Cox’s Bazar Panel Survey

The Cox’s Bazar Panel survey is a comprehensive dataset representative of the displaced Rohingya and host communities.

The Cox’s Bazar Panel Survey (CBPS) is a comprehensive panel dataset tracking social, economic, and health outcomes from a representative sample of recently arrived Rohingya (displaced after August 2017), and Bangladeshi households residing in close proximity to the refugee camps in Cox’s Bazar and Bandarban. The first wave of data was collected in partnership with Yale University researchers and researchers at World Bank and the Gender and Adolescence: Global Evidence project (GAGE). Data collection for the first wave of CBPS was conducted by Innovations for Poverty Action (IPA) Bangladesh between March and August 2019.

Data can be downloaded from the following links:
1. First round of in-person survey (2019):
2. Tracking Surveys (2020-2021)

The design of the Cox’s Bazar Panel survey is useful for practitioners and policymakers to understand the existing landscape, to monitor indicators and prioritize programming, and to design evaluations of future programs.

In order to carry out the first wave of the survey, IPA first conducted a census of 45,916 households, equally divided between Rohingya and host communities, and randomly selected 5,020 households to implement the in-depth survey.

This design allows for the data to be used for three objectives:

  1. The first wave of the survey of 5,020 households provides rich descriptive statistics that policymakers and practitioners can use in designing their programs and policies.
  2. Future waves of the survey of 5,020 households will provide data on how both Rohingya and host communities change over time, allowing practitioners to monitor indicators and prioritize their programming accordingly.
  3. The 40,896 households who participated in the census, but not in the full survey, may become participants in future impact evaluations in order to identify the effectiveness of specific programs. As IPA has already performed a census of this population, the cost to implementing an impact evaluation is marginal.

Sample and Objectives

45,916 households participated in census

5,020 households participated in survey

40,896 households participated in census, but not surveyed

Used to achieve objectives 1 and 2

Used to achieve objective 3


The CBPS sample is representative of three populations: i) camp residents, ii) residents of host communities within 15 kilometers of the camps, and iii) residents of host communities further than 15 kilometers from the camps. The camp sample frame is based on Round 12 of the IOM Needs and Population Monitoring data, and the host community sample is based on the 2011 National Census. The camp sample covers all camps, and the host community sample covers communities in the upazilas of Bandarban, Chakaria, Cox’s Bazar Sadar, Pekua, Ramu, Teknaf, and Ukhia.

Information Collected

The survey consists of two sections: (1) a household interview conducted with one adult household member and (2) an adult interview conducted with two randomly selected household members age 15 or above.
The household interview collected the following information:
  • Household Roster: A listing of all household members including socio-demographic information, education, and employment or activity status. A retrospective module captures the same information on household composition prior to displacement, i.e., as of July, 2017.
  • Housing Characteristics: Dwelling’s building materials, intra-dwelling population density, number of households sharing the same dwelling and households’ rental, and access to water, electricity and toilet facilities.
  • Food Security: General availability and variety of food, coping strategies for food scarcity, identifying intra-household inequalities in nutrition and resource allocation.
  • Consumption: Detailed item-by-item information on consumption of food and non-food items, with a seven-day recall period for food and a thirty-day recall for non-food, as well as access to markets.
  • Assistance: Types of assistance or aid received by the household, quantity of assistance bartered, and goods/services receiving in exchange.
  • Assets: Key assets owned, perceived resale value, ownership in July 2017, and status of items in Myanmar.
  • Anthropometrics: Height and weight for one randomly selected child under the age of 5.

The adult interview collected the following information:

  • Labor Market: Adult respondents’ employment activity in the past week and past year, including whether the respondent is a wage worker or self-employed and occupation, job search methods, monetary or in-kind compensation, and profits.
  • Labor Market History: A retrospective employment details for the period July 2016 to July 2017.
  • Migration History: Short-term (between two weeks and one month) and long-term migration history (more than one month), to identify variation in adult respondents’ area of settlement before and after July 2017.
  • Use of Health Services: Health care utilization and health status, commuting time to health care centers, medicine expenditures, and whether patients pay for their visit or go to public clinics.
  • Crime and Conflict: Exposure to crime and conflict issues in their neighborhood, including bribery, any type of harassment and family disputes, among others, as well as whether victims look for help, and if so, where they seek assistance.
  • Trauma and Mental Health: Depressive symptoms, mental health, traumatic experiences, and symptoms of trauma, using the Patient Health Questionnaire 9 (PHQ-9) and Harvard Trauma Questionnaire (HTQ).


The CBPS is funded by the Yale Macmillan Center; IPA’s Peace and Recovery Program; the Gender & Adolescence: Global Evidence (GAGE) program (the prior two funded by UK Aid); the Poverty and Equity Global Practice of the World Bank; and the State and Peacebuilding Fund (SPF) administered by the World Bank.
Researchers: Paula López-Peña (Yale MacMillan Center), Austin Davis (American University), A. Mushfiq Mobarak (Yale University), Sarah Baird (George Washington University), Jennifer Seager (George Washington Unviersity), Nethra Palaniswamy (World Bank), and Tara Vishwanath (World Bank).
Contact: A. Mushfiq Mobarak, Professor of Economics, Yale University,