@Anne
Well, I'd say that I'm not a mathematician either, nor a statistician. However, the bulk of my research deals with quantitative data, so I've had to learn some statistics. There really is no other alternative. If you intend to make some sort of predictions based on a quantitative body of data, utilizing some kind of statistical technique is the only valid way to do so.

What you really need are suggestions for good intro to stats books. I found Introductory Statistics with R by Peter Dalgaard to be a good primer.


@Yuri
I fully agree with you that the simplest appropriate method is the best. However, the key point there is that not all simple tests are appropriate for all kinds of data. For instance, once you have more than two comparison groups, a t-test is no longer appropriate, so you'll need an ANOVA. Furthermore, if your data violates the independence assumption, even more complex models are needed to appropriately analyze your data. So, if you have 500 observations from 50 speakers, and you want to see if men are significantly different from women, you'll have to use a more complex mixed-effects model to appropriately analyze that data.

I agree that it is not good to use statistics as a black box. So, while I don't understand all of the mathematics behind, say, a mixed effects model, there are a few things I do try to understand before I'll use a model. First, I need to know what the assumptions of the test are, and how my data may violate them. Second, I need to know if there are common pitfalls or confounds with this test that might arise given my data. Third, I need to understand some basics about how the test is computed, if not the mathematical details. Fourth, I need to understand how to properly interpret the outcome of the test.

Steps 1 through 4 usually involve repeatedly going back and forth from running the test on my data, and reading about the test.

-Joe

On Wed, Sep 29, 2010 at 5:30 PM, anne marie devlin <[log in to unmask]> wrote:
Dear colleagues,
this has turned into a very interesting and far reaching discussion.  Although not completely mathematically illiterate, I am certainly not a mathematiciian.  I have been dealing with my data in a simple and 'manual' way as I understand fully what is going in and what is coming out.  I am  nervous about the use of statpacks as I do not fully understand how the results occur and the significance of them.  This leads to a feeling of lack of control.  It would be very beneficial to hear how other non-mathematicians have dealt with these issues.
Regards
Anne Marie
 
Date: Thu, 30 Sep 2010 03:06:09 +0700
From: [log in to unmask]
Subject: One should be aware
To: [log in to unmask]


Dear colleagues, I started the discussion what tools to use for the analysis because many of linguists do not use all these complex statistical packets correctly. The other thing is who knows what and how the data are being analysed in them. We have a sort of a "black box" which has the entrance in which you put your data and the outcome where you receive your results. You must be quite sure that the data are analysed correctly. The more simple criteria you use, the better. This is why I stopped using all the stat. packets and began using very simple criteria like the coefficient of variation, the Chi-square and the t-test. At least, I am sure about the outcome. More often than not, it is advisable to try you data manually than to use the statpack. Looking forward to hearing from you if you agree with me that using statpacks may give you strange results. The fact that so many scholar spoke on the list about how to use the statpacks showed that it is a burning question for linguists and other researchers who are not specialists in mathematics. Yours sincerely Yuri Tambovtsev, [log in to unmask]

The Variationist List - discussion of everything related to variationist sociolinguistics.

To send messages to the VAR-L list (subscribers only), write to:
[log in to unmask]

To unsubscribe from the VAR-L list, click the following link:
http://jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=VAR-L&A=1


The Variationist List - discussion of everything related to variationist sociolinguistics.

To send messages to the VAR-L list (subscribers only), write to:
[log in to unmask]

To unsubscribe from the VAR-L list, click the following link:
http://jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=VAR-L&A=1



The Variationist List - discussion of everything related to variationist sociolinguistics.

To send messages to the VAR-L list (subscribers only), write to:
[log in to unmask]

To unsubscribe from the VAR-L list, click the following link:
http://jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=VAR-L&A=1