Census documentation


The censobr package includes a few functions to help users navigate Brazilian census data, its variables and its methodology. All of these functions download supporting documents in either .html or .pdf format, and opens the documents on a web browser.

Like all other datasets in censobr, these support documents are downloaded and cached locally the first time when users run the function. This way, when users need to take a peak at the documentation again, the functions simply load local files, making it super quick and convenient.

Data Dictionary

The data_dictionary() indicate the definition of each variable, and the meaning of their categories in the case of categorical variables. The function currently covers the data dictionaries for all Brazilian censuses since 1970 (1970, 1980, 1991, 2000 and 2010), and it includes dictionaries for the variables of both microdata (sample portion of the census) and for the variables available in census tract-level aggregate data.

# Microdata variables
data_dictionary(year = 2010, table = 'microdata')

# Census tract-level variables
data_dictionary(year = 2010, table = 'tracts')

Obs. For the years 2010 and 2000, the the dictionaries of variables are also available online from the censobr website. Other years will also be eventually included on the website.


Oftentimes, it is really important to understand the structure of the questionnaire used in surveys. The questionnaire() function includes the questionnaires used in the data collection of all Brazilian censuses since 1970.

Here, users only need to passe the year parameter indicating the edition of the census of interest.

# 2010
questionnaire(year = 2010)

# 1970
questionnaire(year = 1970)

Interview manual

Finally, the interview_manual() function downloads and opens on a browser the “Manual do Recenseador”, i.e. the manual of instructions for IBGE’s census takers (recenseadores) on how to collect the census data.

Manuals for all censuses since 1970 are available.

# 2010
interview_manual(year = 2010)

# 1970
interview_manual(year = 1970)