Psycho-Babble Books Thread 774069

Shown: posts 1 to 6 of 6. This is the beginning of the thread.

 

Looking for a good book on statistics

Posted by Quintal on August 5, 2007, at 0:48:58

The problem I've always had with maths is that it's relatively meaningless so I tend to forget everything I've learned in a short space of time. My physics teacher said she thought my mathematical ability was fine, but I'm not so sure. I've developed quite a passion for statistical analysis with all this PubMed research I've been doing and I'd like to learn how to do things like meta-analysis properly, and also what all those funny statistical terms really mean. At present I can only divine their meaning from context so it would be helpful to have an easy-to-read manual or reference guide at hand.

Does anyone know of a such a document? I don't suppose there are many aimed at beginners? Maybe there's one on statistics in that 'For Dummys' series or something?

Q

 

Re: Looking for a good book on statistics » Quintal

Posted by Racer on August 5, 2007, at 1:06:37

In reply to Looking for a good book on statistics, posted by Quintal on August 5, 2007, at 0:48:58

I'm not familiar with any books about statistics other than my textbooks, but a quick Amazon search led to "How to Lie with Statistics," which would probably be helpful -- it should show some of the pitfalls to avoid in reading studies. And there is a "Statistics for Dummies" book, too.

As far as recommendations, though, I very strongly recommend Ian Stewart -- his writing is remarkably accessible for non-mathematics types. In fact, he's a charming and engaging writer, who also manages to explain concepts that I haven't managed to "get" from other sources. Including the maths classes I've taken. "Concepts of Modern Mathematics," "Does God Play Dice," "What is Mathematics," "Game, Set, and Math," and many other titles. In fact, I had to stop searching Amazon for him, because I can't afford to buy any more books right now!

Another possibility for you, if you choose to try it, is to babblemail me with specific questions. I'm a dork, so I can probably answer a fair number of them...

 

Re: Looking for a good book on statistics » Racer

Posted by Quintal on August 5, 2007, at 5:45:20

In reply to Re: Looking for a good book on statistics » Quintal, posted by Racer on August 5, 2007, at 1:06:37

Thank you Racer for the excellent suggestions. I settled on "How to Lie With Statistics" and another called "Statistics Without Tears" in the Penguin Science series, which is more on my ability level. The "Statistics for Dummies" got some shocking reviews so I stayed clear of that. I will look into Ian Stewart's work when I've learned to add up and do my times tables and square roots etc off by heart. Someone recommended a basic grasp of calculus as a starting point for one book. I'm sure I've covered it and probably even used it in physics before but I have no idea of what it might be or what it does. Something to do with Issac Newton I believe? He invented calculus did he not?

Well while I await delivery could be so kind as to you enlighten me on the meaning of the term 'P' in this study:
__________________________________________________

Hum Psychopharmacol. 2007 Jul 24; [Epub ahead of print]Click here to read Links
Effectiveness and acceptability of sertraline and citalopram in major depressive disorder: pragmatic randomized open-label comparison.
Matreja PS, Badyal DK, Khosla P, Deswal RS.

Department of Pharmacology, Christian Medical College and Hospital, Ludhiana, India.

OBJECTIVE: Citalopram and sertraline are widely prescribed selective serotonin reuptake inhibitors (SSRIs). There is no conclusive evidence to show superiority of citalopram or sertraline in terms of efficacy or tolerability. Hence this study was designed to compare short term efficacy and safety of citalopram and sertraline in major depressive disorder (MDD) in Indian patients. METHODS: In an open, randomized study, 100 patients were divided into two groups. In Group A (n = 50) patients received citalopram (20-60 mg/day) for 6 weeks. In Group B (n = 50) patients received sertraline (50-150 mg/day) for 6 weeks. Patients were evaluated at baseline and then at 1, 2, 3, 4, 5, and 6 weeks. RESULTS: There was significant improvement in Hamilton depression rating scale (HDRS), Montgomery and Asberg depression rating scale (MADRS) and Amritsar depressive inventory (ADI) scores (p < 0.05) with both the drugs. However, the decrease in score was more with citalopram (p < 0.05). Onset of action of citalopram was earlier as compared to sertraline (p < 0.05). The number of responders and remitters was also more with citalopram (p < 0.05). No serious adverse event was reported in either of the groups. CONCLUSION: Citalopram had shown better efficacy, earlier onset of action and more number of responders and remitters as compared to sertraline in MDD in Indian patients. Copyright (c) 2007 John Wiley & Sons, Ltd.

PMID: 17647298 [PubMed - as supplied by publisher]
__________________________________________________

Q

 

Sorry, that should have been a little p » Quintal

Posted by Quintal on August 5, 2007, at 5:49:36

In reply to Re: Looking for a good book on statistics » Racer, posted by Quintal on August 5, 2007, at 5:45:20

I exaggerated the importance of p. Please accept my sincere apologies for this error of judgement.

Q

 

p-values... » Quintal

Posted by Racer on August 8, 2007, at 13:24:16

In reply to Re: Looking for a good book on statistics » Racer, posted by Quintal on August 5, 2007, at 5:45:20

OK, the p value, or probability value, has to do with hypothesis testing. Basically, the smaller the p value, the more significant. In most sciences, .05 is pretty much the bottom end of the significant range, with p values of .01 to .001 considered pretty strong evidence. A p value of .05 means there's about a 5% probability of error.

I can't find my Stats text, but I'll look for it later and try to give a better explanation. Statistics isn't really math-like, it's much more about logic and thinking, so it's harder for me to remember or understand well enough to explain to others... Man, who knew I'd be such a dork?

 

Re: p-values... » Racer

Posted by Quintal on August 9, 2007, at 14:30:40

In reply to p-values... » Quintal, posted by Racer on August 8, 2007, at 13:24:16

Thank you very much Racer. The 'Stats without tears' book was delivered today and I shall set about it when I have a spare moment. I'm relived to hear stats is not very math-like, who knows, I may even enjoy it?

Q


This is the end of the thread.


Show another thread

URL of post in thread:


Psycho-Babble Books | Extras | FAQ


[dr. bob] Dr. Bob is Robert Hsiung, MD, bob@dr-bob.org

Script revised: February 4, 2008
URL: http://www.dr-bob.org/cgi-bin/pb/mget.pl
Copyright 2006-17 Robert Hsiung.
Owned and operated by Dr. Bob LLC and not the University of Chicago.