The Human and Ethical Aspects of Big Data

For the 2011 census, India began a grand experiment wherein every citizen was to be photographed, fingerprinted, and questioned regarding their marital status, education, and occupation. Mind you, such data gathering is not unusual—nations have long conducted censuses—but the scale to which India chose to under take such a vast digitization of information was unprecedented. In earlier times, the Roman Empire was particularly detailed in the matter, as were the Normans, as reflected in the Domesday Book. In the US, Article 1, Section 2 of the Constitution empowers Congress to carry out a census. Data obtained from the census are protected under Title 13 of the US Code, a provision of which calls for census information to be kept private for 72 years. It’s particularly interesting to study how the questions asked by that census have changed over the years. In 1790, we counted the number of free white males as well as the number of slaves in a given household. In the census of 1890, questions included, “Is the person defective of mind, sight, hearing, or speech?” and “Can the person read or write?” In 1990, the questions reflected a very different culture: “What time did this person usually leave home to go to work?” and “What was this person’s total income?” For 2000, partly in reaction to the growth of the immigrant population, we asked, “Is this person a citizen of the United States?” And in 2010, we asked even more directly, “Is Person 1 of Hispanic, Latino, or Spanish origin?”