Read Elementary Statistics: Picturing the World by Ron Larson Free Online
Book Title: Elementary Statistics: Picturing the World|
Date of issue: January 14th 2005
ISBN 13: 9780131483163
The author of the book: Ron Larson
Format files: PDF
The size of the: 440 KB
Edition: Prentice Hall
Read full description of the books Elementary Statistics: Picturing the World:Even for a textbook I thought this was bad. I haven't ever taken statistics before, and almost everything I know is stuff I've picked up either from studying trading, or from trying to figure out the meaning of scientific studies. I had hoped that a textbook would clear up some of the things I've always found confusing. No such luck.
I would have liked the book to explain the formulas, give some insight into how they were developed, and discuss why they were needed. Instead, the book typically presents a fairly complicated formula, without anything in the way of prior explanation. Sometimes it will present the formula before it even defines the terms that the formula contains, a practice that makes me want to shove needles into my eyes.
Then, the book will go straight into examples of how to use the formula. Most of the time, this means a few pages of examples where the only engagement by the reader is going to a table and looking up a number from rows and columns. Sometimes you go to Table 6 in Appendix B. Other times you go to Table 8. And that's about the depth of the insight you get. I really did not need to have extensive instruction in how to look something up in a stupid table. And frankly, the table look-up is pretty pointless nowadays, because anyone with access to a computer and the internet can probably do better than the tables here.
Then there are the ideas that I always found mystifying when reading statistics, like "degrees of freedom." There are many formulas presented here that contain an entry where you apply the appropriate degrees of freedom. But nowhere is there even a word about what a degree of freedom is, why it might be needed in a formula, or anything else to give a person an inkling of what is really going on with any formula involving a degree of freedom. (In my frustration I did some looking on google to get some idea. It involves n-dimensional vector spaces and constraints along some dimensions within that space. This is a pretty advanced mathematical idea, and I understand why the book doesn't go into any detail. But it wouldn't be too hard to at least mention the idea. It wouldn't be that hard even to explain what happens if you take an object in 3 dimensional space and limit it to motion on a flat plane. What you have done is removed one of the "degrees of freedom" of that object. There! Was that really so hard that it wouldn't even deserve a mention.
Then, even with ideas that are pretty simple, the book offers no explanation for how it gets to the formulas a person is to apply. For example, late in the book we get the Sign Test for a population median. Here's how the book presents it without any other explanation. Someone claims that X is the median for some population. Take a sample and assign + signs when the sample datum is greater than X and a - sign when its less than X. Count the number of + and - signs. If the sample size is 25 or less, then choose the lower count. If greater than 25, then calculate (x+.5) - .5n/(square root n/2). Then do a table lookup depending on the level of significance important to you, but you go to different tables depending on whether the sample size is less than or greater than 25 (without explaining why). Then you compare your sample calculation with the number from the table. The examples then walk you through the procedures of the algorithm. Even on its own, this method depends almost entirely on rote, and doesn't give any understanding about what's going on.
But in the case above, its even worse because it would be very simple to explain what's going on with the sign test. First, remind people of what a median is. It's the number in the population where half the population lies above the number and half lies below it. This means that if you take a random sample of one from the population, there's a 50/50 chance of him falling above or below the median. For any random sample, it's a coin toss. Thus, the derivation of the formula should be really simple. Thus, this problem can be reduced to a simple probability problem involving throwing a coin. If the results deviate too much from 50/50, that suggests that the stated median is probably incorrect. This book doesn't even attempt this, or any other, sort of explanation.
If anyone knows of a good book on probability and statistics, I would welcome the suggestion. This ain't it.
Read information about the authorRoland "Ron" Edwin Larson
Add a comment to Elementary Statistics: Picturing the World
Read EBOOK Elementary Statistics: Picturing the World by Ron Larson Online free
|Download Elementary Statistics: Picturing the World PDF:||elementary-statistics-picturing-the-world.pdf|