Type I and II

From time to time, it's worth revisiting our old friends Type I and Type II:

In general, a Type 1 error makes you depart from the status quo and take some action that is wrong, while committing a Type 2 error means that you stick with the status quo when you should have acted.

Maintain control
Decision theory is always a trade-off.

In hypothesis testing, you make sure that the Type 1 error is the severe error, so you can control the probability of making it.

As a result, the Type 2 error may in fact have a larger probability of occurring (but it is the lesser of the 2 errors).

Examples:
A pharmaceutical company is testing the safety of a new drug:
Type 1 error: concluding it is safe when it’s not (kill people)
Type 2 error: concluding it is not safe when it is (lose some money)

The criminal justice system:
Type 1 error: convicting an innocent person (terminology: the defendant is guilty)
Type 2 error: not convicting a guilty person (terminology: the defendant is not guilty. It’s important to note that you never say the defendant is innocent).

A scientific researcher
Type 1 error: publishing your research when it’s wrong (embarrassment)
Type 2 error: not publishing your research when it’s correct (usually this means continuing your research until you can show progress)

Important terminology:

• There is sufficient evidence to reject the null hypothesis in favor of the alternative.
• The is insufficient evidence to reject the null hypothesis (NEVER ACCEPT THE NULL HYPOTHESIS – no one is ever declared innocent; they are declared not guilty)

I didn't make this stuff up. I credit a good friend, Dr. Pat Bond, a professor at Florida Tech, for these insights

 Bookmark this on Delicious  

Are you on LinkedIn?    Share this article with your network by clicking on the link.

Gilb on Agile

Tom Gilb is a system engineer; he is also a project manager and someone with deep understanding of software engineering. His most prominent work is his 1988 book: "Principles of Software Engineering Management".

He is the inventor and thought leader behind an agile-like methodology named EVO (for evolutionary).

In my own book, "Project Management the Agile Way: making it work in the enterprise", I include a lot of material taken from Gilb's work. I like what he says, though frankly a better spokesman for what Gilb advocates is Niels Malotaux: he writes with clarity and simplicity, something that eludes Gilb himself.

Recently, I fell onto a string of Gilb'isms, starting with a 2004 keynote address to an agile conference entitled "What is missing from the conventional agile and eXtreme methods?" His answer, in one word: quantification. And by this he really means quantification of business results, outcomes generated by project outputs, in business terms. He more or less sums up the idea with the words value and quality.

He really takes agilists to task for promoting the idea of a single customer/user as the arbiter of product requirements; instead, agile teams should be consulting--weekly--with the business stakeholders, a diverse group Gilb says can be more than a dozen in any real enterprise.

And, he has heartburn with user stories and other low level practices, like TDD, that he says unnecessarily constrain the inventiveness of developers and leads to "amateur" engineering. Gilb's formula: write down the business quality values on not more than one sheet of paper, and let the developers go from there.

In 2010, he wrote a couple of magazine articles in "Agile Record". The first part, entitled "Value driven principles and values" is a better explanation of what was in the 2004 keynote.  The second part, entitled "Values for Values" builds out his ten principles--first presented in the keynote address--and given also in the first part article.

Gilb's principles are ones we can live by. Here's a reproduction of one version:

As Gilb says of the Agile Manifesto: "it's heart is the right place". Even if you don't buy all that brother Tom says, his heart is in the right place: better quality and value for stakeholders through projects and software.

 Bookmark this on Delicious  

Are you on LinkedIn?    Share this article with your network by clicking on the link.

Building your personal brand

I recently had an the opportunity to address a group of project managers about how to build a personal brand.

Ah yes! The personal brand. And what is that exactly? I told the group: these days, it's whatever the web search says it is. Your identity is more or less your public identity on the internet: it's whatever HR or the hiring manager, or anyone else can find about you as they figure out who you are.

So my advice was: take the initiative. Build your brand on the web yourself and get out in front of the web search. And, there's a lot of ways to do it. Of course, own your ownname.com; and create profiles on the professional network sites that will show up in web searches.

Another idea: have a friend interview you and create your own audio blog. It's cheap and its easy.

And, more, so here's the presentation as available on slideshare.net/jgoodpas:

Building Your Personal Brand

View more presentations from John Goodpasture

 Bookmark this on Delicious  

Are you on LinkedIn?    Share this article with your network by clicking on the link.

Coming in at 95

The good news: we're still on the Top 100 Agile books list
The bad news: we've slipped a bit from last year. Oh well.

Still, this is the book we're talking about:

 Bookmark this on Delicious  

Are you on LinkedIn?    Share this article with your network by clicking on the link.

Bow-ties for risk managers

Have you seen this one before? I had not until I was chasing down ISO 31000:2009, the ISO standard on risk management. In a slide show about that, I found this:

The left side is 'likelihood controls' and the right side is 'consequence controls'. Kinda neat, I thought. Certainly, a unique display of a risk matrix.

 Bookmark this on Delicious  

Are you on LinkedIn?    Share this article with your network by clicking on the link.

A touch of madness

Madness! Madness!!

Last two words spoken

"Bridge over the River Kwai"

Sane was the watchword

In several posts, I've made the point that the passionless rationalist will often find it hard to reach a decision. They are wrapped up in the paralysis of analysis--they keep going over all the data and all the alternatives and never reach a consensus--with others or with themselves--on what to do.

My contention, said by many others as well, is that it takes a dose of passion for a vision to break the analysis cycle and reach a decision. And, note this: the passionate have stickiness--once decided, it's hard-to-impossible to move someone off their position.

How many times, on a big decision, have you had an instinctive feeling: this is the right thing to do!  Once felt, there's no more paralysis and there's no more dithering.

That stickiness we spoke of is an elixir to  the follower community: Yes! the leadership knows where we need to get to; there will be certainty and willingness to put it on the line to get there.

Now, I've also written about innovation, and the role leadership plays inspiring the creative to innovate something new to the world.

Along comes a touch of madness

But what luck! Now we learn that there's something that ties leadership and innovation together--something unique when a leader is both inspirational and innovative. In a book that ties it all together, it turns out that truly inspired and innovative leadership have a common root: it is the byproduct of mental disorder and mild manic depression. Who knew? In a stroke (no pun), we've now got the whole picture!

In a new book, author Nassir Ghaemi explains in "A First-Rate Madness: Uncovering the Links Between Leadership and Mental Illness" that some of our best and brightest are just a little mad. Is this surprising?
Perhaps not.  But Steven Colbert was particularly insightful when he asked whether we should be worried that a madman may have his finger on the button (even if only the project button).  Ghaemi said: "That's one way of putting it".  But in his telling, benefits outweigh risks.

God, I hope he's right!

 Bookmark this on Delicious  

Are you on LinkedIn?    Share this article with your network by clicking on the link.

More about our friend Bayes

When the facts change, I change my opinion.
What do you do, sir?

John Maynard Keynes

In a review of a new book "The Theory That Would Not Die" (S. McGrayne, May,2011) mathematician John Allen Paulos says this:

"At its core, Bayes’s theorem depends upon an ingenious turnabout: If you want to assess the strength of your hypothesis given the evidence, you must also assess the strength of the evidence given your hypothesis.


In the face of uncertainty, a Bayesian asks three questions:

1. How confident am I in the truth of my initial belief?
2. On the assumption that my original belief is true, how confident am I that the new evidence is accurate?
3. And whether or not my original belief is true, how confident am I that the new evidence is accurate?"

In other words, the Bayesian idea is this:

1. Form a hypothesis; assess the probability of its being true
2. Seek evidence that the hypothesis is true
3. Recalculate the probability of the hypothesis, given the evidence (assuming the hypothesis is TRUE) and the probability of finding the evidence

There are some obvious "calibration" problems, the same sort that show up in the risk matrix, and these calibration issues have challenged the efficacy of Bayesian reasoning in the same way:

• It's difficult to get a calibrated probability for the initial hypothesis--what's the benchmark here, and can you be sure that biases are removed?
• It's difficult to find the supporting evidence with clarity of cause and effect
• It's difficult to assess the probability of finding the evidence--again, there may be no historical basis for such a probability.

Nevertheless, according to McCrayne's history, Bayes has these credits:

"It was used to search for nuclear weapons, devise actuarial tables, demonstrate that a document seemingly incriminating Colonel Dreyfus was most likely a forgery, improve low-resolution computer images, judge the authorship of the disputed Federalist papers and determine the false positive rate of mammograms. She also tells the story of Alan Turing and others whose pivotal crypto-analytic work unscrambling German codes may have helped shorten World War II."

For more, see my prior posts.

 Bookmark this on Delicious  

Are you on LinkedIn?    Share this article with your network by clicking on the link.