Statistical Methods, HT2012

Multilevel Analysis

This webpage gives the material for the presentations in this course by Professor Tom Snijders in weeks 2 and 4 of Michaelmas Term 2012.

Literature

Note the website associated to this textbook, with data sets, R scripts, and further material.

Meetings

1. Week 2: lectures Tue 10-12, Wed 10-11.
2. Week 4: lectures Tue 10-12.
3. Week 6: practical Fri 9-13.

Lecture Notes

Practical

Text of the non-assessed part of the practical, week 6, HT.
Data set.

Text of the assessed part of the practical, week 6, HT.
Data set.

Background material

Multilevel analysis

1. A non-technical introduction (not a single formula) to the ideas of multilevel modeling:
   Craig Duncan, Kelvyn Jones, & Graham Moon, Context, composition, and heterogeneity: Using multilevel models in health research, Social Science and Medicine, 46 (1998), 97-117.
2. Tutorial with examples using nlme and lme4:
   Daniel B. Wright and Kamala London, Multilevel modelling: Beyond the basic applications, British Journal of Mathematical and Statistical Psychology, 62 (2009), 439-456.
3. Lisa M. Sullivan, Kimberly A. Dukes and Elena Losina, Tutorial In Biostatistics. An Introduction To Hierarchical Linear Modelling. Statistics In Medicine 18, 855-888 (1999).
   This is a good tutorial, except that what is stated in Section 4.2.2 about testing variance components is wrong; and Section 4.2.3 about testing random effects is potentially misleading, making no distinction between comparative and diagnostic standard errors. Section 5 about software is not relevant for this course and, as is natural for a paper from 1999, out of date.
4. Preprint of chapter 3 of Jan de Leeuw & Erik Meijer (eds.), Handbook of Multilevel Analysis. Springer, 2008, Diagnostic checks for multilevel models by Tom Snijders & Johannes Berkhof.

Some examples of applications of multilevel analysis:

5. A. Gelman, B. Shor, J. Bafumi, D. Park. Rich state, poor state, red state, blue state: what's the matter with Connecticut? Quart. J. Polit. Sci. 2007: 345-367.
6. Dietrich Oberwittler, A Multilevel Analysis of Neighbourhood Contextual Effects on Serious Juvenile Offending; The Role of Subcultural Values and Social Disorganization, European Journal of Criminology, 1 (2004), 201-235.
   See also the correction note to this article in European Journal of Criminology, 2 (2005), 93-97.
7. Mark Levels, Jaap Dronkers, and Gerbert Kraaykamp, Immigrant Children's Educational Achievement in Western Countries: Origin, Destination, and Community Effects on Mathematical Performance, American Sociological Review, 73 (2008), 835-853.
8. Nigel Rice, Roy Carr-Hill, Paul Dixon and Matthew Sutton, The Influence of Households on Drinking Behaviour: A Multilevel Analysis. Social Science and Medicine, 46, 971-979, (1998).
9. Marcel Simard and Alain Marchand, A multilevel analysis of organisational factors related to the taking of safety initiatives by work groups. Safety Science, 21 (1995), 13-129.
10. Hugo Quene and Huub van den Bergh, On multi-level modeling of data from repeated measures designs: a tutorial. Speech Communication, 43 (2004) 103-121.
11. Robert M. Kunovich, Social structural position and prejudice: an exploration of cross-national differences in regression slopes. Social Science Research , 33 (2004), 20-44.

Supporting R materials

1. Introduction to using the nlme package in R for multilevel analysis:
   John Fox, Linear Mixed Models. Appendix to `An R and S-PLUS Companion to Applied Regression'.
2. Another extended example of using R for multilevel analysis Douglas Bates, Examples from Multilevel Software Comparative Reviews.
3. A good introduction that illustrates the use of the R package lme4:
   Maindonald, J., and Braun, J. (2007) Data Analysis and Graphics Using R, 2nd ed. Cambridge: Cambridge University Press.
4. Description of the nlme package.
5. Description of the lme4 package.

The lectures are introductions/overviews and you should expect to do your own reading of some of the additional material given above.