Cause and Effect and the Scientific Method

Experiments and studies are often conducted to identfy causes for observed effects.  'Why does this symptom appear?' or 'Why are oranges orange?'  The problem is, it's easy to innocently create causes that are not really there.  Sometimes the identified 'cause' may be something that is present, but doesn't have anything to do with the effect you are studying. "Oranges are orange because they have lots of vitamin C.'  Well, lots of other things have lots of vitamin C but are not orange.  Look at Kiwi.
 
There are so many potential causes for so many occurrances, that assignment of cause is often difficult.  You need to look for differences.  'How are the ingredients of an orange differentent than those of a Kiwi?' or similarities 'How are the ingredients of an orange like other orange things?'

Then there's post hoc ergo propter hoc': After this, therefore because of this.  In the middle ages, scientists thought that fruit flies spontaneously generated from garbage.  How did they know this? They put fruit in a jar and closed the lid. Lo and behold, fruit flies appeared!  Proof?  The actual cause was discovered in later years.  The fruit flies had laid eggs in the fruit prior to it being placed in the jar.  Post hoc ergo proter hoc.

The Kingdom was Lost

Epstein mentioned that one must accept 'The Normal Condition" as background for any discussion of cause and effect.  Otherwise, it's difficult to tell how far back to go. This is something I haven't considered before. Most often, I'm after root cause. You see the effect, an injury.  Why did it happen? Trace that back far enough and you can do a lot of prevention.


Why was the Kingdom lost?


For want of a nail the shoe was lost.
For want of a shoe the horse was lost.
For want of a horse the rider was lost.
For want of a rider the battle was lost.
For want of a battle the kingdom was lost.
And all for the want of a horseshoe nail.(Gower, 1390)


So the next question is Why was there Want of a Nail? Where to stop is a judgement call.  How far back is reasonable? Epstein says "For a causal claim, the normal conditions are the obvious and plausible unstated claims that are needed to establish that the relationship between purported cause and purported effect is valid and strong."  Is this string of circumstance strong and valid enough to say the country was lost because of lack of a nail?  Probably not.  Starting with the rider, he might well have been able to find another horse or another means of transport.  One the other hand, how much effect can one sholdier have?  Maybe we can't go any further back than the battle....  

"The earliest known written version of the rhyme is in John Gower's " Confesio Amantis dated approximately 1390"  http://www.rhymes.org.uk/for_want_of_a_nail.htm

Mission Critical...

I found the entire Mission Critical website very interesting.  I hope it is available to us after this class is over.  It provided a different way to look at the concepts we've covered.  I particularly liked the quizes.  They did a pretty good job of explaining why wrong answers were wrong - and they explained in less detail why the right one was right.

It was a very inclusive webiste. I found the table of contents almost overwhelming...so many choices, so little time.

The Causal argument exercise was very good. These types of things make it much easier for me to learn than reading a chapter in a book. I have some vision problems that make reading difficult, and reading from a screen even harder. However, the short amount of large text used in the exercises was excellent.

Clear definitions, separate from text are great.

"A word or phrase is said to be ambiguous if it has at least two specific meanings that make sense in context.
A word or phrase is said to be vague if its meaning is not clear in context. "

Well, Epstein is fairly good at that too.

The Group Paper Experience - Synergy - or the Lack of It

What did I learn?  I learned a lot about myself.  I found that the less response I felt from the group members, the more I tended to take over and try to do it myself. Not good. Each member of the groups I was in did participate and gave good input. However, my current experience is with a stable team, and that's different.

The experience took me back to the first group projects I worked on, and to the sense of ownership I had over my one part.  I didn't want it changed.  It was MINE! To give up control and allow changes to happen is very difficult.  But I now feel that group projects work best when there is give and take among everyone involved.  So, I can change your piece, you change mine, and we end up with a product that is better than either one of us could do alone.  Unfortunately, that's difficult to achieve unless the work group has time to develop cohesion and and group vision.

Groups have development cycles, starting with forming - getting to know one another.  From there, a storming period, where personalities and ideas clash usually occurs.  Next, progression can occur to what is called norming.  Here, the group interacts well and real working together can begin.  Finally, the group hopefully reaches the performing level.  This is where inventiveness and experimentation occur.  The team trusts each other well enough to allow true collaboration. 

So why was I surprised that these papers were so hard?

Median, Mode, Average

These are interesting concepts that I remember from my beginning statistics class. 
Median:  If you have a bunch of numers in set   (1,3,4, 3,2,5,4,2,3,4,5,3,6,3) and put them in order, you have one 1, two 2s, five 3s, three 4s, two 5s,, one 6.  How many numbers do you have? 13.  So, if you divide 13 by 2, you get 6.5.  The seventh number from the top or bottom is smack in the middle. That is the median.  The number that occurs most often (in this case, 3) is the mode.

So the average is left.  That's when you add all the numbers up and divide by the number of numbers. So let's see, added together, 1 get 48 divided by 13 = 3.69 or round up to 4.

median = 7
average = 4
mode = 3

Depending on what your're talking about, using the right one can make a lot of difference.

The Devil is in the Details

In Epstein's Chapter 13, NUMBERS, the Devil may be in concealed details, or details that aren't readily apparent. Having worked a bit with statistics, I can say that it is easy to accidently mis-lead someone by missing details. 

Over the past year, I was trying to understand why there were more calls answered than cases about the calls concerning people who move within San Jose.  There should have been about twice as many cases.  I thought there was something wrong with the query I was using to get my figures or that the Customer Service group was not entering notes as they should.

In March this year, I was finally able to talk the technical team into changing the names for the notes to make them more descriptive.  To my surprize, this seems to have fixed the problem.  Now, we have about twice as many notes having to do with moving than we have calls.  It turns out that notes were being entered, but they were in the wrong categories - so they didn't come up when I searched for moving statistics. Duh!

With statistics, you have to be sure you're measuring what you think you are.

Apple Polishing and Advertising

Apple Polishing is essentially a term meaning to give false praise in order to gain favor.  Wise Geek.com lists synonyms such a brown-nosing toadying, etc.  People who use this technique extensively are called sycophants. Their attitude is often obsequious (fawning or servile).

I remember "bring an apple for the teacher" as a phase I heard growing up.  I kept expecting to see the teacher's desk covered in apples.  Indeed, this is where the phase 'Apple Polishing' comes from (Wise Geek).

In advertising, the Apple Polishing is sometimes fairly open:  "Victoria's Secret is THE destination for chic women's clothing, sultry swimwear, must-have shoes ..."  In other words, shop here and you will be chic or sultry.  Apple Polishing is even used in product names: "Weight Watchers® Smart Ones® - Low Fat, Low Calorie and High Fiber ...Weight Watchers® Smart Ones® Low Fat, Low Calorie and High Fiber Meals has taken eating smart to a whole new level of deliciousness" .saying Eat this food and you will be a Smart One.

The use of celebrities in advertising is more of a analogy.  If you want to be like this person, use my product..but it is also a more indirect form of apple polishing, because it intimates you CAN be like the celebrity. 

Finding one advertisement to speak to seemed impossible.  Each one only seems to use apple polishing in one or two statement:  And written or on-line adverts are not primarily words. But Apple Polishing seems to be everywhere.



http://www.wisegeek.com/what-does-apple-polishing-mean.htm
http://dictionary.com/
http://eatyourbest.com