> ## Documentation Index
> Fetch the complete documentation index at: https://methodology.china-index.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Data collection

> China Index methodology, 2021 edition, as published.

## Regional partner selection

In this pilot stage, we will select 7 regional partners responsible for managing the research and coordinating with local experts to fill out the questionnaire. Each partner is expected to conduct this research in 5 countries.

## Eligibility requirements for regional partners and local experts

* They should be academics or professionals with expertise in certain domains, independent journalists or nonpartisan researchers, think tanks or civil society organizations, and community leaders.
* They must be nonpartisan in their own countries and in the countries surveyed by them.
* Their employees or members must not be on the U.S. Department of Treasury’s list of the Specially Designated Nationals and Blocked Persons ([http://sdnsearch.ofac.treas.gov](http://sdnsearch.ofac.treas.gov)).
* They must not have directly or indirectly received funding, donations, any form of payments, or interests from the government of the People’s Republic of China or the apparatus of the Chinese Communist Party (CCP).

## Selection criteria for regional partners

* Expertise in China studies, such as Chinese politics, economics, cultural influence, or other relevant fields.
* At least one year of work or research experience in China-related affairs.
* Experience in coordinating transnational project(s), especially experience in collaborating with civil society from several countries (proven skills in networking, administration, and accounting preferred).
* Expertise or appropriate networks in their regions.

## How to answer the questions?

Each of the indicators will include 4 fields, all of which must be completed by respondents:

* **Influence measure:** The response to the indicator question. For binary questions, YES indicates observed PRC influence; NO indicates no observed influence. For graded questions, a score of 0 indicates no observed influence and a score of 4 indicates high observed influence.
* **Supporting Evidence:** References supporting the chosen response. The evidence provided must be appropriate and sound. Accepted forms of evidence will be limited to a functional URL that directs to news accounts, official documents, think tank reports, NGO publications, academic papers, interviews, and books published within 4 years. Social media posts or other sources may be accepted if deemed appropriate but are generally discouraged.
* **Confidence rating:** A percentage score that indicates how confident the respondent is in their response.
* **Supplementary Notes (Optional):** Any additional information that the respondent may wish to provide for this question. The content of these notes may include but is not limited to:
  * **Elaboration:** Detailed information on the rating or supporting evidence if necessary.
  * **Limitations:** Reasons for any limitations encountered in providing evidence or securing high quality evidence.
  * **Local contexts:** a) the local current situation is far more extreme than the rating suggested, b) different interpretations regarding the option usually implied, or c) other local factors that may elevate or suppress the impact of the indicator.
  * **Evidence to the contrary:** If a respondent dissents with their country’s government and believes they’ve been less than sincere in providing accurate information (in the extreme case, being the source of disinformation per se) to reflect China’s influence, the respondent should elaborate on the situation and explain how the government has been unreliable or insincere.
  * **Feedback:** Reflections on the design of this question.

## Reducing the Impact of Bias

* Regional partners are responsible for collecting data by consulting with local experts. Questions should be answered by consulting at least one local expert. There is no limitation for the number of questions for which an expert could be consulted, as long as they have qualified expertise.
* Questions that the regional partners consider irrelevant or can either answer “yes” or “no” may be skipped. Respondents will be requested to skip no more than 10 questions (10%) in one questionnaire.
* Regional partners should find as many qualified and trustworthy local experts within a country as possible. Considering this might be a challenge in some countries, we set the minimum limit for a total of one expert for one country.
* **Access to local experts to fill out the questionnaire:** selected local experts may or may not be affiliated with the corresponding regional partner. If they are affiliated with the regional partner, these experts must be excluded from the administration of this project and the data collection and verification process, including the weighting of domains, in order to avoid role conflicts.
* Situations that regional partners may reach out to a secondary expert to answer the questionnaire: (a) If the confidence of answers from the first respondent are low; (b) if the results from the first respondent contain too many unfilled questions.
* Local experts may choose not to reveal their identity or association with this project for security reasons.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.