ODIN assessments review published statistics in 22 topical categories, grouped as social statistics, economic and financial statistics, and environmental statistics. The default ODIN overall score weighs the three groups equally. More information about the indicators in each data category and how data categories are scored can be found on the category pages in the Methodology Guide.

The data categories in ODIN 2024/25 are:

Coverage of data is measured by five elements that measure how complete the country’s data offerings are. For more information on these, visit our Methodology Guide.

The five coverage elements are:

1. Availability of indicators and disaggregations

   How many indicators are available and disaggregated? Requirements vary by category.

2. Availability of data in the last five years

   How many years are available for an indicator between 2019-2023? 3 years are required for full credit.

3. Availability of data in the last ten years

   How many years are available for an indicator between 2014-2023? 6 years are required for full credit.

4. Availability of data at the first subnational level

   How many indicators have data at the first subnational level (not required for many economic and environmental indicators)?

5. Availability of data at the second subnational level

   How many indicators have data at the second subnational level (not required for many economic and environmental indicators, as well as small countries)?

Openness of data is measured by five elements that assess how well a country’s data offerings meet international standards of openness, as defined by the Open Definition. For more information on these, visit our Methodology Guide.

The five openness elements are:

1. Availability of data in machine readable format

   Are all data published in at least one machine-readable format? Such as XLSX and CSV.

2. Availability of data in non-proprietary format

   Are all data published in at least one non-proprietary format? Such as XLSX, CSV, and PDF.

3. Availability of reference metadata

   Do all datasets have complete reference metadata? Which includes the responsible agency for data collection, the date of upload, and a definition of the indicator.

4. Availability of download options that make the data more accessible: bulk downloads, APIs, customizable data export options

   Are all datasets available via bulk download and an API or customizable export?

5. Availability of an open data license or open data terms of use

   Are all datasets published under an open data license or terms of use?

Yes, in ODIN countries with a surface area of 1,500 sq. km. or less are not scored in many categories for subnational data. In ODIN 2024/25, this includes the countries: Andorra, Anguilla, Antigua and Barbuda, Aruba, Bahrain, Barbados, Cook Islands, Dominica, Grenada, Hong Kong, Kiribati, Liechtenstein, Macao, Maldives, Malta, Marshall Islands, Micronesia, Monaco, Nauru, Palau, Sao Tome and Principe, San Marino, Seychelles, Singapore, St. Kitts and Nevis, St. Lucia, St. Vincent and the Grenadines, Tonga, and Tuvalu.

None of these countries receive scores in any category for coverage element 5- data availability at the second subnational level. Also, they are not scored for data disaggregation at the first subnational level for the following categories: Food Security and Nutrition, National Accounts, Price Indexes, Government Finance, Money and Banking, International Trade, Balance of Payments, Energy and Pollution. Some individual indicators in categories that otherwise receive scores are also not considered in scoring (Indicators 2.3, 4.3, 6.1, 6.2, 8.2, 9.3, 10.2, and 18.2). See more about these indicators in this section of the Methodology Guide.

The ODIN 2024/25 assessments were carried out between June and October 2024. Only data published before October 15 2024 are considered. ODIN 2024/25 assesses datasets with data points from the 2014-2023.

In ODIN 2024/25, there have been the following minor methodology changes:

Expanded the list of sources used for data collection: The purpose of ODIN has always been to measure the availability and openness of official statistics produced by the national statistical system (NSS), not only the national statistics office (NSO). To better serve the needs of SDG monitoring and to further expand country coverage, data will now be collected from SDG data portals and line ministry government websites even if they are not linked from the NSO website.

Broadened indicator names and categories: The structure of how indicators are presented in ODIN has been slightly modified. We have recategorized the presentation of indicators as “topics” instead of single-standing indicators. This allows for more flexibility in the types of datasets that could be collected for ODIN for each indicator.

Piloting new indicators on Digital Connectivity: Connectivity is critical for accessing health information, using financial services, or networking with peers, and can be used to gain further skills to advance education and work opportunities. This is why for ODIN 2024/25, data on internet use and mobile network access were collected and assessed. For these two indicators the data collected was not scored for this round, but they will be scored moving forward.

Scoring data for the first and second subnational level: Data availability at the first and second subnational levels have been the lowest scoring coverage elements across most countries. Previous ODIN editions have required that countries publish every indicator and disaggregation with subnational level data for full credit. Now, this criterion has been lowered to only require subnational level data for all indicators and at least one disaggregation for full credit. For the vast majority of countries, this change will be insignificant, but it will better recognize the efforts of countries who are already publishing much data at this level.

Open Data Watch invites all countries’ national statistics offices (NSOs) to provide feedback on the datasets recorded in ODIN. This feedback can include information on datasets they believe should be considered. Each NSO is contacted at least 3 times between March and May 2024 by email. If NSOs agree to participate, they are provided a spreadsheet identifying the datasets evaluated by the ODIN team and given a month to provide any comments. Their feedback is reviewed and incorporated into the final assessment, if in line with ODIN methodology. If you are an NSO that would like to participate in the review process, please email us ODIN-input@opendatawatch.com.

ODIN assessments reflect the state of open data at the time the assessment was completed. Since then, many countries may have uploaded new data, removed datasets, or even restructured their website or data portal. Because of this, it can be very difficult to confirm whether datasets were available at the time of the assessment when presented at a later date.

Nonetheless, if you believe we made an error, please email ODIN-input@opendatawatch.com. The most common reason for a dataset to not be considered is because it does not contain one of the many required disaggregations. Unless an error is egregious, typically changes are not made to scores after they have been published. Countries’ opportunity to comment on their results is limited to the review period.