1980 Integration: SPSS files for Individuals and for Households

Then, as we did the 1980 census, a new ID number came into being for each person censused in the village. All of the individual sheets were updated with the new numbers and the fichero grew with the addition of newly arrived (births and inmigrants) persons to Tzin. But, for 1980, we now had the mainframe version of SPSS to work with. I prepared separate data sets for individual data and for household data. And for the first time, we used both a jefe ID number and a household ID number (tied to the geographical grid established for the 1960 census and continued thereafter in slightly modified form). It was with the 1980 census that I added to idea of having a side on each block -- i.e., 20A-01 instead of just going around the block counterclockwise, as Foster had created the system.

1980 cont

Continuing on with the story of the household data --

For 1980, we did a much more comprehensive data entry job, which resulted in the files called TZINHI1 and TZINH3. These files contain 47 varaiables and a total of 450 cases (less two cases not counted), for a real total of 448 cases. As you look at these files, you will see that they are arranged geographically from block 01A01 (record 001) through block 61L45 (record 431). Following record 431 are records 432 through 445, which involve households in which all of the persons are treated as migrants -- i.e., they were not counted as part of the core population of the village for the 1980 census. These included hh numbered 20B02 (with jefe 3149), etc. Then records 446 through 450 involve five cases for which we had jefe numbers and there out-migrant families but no physical household location to tie them to. These are jefes 4013 through 2681.

I assume that you can see all of these data in the TZINH3 computer file, for example, just as I am looking at the hard copy here.

What is important (among other things) is that this file has all of the hh items for the standard of living score for the 1980 40-point scale. One of the printouts I have here shows the distribution of the CC-scores, from a low of 1 to a high of 34.

The first (lowest) quartile went from 1 to 7 points,
the third quartile went from 14 to 19 points,
and the fourth (highest) quartile went from 20 to 34 points (since no hh scored higher than 34 out of the max. possible 40 points).

So, when we look at the problem of stratification, esp. the "submerged quarter" we can do a quartile analysis for ALL of the village censuses as a way to level out the differences in the different scales employed in 1945 (10 pts), 1960 (20 points), 1970 (30 points), 1980 (40 points), 1990 (also 40 points, but with a couple of items replaced), and 2000 (50 points).

By the way, I see a note here that the 1980 household data can be found in a file called TZIN80HH.SYS -- I assume that this is a SPSS system file which includes the raw data file.

Random thought: If, for the time being, we stick to the 1945-1990 data series, then we ought to be able to use the 1990 ID numbers to substitute for the older ID numbers for individuals (e.g., jefes in the HH files). Then, we have the problem of converting from the fichero ID for 1990 to the system developed by Eric for his analysis of parents, etc. But, finally, we should be able to have a series of data files which are compatible in terms of ID code numbers.

For Is the
1980 1970
Id: Id:

424 307? -- YES, it is 307 (Jose Calvo Melchor)
649 4?? -- not applicable (Alfredo Cira Cornelio, born 12 mar 1977)
1531 1170? -- YES, it is 1170 (Evaristo Francisco Tovar)
1680 1283? -- YES, it is 1283 (Francisco Guillen Maturino)
2038 ???? -- not applicable, (Jose Raul Lara Aparicio, born 6 dec 1978)
2099 1630? -- YES, it is 1630 (Nicolas Lopez Huipe)

Dear Doug,

Let me attempt to respond to your several sequential queries.

First, I will bring with me all of the references you listed on your email of 21 April, excepot for my article in the KAS (1974) which is not helpful in the present context.

Second, I also have the published Spanish version of the "Extending the Boundaries..." article.

Third, it appears that you (Doug) now have found the 45, 60, and 70 hh files. Is this correct? I am unable to open the attachment that appears to deal with this (a file called Tz45-70h.sav), but if you have it in good shape then we have no problem.

Fourth, you ask about the ID correspondences for the HH data and the last 3 fortran pages. Elsewhere, you refer to there existing 411 hh heads for 1970. I do not think so. I have only 360 entries in the paper copy of this file and the same 360 in the tables in my book on Migration and Adaptation. I think that the 411 figure must be the total number of family heads, since there are a goodly number of multi-family households.

Fifth, please tell me what you mean by "ID correspondences." Are you referring to the ID # of the head of the specific household -- which is the four digit number listed in cols 4-7 of the fortran coding sheet for 1970, 1960, 1945. This number should tie in to the master fichero with the 1970 ID codes; i.e, the second number from the left at the top of each fichero sheet for persons with a 1960 ID. Thus hh 001 has as its head Maria Socorro Zaldivar Aparicio (with ID 3049 in the 1970 fichero).

If these personal ID correspondences is what you need, then perhaps the best thing to do is get a secretary to create a file (e.g., in Excel) which has all of the extant ID numbers for 1960, 1970, 1980, and 1990, with provision for 2000 to be added. Some of this already exists, of course, and could serve as the basis for such a file. Then, this file could serve as a giant look-up table for tying together all of the different ID codes by decade. Is this what you have in mind?

Sixth, a small point about file names: the Paradox files for MIG90.db etc. are about the Migrants outside of the village, whereas the TzMig files are about the persons inside the village who have become migrants. There is some overlap in persons in these files, but most of the MIG90 folks are NOT in the master fichero because they were never living there during a census.

Well, that¥s all for now. Sandra is calling me to come to dinner. Get back to me tonight or in the morning on all the above and I can get another response to you before I leave for Mexico on Thursday am.