way to do it.  Well, there is.

 

In a column to the side of the table, insert the following equation and drag down for all 28 rows.

 

 

We use the =MAX() formula to find the highest count in each row.  Note:  hide this column when publishing the report.

 

Next, we want to use the =MAX() column to conditionally format each row.  Go to Conditional Formatting > Manage Rules > New Rule > Format only cells that contain.  Under Format only cells with:  choose Cell Value > equal to.  For this example, the formula is =IF($I3>0, $I3, “”).  Choose your format for the cells and click OK.  Under Applies to click the selection box on the right and highlight your entire table.  For this example, we choose =$B$3:$H$30.  Click Apply then OK and you are done!

 

This is what the above steps should look like (in Excel 2010):

 

 

 

 

Crime Prediction – Predictive Modeling in Law Enforcement

No predictive model is going to be 100% accurate unless by chance.  The nature of predictive modeling is to learn from the past and see into the future.  Essentially, predictive modeling is just modeling.  Think about why we use statistical models – so we can fit the data into a pattern of behavior and anticipate future results.  It’s all about how you use and interpret this model.

 

Crime analysts may use a tool similar to the following example on a robbery series:

 

Data

Date DOW DBO Time Location Money Stolen
3/1/2005 Tue 18:30 1502 E Broadway $1,120
3/8/2005 Tue 7 17:25 1454 S Main St $1,130
3/12/2005 Sat 4 19:00 1903 S First St $1,370
3/17/2005 Thu 5 20:00 1989 N Fourth St $1,420
3/24/2005 Thu 7 20:25 1048 N Third St $1,700
4/2/2005 Sat 9 21:45 1336 E Kimberly Ave $980

 

Average 6 20:09
Minimum 2 17:18
Maximum 9 23:47
St. Dev 1.91 1:36

 

Predicting Next Hit

Min Earliest 7/14/2005 17:18
1 St. Dev Earliest 7/16/2005 18:33
1 St. Dev Latest 7/20/2005 21:45
Max Latest 7/21/2005 23:47
Correlation 0.79381602
Slope 0.00588819
Y-Inctercept -1.231583
# of Days til Next 3.18
Predicted Date 7/15/2005

 

Based on the frequency of the robberies and the amount stolen, the next predicted date is in 3.18 days.  Now, would you tell your patrol officer that they can expect another hit in exactly that time?  No.  But you can use this information to provide a possible date range.  You can also provide the most likely time for a robbery to occur.  Past data helps us to identify patterns for the future but it is not always the rule to follow for future events.  This data is used in conjunction with intelligence reports from patrols.

 

See the Resources page for other templates like this one this template on Robbery Series.

 

Data is concrete.  It gives us the probability of future events occurring.  Add in variables that you know will affect the probability, and you have a predictive model that is even less concrete.  Explaining this to your “customer” should go something like this:

 

This is what happened [insert data here].  This is the probability of it happening again [insert probability here].  Based on the pattern, the next most likely event occurrence is [insert data range]. 

 

In conclusion, it’s all about conclusions.  You will not be able to give a definite “next step”, though you can come pretty close.  A marriage of data analysis and intelligence analysis will give you a higher probability of catching and preventing criminal actions.

What is Lean/Six Sigma?

What is Lean?  Ask this to five people and receive five different responses.  In short, lean is waste elimination and it is an ongoing process.  You will never have officially completed a lean transformation and it may take years to be successful.  This isn’t a simple statistical model where you input some data and receive an answer.  Nor is it just a set of tools you apply to a company.  A lean transformation involves a culture change and it requires buy-in from the lowest level through middle management and all the way up to the big boss.

Alternatively, no project is too small.  In fact, you should probably start small.

You will also see the term Six Sigma associated with lean.  Same goal, more statistics.

Lean/Six Sigma Certifications

If you are interested in becoming certified, there are a number of options available to you.  A certificate course offers you a structured environment and support from your instructors and coaches in learning how to facilitate a lean transformation.  However, pieces of paper do nothing for an organization.  You can be a Green Belt, Black Belt, Master Black Belt or whatever…who cares about a cert if you do nothing with your knowledge?

Where do I obtain a certification?

Many colleges and universities have certificate programs.  There are also several online programs.  My recommendation is to go through a brick and mortar program where you can receive interaction with your instructors and other students.  Also, no matter what type of location, DO A REAL PROJECT.  A simulated project (as you will find in many online programs) does not begin to touch upon the issues you may have while trying to institute a lean transformation.  A main component of lean is the culture – you can’t simulate that!  There are plenty of forums dedicated to arguing over which online university sucks more and why.  It’s up to you.  I’m not against online certifications if you have a job where you can apply your skills as you learn.  If your employer supports your educational aspirations, then go for the online cert.  It’s all about experience.

How long does this take?

It depends what certificate you want to receive.  A Green Belt takes the least amount of time – maybe six months.  A Black Belt or Master Black Belt might take at least a year.  However, this varies depending upon which program in which you are enrolled.

Organizations

ASQ

IASSC

University Certifications Programs

Arizona State University

Brigham Young University – Hawaii

Boston University

Cal State Fullerton

Case Western Reserve University

Cornell University

Emory University

Franklin University

George Washington University

James Madison University

Kent State University

Louisiana Tech University

The Ohio State University

Rutgers University

San Jose State University

University of Colorado – Boulder

University of Houston

University of Michigan

University of South Carolina

University of Texas

Villanova University

 

Here’s a great book resource for Lean:

 

 

If you know of more courses available (and there are a ton!) please comment!  Or, email DD365 and I will update the post.

R Installation Instructions

R logo

 

R is a free software environment for statistical computing and graphics. It compiles and runs on a wide variety of UNIX platforms, Windows and MacOS.

Instructions

  1. First, go to the R-Project for Statistical Computing.
  2. Select one of the links under “Getting Started”.
  3. Scroll down to USA (or your current location) and select one of the links to download the mirrors.
  4. Under “Download and Install R”, select your operating system.
  5. Click on “Install R for the first time”.
  6. Click “Next” and customize as you wish.  Defaults are automatically set.
  7. Click “Finish”.

Or, skip those steps and download R from Michigan Technical University – Download R 3.0.1 for Windows.