The World Bank is providing technical and financial support to countries to help mitigate the spread and impact of the coronavirus disease (COVID-19). One area of support is for data collection to inform evidence-based policies that may help mitigate the effects of this crisis. Towards this end, a phone survey of 4 rounds is expected to be implemented in Djibouti. The first round of data was collected in July 2020 by the National Institute of Statistics of Djibouti. The second round was collected between September and October 2020. The third round of data was collected in December 2020 and January 2021 by the National Institute of Statistics of Djibouti.
To understand the socio-economic impact of COVID-19 and associated government measures over the long term, the third round of the COVID-19 National Panel Phone Survey 2020 was collected by the National Institute of Statistics of Djibouti (INSD) between December 20, 2020 and February 2, 2021. Various channels of impact are explored such as job loss, availability and price changes of basic food items, ability to access healthcare, and food insecurity. New sections on the attitudes toward a potential vaccine and shock coping strategies have also been added (compared to the second round).
Note that a sample of 564 refugee households living in Djibouti has been collected during the same time frame and using the same questionnaire. This data set will be available for download separately on the microdata library.
Kind of Data
Sample survey data [ssd]
Unit of Analysis
Version 01: Edited, anonymized dataset for public distribution.
The COVID-19 National Panel Phone Survey 2020 Djibouti wave 3 covered the following topics:
- Household Roster
- Household's Income sources
- Access to Basic Goods
- Access to Healthcare
- Safety Nets
- Food Insecurity
- Shock Coping Strategies
- Vaccine Attitudes
Urban areas only.
The survey is representative of the bottom 80 percent of the consumption distribution of the national households (thus the top 20 percent are excluded). It is representative by poverty status and by three domains of Balbala, rest of Djibouti city and urban areas outside Djibouti city.
The survey covers national households that reported telephone numbers, are included in the social registry data collected by the Ministry of Social Affairs and Solidarity (MASS) and have been interviewed after 2017.
Producers and sponsors
Poverty and Equity GP
Institut National de la Statistique de Djibouti
Implementation partner and collaborated in survey design and analysis
The World Bank
Ministry of Social Affairs and Solidarity, Djibouti
Sharing the social registry data with INSD to draw a sample
As a recently conducted representative household survey with telephone numbers was not available, data from the national social registry collected by the Ministry of Social Affairs (MASS) was used as the sampling frame of the national sample. The social registry is an official database of households in Djibouti that may benefit from public transfers and be particular targets of poverty alleviation efforts. The sample consists of households drawn randomly from the social registry data restricted to urban households having at least one phone number and interviewed after July 1, 2017. The sample design is a one-stage probability sample selected from the sampling frame and stratified along two dimensions: the survey domain (three categories) and the poverty status (binary). This yields six independent strata. Within each stratum, households are selected with the same ex-ante probability, but this differs across strata. With a non-response rate averaging 26 percent for the national households, the third wave consisted of 1,383 interviewed national households with complete information that were representative of the urban national population, out of which 990 households were also interviewed in the two first waves, 190 were added as replacement households in the second wave and re-interviewed in the third one, and 203 were added as replacement households in the third wave.
The response rate among the national sample was about 74.3 percent with 1,383 interviewed national households. Slight differences were observed across location, with districts 1, 2 and 3 of Djibouti city more likely to respond than other locations (response rate at 76.4 percent versus 75.9 and 70.5 percent, respectively in Balbala and the other urban areas).
For the national sample, both cross-sectional and panel weights are designed to adjust for differences in selection probability due to either design or non-response. In addition, further adjustments in sampling weights were made to ensure that indicators produced are representative of the country’s population, by poverty status and by location, and of the refugee population present on the three main refugee villages and Djibouti City. The sampling frame of the Djibouti nationals, the social registry of the Ministry of Social Affairs, over-represents the poor and has an incomplete coverage of the upper distribution of income. To correct for these biases, we rely on a post-calibration approach, using the household budget survey of 2017 (EDAM 2017) as the reference data source. This is because EDAM 2017 survey was representative of the country’s population by poverty status and survey domains. However, EDAM 2017 survey is restricted to the first four consumption quintiles to ensure sufficient overlap of the universes covered by both surveys.
Dates of Data Collection
Data Collection Mode
Computer Assisted Telephone Interview [cati]
Data Collection Notes
1. Organization of the fieldwork: The survey team was composed of 27 surveyors and 4 supervisors. Each enumerator was given a tablet and mobile phone (including sim card and data bundles) to be used for the interviews. The questionnaire was implemented using CSPro's CATI capabilities. Data were collected by trained INSD interviewers who individually made phone calls from their respective homes. Data from completed and partially completed interviews were synchronized each evening.
2. Pre-loaded information: Basic information on each household (such as location, household head name, phone number, etc.) was pre-loaded in the CATI assignments for each interviewer. The list of household members and their basic characteristics were uploaded from the previous rounds and the social registry data. The aim of pre-loaded information is to assist interviewers in calling and identifying the household and ensure that each pre-loaded person is properly addressed and easily matched to the most recent interviews. Moreover, the names of the respondent and the breadwinner from the second round were uploaded to ensure an easier follow-up.
3. Respondents: The survey had one respondent per household, who was the knowledgeable adult household member or the head of the household. The respondent must be a member of the household and must be an adult. The respondent may still consult with other household members as needed to respond to the questions. Whenever possible, the respondent from the second round is preferred to answer questions of the third round.
Institut National de la Statistique de Djibouti
The questionnaire of the third round is adapted from the questionnaire of the second round and in accordance with the template questionnaire prepared by the Poverty and Equity GP to measure the impact of COVID-19 on household welfare. It was designed in French and dispensed in local language (Afar, Arabic, Somali, French or other). The questionnaire includes the following sections:
- Household roster
- Household's income sources
- Access to services
- Safety nets
- Food insecurity
- Shock coping strategies
- Vaccine attitudes
The CSPro CATI data entry application helped to enforce skip and range patterns during data collection. Standard consistency checks (like age differences between parents and children and unicity of household heads) were carried out at the time of the data collection. Because the entry application was strictly system-controlled, complete cases including missing items were avoided. The various checks resulted in a limited need for secondary data editing, which eventually entailed two main steps from the WB team. First, duplicated names of household members, who were otherwise distinct, were corrected by adding a suffix “bis” to the names. Second, after analysis of text responses mentioned in the residual “other” categories, a few items codes were adjusted (not exceeding 10 in any category).
The dataset has been anonymized and is available as a Public Use Dataset. Before being granted access to the dataset, all users have to formally agree:
1. To make no copies of any files or portions of files to which s/he is granted access except those authorized by the data depositor.
2. Not to use any technique in an attempt to learn the identity of any person, establishment, or sampling unit not identified on public use data files.
3. To hold in strictest confidence the identification of any establishment or individual that may be inadvertently revealed in any documents or discussion, or analysis. Such inadvertent identification revealed in her/his analysis will be immediately brought to the attention of the data depositor.
Use of the dataset must be acknowledged using a citation which would include:
- the Identification of the Primary Investigator
- the title of the survey (including country, acronym and year of implementation)
- the survey reference number
- the source and date of download World Bank.
COVID-19 National Panel Phone Survey 2020 Djibouti Wave 3. Dataset downloaded from www.microdata.worldbank.org on [date].
Disclaimer and copyrights
The user of the data acknowledges that the original collector of the data, the authorized distributor of the data, and the relevant funding agency bear no responsibility for use of the data or for interpretations or inferences based upon such uses.