Hello Cave, Wednesday, October 29, 2003, 2:09:49 PM, you wrote:
C> I've got a 4*2*8 table where: C> - The first dimension is the year of the data (4 values: 1999-2002). C> - The second dimension holds number of occupational injuries and total C> number of workers (a worker can experience more than one injury in the C> same year, but each injured worker is counted appart). C> - The third dimension represents groups according to experience at C> work (less than 6 months, from 6 to 12, ... etc., a total of 8 C> groups). C> I'm interested in analizing the relationship between the second and C> third dimensions, that is, the relationship between injuries by worker C> and experience. C> My question is �wich is the appropiate analysis for my study? C> I would thank very much any help, and also if you have references of C> this kind of analysis. C> Cave. Hi, I would build a Logit model for your data. This will tell you (in terms of the crossproduct ratios) what is the impact of experience on injury. And wether the particular association pattern between these two is stable or not over time (the third variable). As for the references, here are some. Please forgive my bias towards social science applications :). There are a lot of books and articles on Logit models. I advise taking a look, for instance, on "Applied Regression Analysis, Linear Models, and Related Methods" by John Fox. There is a chapter on Logit models. There is also a book released in Sage series "Quantitative Methods in Social Sciences": "Logit Modelling, practical applications" by Alfred Demaris. You may try Goolging for "Logit model" some online introductions will come up. I hope this helps. Cheers Micha� . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
