Mexico population census

I’m working with data from the Mexican Census (1990 & 2000) and focusing on economic variables, specifically those related to economic activities. In the databases, I have the variable “IND” which, as described, denotes the industry of the person’s workplace, represented by 5 digits for 1990 and 3 digits for 2000.
I have a couple of questions regarding this variable:

  1. What classification system does the code follow? Is it based on CMAP or another classification scheme, or is it arbitrarily devised by IPUMS? I’m asking because I’ve noticed that there’s no match for the CMAP94 and the code that is provided in the IPUMS.

  2. From which source or question does this disaggregation of industries originate? I’m asking because the data available on the INEGI website seems to only provide industry information at an aggregate level (similar to the “INDGEN” variable in IPUMS). So, is the “IND” variable sourced from INEGI? If so, I’m curious about which question it corresponds to in relation to the questionnaire available on the INEGI website.

I appreciate any insights you can provide to clarify these points. Thank you!

The IND variable in IPUMS International uses codes that are specific to each sample. In order to retain the maximum amount of information for each sample’s specific industry codes, we do not harmonize IND across samples. The sample-specific IND codes for Mexico 1990 and Mexico 2000 can be found in the variables MX1990A_IND5 and MX2000A_IND3, respectively. These are unharmonized variables that come from the original Mexico census data. I am not sure whether the designers of the Mexico census modeled their industrial classification system after an existing classification system, or whether it was developed specifically for the Mexico census. This documentation from INEGI suggests they may have implemented a classification system similar to NAICS in the 2000 Census, which is supported by the similarities between those industry codes and the NAICS codes. However, that document does not seem to provide information on what system was used in the 1990 census. INEGI may be able to provide you with more information.

IPUMS also provides INDGEN, which is a very general, harmonized industry variable. The groupings roughly conform to the International Standard Industrial Classification (ISIC). I would guess that any published estimates you are referring to from INEGI use their own aggregation of the original industry codes, which may coincidentally bear some similarity to the IPUMS International INDGEN variable codes.

Thank you for your reply. Regarding the variable “INDGEN”, there are no problems. The label is well defined.
My question is about the origin of the variables MX1990A_IND5 and MX2000A_IND3. Are these sample specific IND codes from IPUMS or from INEGI? If the data (the label) comes from the original Mexican census, I am wondering if these codes come from the data itself or if they were constructed by IPUMS?
I am trying to manually match these codes to the CMAP classification using the label provided by MX1990A_IND5. However, it would help me to know if the codes in IND come directly from INEGI or from IPUMS.

Thank you for your help.

As I indicated in my previous post, the variables MX1990A_IND5 and MX200A_IND3 are the original industry variables in the original Mexico census data. IPUMS commissions professional translation of value labels into English, but does not aggregate the categories, recode, or harmonize these specific variables. While IPUMS processes all data that are offered on our platform, these are the original unharmonized variables as they came to us in the Mexico census data from INEGI. You can find links to the Spanish language versions of the codes under the codes section of IND. I am not sure whether INEGI attempted to model the industry codes in the 1990 censuses after the CMAP classification system or some other existing system, or whether they created their own occupation classification system for the purpose of the Mexico census; it does appear that the industry classification system used in the 2000 Mexico census is based on the NAICS.

Thank you for your assistance.