Malcolm M. Dow
Contents |
Wikisites
Publication indices
- Google scholar indexed publications: also Google http://www.google.com/search?q=Malcolm+M.+Dow&ie=utf-8&oe=utf-8&aq=t&rls=org.mozilla:en-US:official&client=firefox-a
Sage publications in Cross-Cultural Research
Publications
Cheverud, James M., Malcolm M. Dow and Walter Leutenegger. 1985. The quantitative assessment of phylogenetic constraints in comparative Sexual dimorphism in body weight among primates. Evolution 39: 1335-1351.
Dow, Malcolm M. 1989. Categorical analysis of cross-cultural survey data: Effects of clustering on chi-square tests. Journal of Quantitative Anthropology, 1, 335-352.
Dow, Malcolm M. 1991. Statistical Inference in Comparative Research: New Directions. Cross-Cultural Research 25(1-4): 235-257.
Dow, Malcolm M., Michael L. Burton, Douglas R. White, Karl P. Reitz. 1984. Galton's Problem as Network Autocorrelation. American Ethnologist 11(4):754-770.
Dow, Malcolm M., & Cheverud, J. M. 1985. Comparison of distance matrices in studies of population structure and genetic microdifferentiation: Quadratic assignment. American Journal of Physical Anthropology, 68, 367-373.
Dow, Malcolm M., Cheverud, J. M., & Friedlaender, J. S. 1987. Partial correlation of distance matrices in studies of population structure. American Journal of Physical Anthropology, 72, 343-352.
Dow, Malcolm M., D. R. White, M.L.Burton. 1982. Multivariate Modeling with Interdependent Network Data. Cross-Cultural Research 17:216-245.
Galton's problem and autocorrelation - see also Anthon Eff
- Dow, Malcolm M. 2007. Galton's Problem as Multiple Network Autocorrelation Effects: Cultural Trait Transmission and Ecological Constraint. Cross-Cultural Research 41(4):336-363. The Cultural Trait Transmission variables here correspond to vertical (language family proximity) and horizontal (special proximity) in the Standard Cross-Cultural Sample.
- Dow, Malcolm M. 2008. Network Autocorrelation Regression With Binary and Ordinal Dependent Variables Cross-Cultural Research 42(4):394-419.
- Dow, Malcolm M., and E. Anthon Eff. 2009a. Cultural Trait Transmission and Missing Data as Sources of Bias in Cross-Cultural Survey Research: Explanations of Polygyny Re-examined. Cross-Cultural Research. 43(2): 134-151. The Cultural Trait Transmission variables here are developed in Dow (2007) and corresponds to vertical (language family proximity) and horizontal (special proximity) in the Standard Cross-Cultural Sample.
- See: http://ccr.sagepub.com/cgi/reprint/41/4/428 Ember Ember and Low
- Dow, Malcolm M., and E. Anthon Eff. 2009 Multiple Imputation of Missing Data in Cross-Cultural Samples. Cross-Cultural Research, Vol. 43, No. 3, 206-229 (2009) http://ccr.sagepub.com/cgi/content/abstract/43/3/206 DOI: 10.1177/10693971093333
- Eff, E. Anthon, and Malcolm Dow. 2009. How to Deal with Missing Data and Galton's Problem in Cross-Cultural Survey Research : A Primer for R. Structure and Dynamics: eJournal of Anthropological and Related Sciences 3#2. Previous draft in pdf
- Eff, E. Anthon. 2008. "Weight Matrices for Cultural Proximity: Deriving Weights from a Language Phylogeny." Structure and Dynamics: eJournal of Anthropological and Related Sciences 3(2), Article 9. http://repositories.cdlib.org/imbs/socdyn/sdeas/vol3/iss2/art9
- Eff, E. Anthon, and Malcolm M. Dow. 2008. Do Markets Promote Prosocial Behavior? Evidence from the Standard Cross-Cultural Sample. http://econpapers.repec.org/paper/mtswpaper/200803.htm.
- Eff, E. Anthon, and Malcolm M. Dow. 2009. Market integration and pro-social behavior. To appear in Robert C. Marshall, Editor. Cooperation in Economic and Social Life. Society for Economic Anthropology Monographs Vol 26. AltaMira Press: Walnut Creek, CA.
- Dow, Malcolm M., and E. Anthon Eff. 2009. Cultural Trait Transmission and Missing Data as Sources of Bias in Cross-Cultural Survey Research: Explanations of Polygyny Re-examined. Cross-Cultural Research May 2009 43: 134-151.
Adjusted Chi-Square for Crosstabulation
- Dow, Malcolm M.. 1984. Categorical Analysis of Cross-Cultural Survey Data: Effects of Clustering on Chi-Square Tests. Journal of Quantitative Anthropology 1(4):335-352. ABSTRACT PDF http://www.quantitativeanthropology.org/index.php?journal=QA&page=article&op=view&path%5B%5D=21
p338-339 "In addition to constructing this fairly large cross-cultural sample, Murdock and White (1969) applied a number of statistical tests to assess the extent to which all effects of historical relationships among the sample units had been eliminated by their sampling procedure. Their conclusion from a series of tests was that if all historical influences were to be eliminated, there would be only about 20 regions of the world from which to choose a sample of completely independent societies. A sample of size 20 is certainly too small for most kinds of statistical analysis. However, grouping the 186 SCCS societies into 20 "clusters" and then taking this clustering into account in subsequent analysis provides one approach to dealing with the problem of sample unit interdependence. This is the approach taken in this paper. The societies in the SCCS are arranged in a series such that each society is placed between the two societies to which it bears the closest overall cultural resemblance. This ordering greatly simplifies the task of clustering the societies. The twenty clusters employed in the analysis reported below were constructed by systematically taking groups of 9 sequentially arranged societies at a time to get the first fourteen clusters, and then taking groups of 10 sequentially arranged societies to get the remaining six clusters. Preliminary results indicate that the analytical procedures employed below are little affected by small changes in the method of constructing the clusters, such as taking groups of 10 as the first six clusters and then groups of 9 for the remaining fourteen. It is also worth noting that Murdock and White (1969) point out that the SCCS sample is geographically divided into six major world regions, with approximately equal representation within each region. Thus the 20 clusters of societies analyzed below are also more or less geographically stratified, so the SCCS sample can be thought of as an approximation to a stratified cluster sample. As Kish (1987) notes, stratification in general will reduce estimates of variances, while clustering generally increases p339 such estimates either mildly or badly, and, those increases generally survive the ameliorating effects of stratification. Both the reductions and the increases in variance estimates are expressed in terms of the design effects for the sample, and these are reported below for a number of variables and contingency tables. To the extent that all cross-cultural samples are either representative samples or random samples drawn from judgemental lists of societies, application of analytical results based on probability sampling to such data sets is not strictly correct. However, it is now clear from numerous analyses of interval level SCCS variables using network autocorrelation methods, that ignoring the lack of independence within the SCCS and other comparative samples is potentially disasterous in terms of drawing valid inferences Dow el al. 1984; Dow 1984). The results reported below suggest that ignoring the interdependencies among the SCCS sample units may also lead to potentially hazardous inferences for categorical variable analyses."
Tests
- Shah’s Wald F-statistic with the appropriate degrees of freedom, and the p-value for the hypothesis test would be computed from the F-statistic. CROSSTAB would print the untransformed Wald chi-square test statistic value (asymptotically correct, but uncorrected for inflated Type I error rates in finite samples), along with the p-value based on the F-statistic. http://www.rti.org/sudaan/pdf_files/SUDAAN_Language_Manual_Addendum_903.pdf
adjusted Wald F (Fellegi, I.P., 1980. Approximate tests of independence and goodness-of-fit based on stratified multistage samples. J. Am. Stat. Assoc. 75, 216–268.
Wald, A., 1941. Asymptotically most powerful test of statistical hypotheses. Ann. Math. Stat. 12, 1–19.
Wald, A. 1943. Tests of statistical hypotheses concerning several parameters when the number of observations is large. Trans. Amer. Math. Soc. 54 426-482.
For Wald test see Wooldridge, J. M. 2002 Chapter 15. Econometric Analysis of Cross Section and Panel Data. MA: MIT Press.
http://en.wikipedia.org/wiki/Effect_size
