Introduction to Research

This section of the Guadalajara Census Project Website is devoted to presenting the basic data from the censuses of 1821 and 1822, in so far as it is practical. Because the data from the two years contain duplicate information on those individuals who remained in the city for both counts, and because 1821 is nearly complete while only about half the returns for 1822 have been located, the data for each variable will be presented as a table (“cross tabulation”) with separate columns for 1821 and 1822. Please note that, despite the official requests for complete data on all individuals (see Censuses/Authorization for the 1821 census), the census takers (who were the political official responsible for each district or cuartel in Spanish) did not uniformly collect all the requested information. For example, while they gave names and marital status for nearly everyone, they gave place of birth (patria) for only about 43 percent of the population and calidad (ethnicity) for just over one-third of the city’s residents. However, because there appears to be no particular pattern as to those districts which did or did not collect certain type of information and because the data we do have is for thousands of individuals, one may tentatively, at least, make reasonable generalizations about early nineteenth century Mexican urban life. ¹

The data for the reliability of the data varies, of course. The so-called “Literal” variables are data written down by the census takers or scribes, and we have simply copied the data as we saw it. Therefore, the reliability depends on the accuracy of the original data, and its faithful copying by our data entry personnel. We have taken particular pains to insure that the error rate for our side of the job is as low as it is possible to make it. (See Guide to Database Designing/Error Detection & Verification Procedures.) The reliability of the original collectors of this information is always an issue, however, and is discussed in the same essay.

The “constructed” variables are those created by the GCP staff from the information contained in the original census manuscripts. These range from the relatively dependable variable such as “sex”. Usually gender was not stated specifically by the census taker but easy to determine because the given name usually (but not always) is gendered (o ending for male, a for female). Or one can use the same reasoning to establish sex by one’s occupation or marital status. Other constructed variables are much more difficult, such as determining the relationship of each individual in the household to that of the head of the household, when the census taker neglected to give that data. In such cases where interpretation is necessary, we have established consistent procedures for each interpretation or assumption. For a more thorough discussion of the types of variables and our interpretations and assumptions, see Codebooks/Description.

While it is not practical to present each piece of information (several million) even in summary form, we have chosen the more interesting and/or useful. (Given names, for example, would if printed separately take up to 90 pages of single space text just to present each different name with the number of persons who had that name.) Frequencies of individual variables are presented by year of the census, for most literal variables. Tables of several variables each are presented, by year, where interpretations might yield some interesting conclusions. For example, marital status by sex reveal the high likelihood of widowhood for the majority of women over 50, compared to widowers. Each table will be linked to the specific variable description in the codebooks in order to clarify questions of coding or interpretation. (Or one may first choose to read the codebooks; each selected variable there is linked to the tables in this section. Codebooks)

Notes:

1. For a technical discussion of the issue of sampling in general, see Alan Bryman and Duncan Cramer, Quantitative Data Analysis for Social Scientists (London, 1990), pp. 98-104. A fine, non-technical explanation is Loren Haskins and Kirk Jeffrey, Understanding Quantitative History (New York, 1990), pp. 121-66. More technical but for students of history is R. Darcy and Richard C. Rohrs, A Guide to Quantitative History (Westport, CT, 1995), pp. 7-26.

Search WWW Search www.fsu.edu