O IDHM é obtido pela média geométrica dos três subíndices das dimensões que compõem o índice: longevidade, educação e renda.

Obtaining data

    In spite of being collected at the residential level, census data is only released in an aggregate way in order to avoid the exposure of personal information. In the case of information contained on the questionnaire used in the domicile universe, data is available only for census sectors. In the case of the sample questionnaire, that from which the Atlas obtained the greater part of its information, data is available only for areas under consideration.

    To obtain access to data from the questionnaire for spatial divisions different than those corresponding to the areas under consideration, the users should submit a project to be evaluated by the IBGE for the proposal of a new aggregation, while observing the demands for statistical reliability and obeying the criteria that are rigorously evaluated by the technical committee. Among the parameters evaluated by the committee, especially highlighted is the demand that areas created should have at least 400 private, permanent residences.

    Once the project is approved for the creation of new spatial divisions, the researchers utilize a special room made available by the IBGE, called the “secret room”, where they have access to the micro data of the census according to the spatial aggregation. The aggregated results obtained based on the micro data have yet to be evaluated for consistency by the IBGE before being finally released to the user.

    The above described process shows the procedures undergone by the project team to obtain the Metropolitan Region Human Development Atlas, in the case of spatial divisions called Human Development Units (HDU).

What is a census sector?

    A census sector is constituted of contiguous areas, delimited to attend the parameter of census collection and control. It is situated in a single urban or rural block. The number of residences contained and its territorial dimensions are defined in such a way as to allow information to be collected by one census taker.

It is defined due to the censor taker route, considering physical barriers and the layout of the streets. Its configuration is similar to the routes taken by meter readers, mail deliverers or garbage collectors (in city blocks, large buildings and etc.). With this, it tends to be more homogenous in areas with greater population density and less in sparse areas.

What is a weighting area?

    A weighting area, in turn, is a geographic unit formed by a grouping of census areas for the application of calibration procedures of the estimates obtained from the sample with the information known to the population as a whole.

Criteria for the Division of HDUs

    The HDUs were outlined seeking to generate more homogenous areas, from the socioeconomic point of view, than the IBGE weighting areas. In other words, they are constructed with the objective of better capturing the diversity of situation related to human development that occurs within the intra-metropolitan spaces, notably in large municipalities, to uncover what is hidden by the aggregate municipal averages, as presented in the Municipal Human Development Atlas.

    While the logic of the IBGE weighting areas meets technical requirements related to the process of collection and samples, the HDUs are directed toward the spatial analysis of the Metropolitan Regions (MR), by means of spatial divisions of greater socioeconomic homogeneity, with the objective of portraying the intra-metropolitan inequalities in a more decisive way.

    To propose these new spatial divisions for the analysis and approval of the IBGE, no econometric model was used that could be applied to generate conformity of all HDUs in all MRs in the country. The urban/metropolitan life characteristics lead to very diverse socio-spatial conformity and the homogeneity obtained by means of variable A or B would not necessarily be equal (or would delineate the same area) as measured by a second variable, considering that the very census areas already bring some level of heterogeneity. Each place has its own morphology and history of occupying that urban space.

    As such, going beyond the possibilities associated with the analysis of the variables available for all census sectors (which could be contemplated by some econometric model) there are diverse situations that could alter the census sector data (such as the existence of an only apartment building that could alter the data average for a census sector) and urban occupation characteristics (such as age and settlement profile), that may not be captured, for example, by the income variable and could interfere with the HDU socio-demographic indicators. This requires that a “customized” proposal be constructed for each HDU for each MR, one that fits the specifications for each metropolitan space considered.

Locale validation

    In the process of delimiting the HDUs, it was necessary to count on the knowledge and technical collaboration of research institutions in all of the MRs covered by the Atlas. This was necessary so that they could, based on socioeconomic information at the census sector level (information was made available from the universal census, such as income, number of bathrooms per residence and more), propose a more homogenous configuration of intra-metropolitan spatial divisions that met IBGE technical demands.

    Besides this, these new spatial units should be recognized, including by denominations already used by the population. As much as possible, these units would be located in aggregations of census sectors that represent contiguous areas, seeking to facilitate their nomenclature and recognition.

    The construction of the HDUs, therefore, was a task that demanded the articulation of an expressive set of partners (articulation by means of the Ipea platform of internet research – Ipea Network). The partners should propose the configuration of these intra-metropolitan spaces, respecting the IBGE criteria and demands, which should be as homogenous as possible, in socioeconomic terms (homogeneity), contiguous (adjacency) and be recognized by the resident population (identity).


    In the process of constructing the new agglomerations that would come to be HDUs, many times it was difficult to attend the criteria of a minimum showing of 400 permanent, private residences and, simultaneously, observe the criteria of socioeconomic homogeneity.

    Considering this technological-methodological problem, an agglomeration of non-contiguous areas that had similarities among themselves was chosen, following the aspects previously described, and that could be recognized independently of size. This process of agglomeration was implemented until the minimum size of 400 permanently sampled private residences was achieved. This procedure is justified, for example, when there are small villas or slums inserted within high-income neighborhoods, or inversely, when there are luxury condominiums located in low-income neighborhoods or that have domiciles with distinct characteristics.

    In the end, it was as if new areas of consideration were formed that had greater socioeconomic homogeneity, but without spatial contiguity.

    The technical limitation described above demanded the construction of the HDUs to follow two stages. In the first stage, the criteria of homogeneity, contiguity and identity should be respected, without necessarily meeting the criteria of forming areas with 400 permanent, private residences (though this was desired).

    The result of this division, proposed by the Atlas coordination and approved by the Ipea Network partners, corresponds to the HDUs as they are presented in the Metropolitan Regions Human Development Atlas.

    In the second stage, it was necessary to meet the IBGE criteria and demanded for extraction of the sample questionnaire data (sample census). This implied the need to aggregate homogeneous HDUs into socioeconomic terms and urban occupation profiles in order to attend the criteria defined for the extraction of information.

    As such, the data presented in the Atlas expresses the average data of the HDUs that were aggregated for the purposes of data extraction from the IBGE. In other words, if an HDU presented in the Atlas did not meet the technical demands imposed by the IBGE, the data presented for that HDU does not pertain exclusively to it. It is the average of results relative to the HDU in question and other homogenous HDUs that also presented this technical limitation and which were aggregated for the purposes of data extraction. In this way, the HDUs whose data was obtained jointly with other HDUs, share the same indicators, except in the case of those indicators that have their origin in the universal census (such as illiteracy and population).

    Another important observation regarding the construction of the HDUs deals with the small municipalities that are part of the MRs. When a municipality has a small population, in general, with a single area for consideration, it also corresponds to a single HDU. In this way, its indicators correspond to those observed for the municipality as a whole, given the impossibility of creating a division of its space, even though recognizing the existence of socio-spatial inequalities in its territory.

Regional Division

    This deals with the construction of the HDUs that appear in the Atlas and in the eventual aggregation of HDUs for the purposes of sample data extraction. The municipalities that have regional divisions compatible with the census sectors for the 2010 demographic census were respected such that their indicators are available for territorial levels (administrative regions, regions, districts, sub-districts, etc.), intermediaries and infra-municipalities that are HDU aggregations that compose these scales.

Source of Information

    Demographic Censuses conducted by IBGE