Largest-ever Survey on Aadhaar Provides Data-Driven Insights on Key Aspects of Exclusion, Privacy and Data Quality

May 16, 2018

Three-state survey highlights differences in state capacity; identifies specific areas to address to tackle data quality and exclusion issues

 New Delhi, India — 17 May 2018 — IDinsight, a global development analytics firm working across India, today released the State of Aadhaar Report 2017-18. The report is based on the largest household survey since Aadhaar’s inception, covering 2947 rural households in 21 districts across Andhra Pradesh, Rajasthan, and West Bengal. The survey was conducted between November 2017 and February 2018. “The goal of the State of Aadhaar Report is to catalyse the Aadhaar debate and policymaking to be more data-driven,” said Ronald Abraham, Partner, IDinsight.

The key highlights of the report are:

  • Aadhaar’s coverage is widespread, but the data quality has room for improvement. The report finds higher coverage of Aadhaar than voter IDs. In addition, the report finds no evidence of differences in enrolment by gender, caste, religion, or education levels. The report highlights that 8.8% of Aadhaar-holders reported errors on their name, age, address, or other information on their Aadhaar letter. Compared to voter IDs, the error-rate on Aadhaar was 1.5 times higher.
  • While exclusion from food ration (PDS) due to Aadhaar-related factors is significant, it is lower than exclusion explained by factors unrelated to AadhaarState capacity has a bearing on the functioning of PDS, with wide variation between Rajasthan and Andhra Pradesh. Overall monthly exclusion from PDS in Rajasthan is 9.9%, whereas it is 1.1% in AP. Of this, Aadhaar-related factors contribute 2.2% and 0.8% respectively. Despite this, the report finds that a majority of food ration (PDS) recipients prefer Aadhaar-based PDS delivery in both states, as they perceive biometric authentication prevents identity fraud.
  • Over 96% of respondents valued privacy and thought it is important to know what the government will do with their Aadhaar data. At the same time, 87% of respondents approve of mandatory linking of Aadhaar to programmes like PDS.
  • Aadhaar's analog version (the letter used as a paper ID) is much more widely used to open bank accounts than its digital version (e-KYC). 67% of bank account holders used their Aadhaar letter to open their most recent account, and 17% used e-KYC.

Abraham added, “Issues such as protecting privacy and eliminating exclusion are serious and require rigorous evidence to carefully unpack and address”.

Dr. Ajay Bhushan Pandey, CEO of UIDAI, said, "IDinsight's latest report highlights that Aadhaar has wide-scale support on the ground. PDS exclusion due to failure of local administration, though small, should be taken very seriously by the concerned agencies. They should ensure that not a single beneficiary is denied. The Aadhaar Act and Government instructions provide for alternate means of identification for genuine beneficiaries who encounter problems in authentication."

To enhance access and agency for individuals, the report points to the need to augment state capacity and providing a robust, credible manual override in case of biometric failure. Other policy recommendations of the report include a national campaign to update incorrect data, strengthening business correspondent networks, and developing robust laws and institutions for privacy protection.

CV Madhukar, Global Digital Identity Lead at Omidyar Network, which sponsored the report, said: “There has been much heated debate about Aadhaar in recent months, which is mainly binary in nature. We hope the findings of the State of Aadhaar Report provide the much-needed empirical base for a more informed and nuanced debate, and sharper insights for policy making on this very important issue.”

What are you afraid of Publish this comment anyways Initial set of questions at Not UIDAI or former UIDAI

1) What about claims of saving from PDS? It is well-known that, of the three reasons, Aadhaar can address only one. 2) It is well known that paper Aadhaar should not be used for e-KYC. This is in direct violation of what Aadhaar is supposed to be. No one verifies residential address when Aadhaar is created - in fact, if the claim is that there is a lot of fake identities floating around, given a majority of Aadhaar numbers (not cards! It isn’t a card) were issued on the basis of an existing identity, what’s the point in using Aadhaar for e-KYC? 3) How many of those who were soooo in favor of mandatory Aadhaar linkage are aware of State Resident Data Hubs? How many are aware of data leaks that have occurred to date? Finally, in the interest of transparency - since this affects an issue that is currently in the Supreme Court and has repurcissions for the privacy and rights of over a billion people, share the questionnaire used with respondents as well as sample frame. Also declare conflicts of interest - doesn’t IDInsight have people who once worked with UIDAI? Aren’t there people who have written publications arguing in favor of Aadhaar?

S Raj Brava! A partial job at Panopticon

Your headline is a mis-characterization. "Largest-ever Survey on Aadhaar Provides Data-Driven Insights on Key Aspects of Exclusion, Privacy and Data Quality". In general, you have looked at a partial list (a very small set) of issues that have been raised by well-informed legal, development, privacy and technology experts, and yet have Tweeted that "#StateofAadhaar Report based on largest survey on #Aadhaar reveals that Aadhaar can’t be given a simple “good” or “bad” verdict.". Seriously?! You want empirical debate on Aadhaar? 'Unpack' one issue after another, in depth. Every. Single. One. Specifically, on your headline and each of these three aspects are treated: 1. Exclusion: does your report capture the cost of exclusion or repeat visits needed when authentication doesn't occur? Did your questions differentiate between the times that connectivity was available and authentication failed; and connectivity was unavailable (and later authentication was fine on first attempt once connectivity is established)? How did you differentiate between the times one household member was able to authenticate (single or multiple tries), and when more than one member had to (or have started to show up, anticipating problems) at the PDS shop? When one speaks about the poor - who have to travel a certain distance or take a day off work (and risk losing a job) - in order to get PDS or any other service, every type of failure has significant costs. And in a country like India, %s adds up. 2. Privacy: the question on privacy in your report was limited to Aadhaar-linked information (name, biometrics, age etc.). There is a broader issue - of how public as well as private sector links Aadhaar number to other types of data, including in ways not permissible (caste, religion data - as was transparent in the AP data leaks). How many of those you interviewed were aware of (a) SRDHs; (b) multiple points of storage of data; (c) leaks of not just Aadhaar-linked data, but also other details link bank account numbers (how will people know, when UIDAI refuses to acknowledge leaks and there is no legal obligation to inform the resident. Not citizen. Resident. Try THAT in one of the OECD countries! Especially under GDPR) ; (d) instances where Aadhaar-number and PIN combination has been misused? For (c) and (d), these are things that can be fixed once billions have enrolled. There were plenty of opportunities since the late 2000s when these issues were obvious or should have been anticipated. I could go on. 3. Data quality: the only aspect you can speak to is what is on the Aadhaar card. Not the quality of biometric data - authentication of which is probabilistic; not whether the current information with UIDAI is accurate. Why? 1000s of operators who were disqualified by UIDAI later but who may still have access and, as journalists have shown, or others who can buy access can/have change/changed the data. The Caravan came out with an excellent piece on Aadhaar and UIDAI in May 2018. Even THAT is a partial list of issues associated with Aadhaar. I am so very pleased that UIDAI and Mr Pandey's communication machinery can now tout this report - much like the other management consultancy influenced number in a World Bank report - as a confirmation of success. At any point, did you ask yourselves the question, we could produce such a report, but should we? Especially, knowing this is a partial list of issues, that UIDAI tends to any distort any information to look favorable, and in the midst of a trial in India. Much like UIDAI never asked itself, we could push this project through, but should we?

S Raj Karnataka at Govt

Here are some numbers from the government's own reports for you folks. You should ask Rajasthan and Andhra Pradesh to release the equivalent numbers. 20%. 1 in 5 families unable to authenticate. Data released by the Food and Civil Services Department of the Karnataka government too revealed that the system is not functioning smoothly. Of the 1.2 crore ration cards linked with the biometrics database in Karnataka, nearly 99 lakh cards recorded transactions in February 2018. Of these, fingerprint authentication did not work for over 18.6 lakh cards, almost one in five families in the state.

Be the first to comment
It looks like there's some information missing
By clicking, you agree to the Terms and Conditions


Investigative reporting co-operative, The Ferret, has secured a $100,000 grant to support its award-winning work

The two-year grant comes from Omidyar Network, a philanthropic investment firm, and will enable The Ferret to expand its journalistic capacity, increase membership and impact, and strengthen its ability to hold power to account across Scotland, the UK and beyond.



Independent Research Commission Partnering with Facebook and 7 Nonprofit Foundations to Study Role of Social Media in Elections and Democracy Reveals New Name and Announces First Data Set is Available for Academic Research

The independent research commission that was announced in April to use Facebook data to analyze the role of social media on elections and democracy revealed today the organization's name — Social Science One. The new organization also provided details about how it will work, and announced that academics can now apply for access and funding to study the first Facebook data set.



Moving from Financial Access to Health

In a new World Bank article, Tilman Ehrbeck shares insights on the change in focus from financial access to financial health.