# Malcolm M. Dow

- Dow-Eff References
- Born 1947 ; Gwen Stern; now Amberg WI
- https://www.mylife.com/malcolm-dow/e187026490668
- https://www.semanticscholar.org/paper/Monogamy-in-the-Pre-industrial-World-Dow-Eff/a7cac8f74dac8bd1030c11afaa40509428aa37ba
- Dow Similarities
- Dow-Eff_References -- Google Scholar Citations missing 2007 Microsoft M.M.Dow
- Dow, M. M. (2007). Galton’s Problem as multiple network autocorrelation effects. Cross-Cultural Research, 41, 336-363.
- https://vpn.nacs.uci.edu/+CSCO+00756767633A2F2F71626A6179626E712E66636576617472652E70627A++/static/pdf/674/art%253A10.1007%252FBF00435200.pdf?auth66=1400095454_248f37614d2f2de7b5d7467f9c197b9d&ext=.pdf D. Beasley 1988 "Two or Three-Stage Least Squares." Computer Science in Economics and Management 1:21-39.

Harry H. Kelejian and Prucha

Wife: Gwen @gmail.com Take care of a faulty address with mmd383northwestern.edu and no @, ie no mmd383@northwestern.edu

- http://scholar.google.com/citations?user=cZfXhJ0AAAAJ&hl=en For the 1985 article see:

- Malcolm M Dow. 1985 Agricultural intensification and craft specialization: A nonrecursive model. Ethnology 24(2): 137-152.
- Abstract. Theory that stems from a non-recursive model developed by E. Boserup, F. Hole, KF Flannery, different assumptions about the relationship between the change in agriculture and the development of specialization of labor: the law of "least effort", the theory of surplus, the generalization of trade (food surplus can be traded against non-agricultural products) and other factors which variables depend on the population density, degree of settlement and level political integration.
- Théorie à partir d'un modèle non récursif élaboré par E. Boserup, F. Hole, K. F. Flannery, des différentes hypothèses concernant les relations entre le changement dans l'agriculture et le développement de la spécialisation du travail: la loi du "moindre effort", la théorie du surplus, la généralisation de l'échange (les surplus alimentaires peuvent être échangés contre des produits non agricoles) et les autres facteurs dont les variables dépendent de la densité de la population, de leur degré de sédentarisation et de leur niveau d'intégration politique.

dépendent de la densité de la population, de leur degré de sédentarisation et de leur

Chevrud, J. M., M. M. Dow, and W. Leutenegger (1985). The Quantitative Assessment of Phylogenetic Constraints in Comparative Analyses sexual dimorphism in Body Weight among Primates. Evolution 39(6):1335-1351.

SUBSAMPLE REPLICATION IN CROSS-CULTURAL SURVEYS Paperback – January 1, 1987 by Malcolm M Dow (Author)

## Contents

## Wikisites

## AIC (Akaike information criterion)

- Schwarz, Gideon E. (1978). [Gideon Estimating the dimension of a model]. Annals of Statistics 6 (2): 461–464. doi:10.1214/aos/1176344136. MR 468014. Wikipedia:Bayesian_information_criterion#References

- Dow, Malcolm M. 1986. Procedures for Network Autocorrelated Disturbances Models. Sociological Methods Research May 1986 vol. 14 no. 4 403-422. http://smr.sagepub.com/content/14/4/403

- Doug: Restriction of slopes of 1&2 to be equal across subsamples as in Table 2 Model 3, Schwarz's AIC approach (Akaike information criterion), is a great resolution to this problem. I don't see an AIC R library that solves his AIC approach -- do you see any way we could implement this option in DEf01?
- Not immediately. Seems to me that it would need some additional R programming before trying to do it with multiple imputed data sets. I also do not know which of the routines in the R library might be useful try some of this - I wrote this paper long before I knew anything about MI, and I don't know much about R programming or the R library. Anthon's the expert. I also don't know at the moment if the R-sqd from 2sls could be used with the F-stat procedures I outlined in my paper. I'll have to re-read the paper quite closely to see what exactly it was I said way back 30 years ago!

## 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

Dow, Malcolm M., and E. Anthon Eff. 2013. "When One Wife is Enough: A Cross-Cultural Study of the Determinants of Monogamy." Journal of Social, Evolutionary, and Cultural Psychology 7(3):211-238. (Link)

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.

Dow, Malcolm M., Michael L. Burton, & Douglas R. White. 1982. Network Autocorrelation: A Simulation Study of a Foundational Problem in the Social Sciences. Social Networks 4(2):169-200

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.
**Abstract:** - Empirical evidence that cultural traits are often nonrandomly distributed because of the individual or combined effects of common history, diffusion, borrowing, and/or other types of cultural transmission processes has been accumulating for decades. Because many cultural traits have recently been shown to be influenced by more than one transmission process, it has become a methodological priority in comparative research to develop statistical methods that can simultaneously incorporate multiple transmission processes. This article proposes a multiple network autocorrelation effects model and associated two-stage least squares (2SLS) estimation procedures. The network autocorrelation effects model offers an alternative interpretation of how cultural trait transmission processes operate than does the network autocorrelation disturbances model. Conceptual differences between the two classes of models suggest that the network effects specification will be more generally applicable in comparative studies. An empirical example demonstrates the substantive value of the multiple network autocorrelation effects model and the widely available 2SLS estimation procedures.

Galton's Problem * network autocorrelation effects model * two-stage least squares * trait transmission * ecological constraint

- 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

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.

- 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

- Imputing_the_data#Eff_and_Dow -- Eff, E. Anthon, and Malcolm Dow.
**2009 pdf quick download**. 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#3 art 1. Structure_and_Dynamics_contents#Issue_3#3_2009

- pp 94 E. Anthon Eff. Updated scripts for R in Eff and Dow (2009) Issue 3#1 art 1. Structure_and_Dynamics_contents#Issue_5#2_2012

- 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