Welcome!  Today is . You are the visitor since 19th March,2007.

Brief Introduction
Reconstructing the age-patterns of marriages, births, and deaths in Qing-dynasty China has depended in the past primarily on genealogies. Here we present a preliminary study of the lower Yangzi valley that demonstrates in action the technical methods needed to use a different sort of information for this purpose: the biographies of virtuous women in local gazetteers. The reader is invited to download our data files containing more than twenty thousand cases, to use or adapt our executable PERL programs, and to examine, improve on, or refute our reconstruction based on the consistent  interlocking four main types of data and probabilistic computer models.

Principal Results as of 16 March, 2007
The project on Qing Dynasty Population Dynamics (QDPD) is continually making small improvements to its calculations and correcting the errors that still turn up occasionally in the database. The principal results, though remarkably stable, thus shift fractionally from time to time. For rapid reference, the current values for the Lower Yangzi Valley around 1800, as of the date shown above, are

• Female expectancy of life at birth:                                 
• Female expectancy of life at age 10:                             
• Male expectancy of life at birth:                                       
• Male expectancy of life at age 10:                                   
• Mean age of females at first (major-mode) marriage:  
• Sample standard deviation in female ages at first marriage:      
• Gross Reproduction Rate (GRR):                                          
• Mean difference in age between husband and wife:              
　

For the current Lower Yangzi Valley Standard Life Tables, see Menu of Downloads.
Major areas of inadequately precisely determined values: (1) the early years of female life (including infanticide, due to a lack of data on surviving daughters at given ages for mothers, as opposed to those for surviving sons); (2) the latest years of male life (above about 65, when parents-in-law data become thin). The relatively low female e(0) for the LYV can be brought to approximate equality with the male e(0) by assuming that the first-year female mortality in excess of the male mortality was due to female infanticide and differential neglect of girl babies. (This is currently 11.5 %, as opposed to the 10.3% stated on page 3 of the second article in the Menu of Downloads.) When we equalize the female and male l(1), but retain the other proportionate relations between the female l(x) values for higher ages (using unsmoothed sui data, and the current Brass's alpha = 0.0012, and beta = 1.0042 to adjust the standard) we find a female e(0) of 28.5 years versus the male 28.4, . There is, however, nothing in the data that directly justifies such a procedure. We have therefore not attempted, for the time being, to create an alternative set of interlocking values for a solution that incorporates this modification, reasonable though it seems.

The male survivorship proportions  for the oldest ages likewise probably need to be brought down, but, so far as we can see, there is no comparably non-arbitrary procedure presently available for doing this. 

The e(0) are probably maxima, as 'gentry families', defined as those headed at some point by a holder of an imperial examination degree or an official rank, are overrepresented in the data by about 5 times compared with the actual population.)

Academic affiliations:
Dr. Fox, Research School of Pacific and Asian Studies, Australian National University;
Emeritus Professor Elvin, Research School of Pacific and Asian Studies, Australian National University, Sinological Institute, University of Heidelberg, and St Antony's College, Oxford;
Dr. Wen, Center for Geographic Information Science, Research Center for Humanities and Social Sciences, Academia Sinica, Taipei, Taiwan.