Palestinians' Psychological Conditions Survey 2022
West Bank and Gaza, 2022
Get Microdata
Reference ID
WBG_2022_PPCS_v01_M
Producer(s)
World Bank
Collection(s)
Metadata
Created on
Apr 03, 2023
Last modified
Apr 03, 2023
Page views
6216
Downloads
190
Identification
Survey ID Number
WBG_2022_PPCS_v01_M
Title
Palestinians' Psychological Conditions Survey 2022
Country/Economy
| Name | Country code |
|---|---|
| West Bank and Gaza | WBG |
Study type
Other Household Survey [hh/oth]
Unit of Analysis
Individual.
Version
Version Description
Edited, anonymous dataset for public distribution.
Coverage
Geographic Coverage
West Bank and Gaza.
Producers and sponsors
Primary investigators
| Name |
|---|
| World Bank |
Sampling
Sampling Procedure
The planned sample size for PPCS was set to 7,057 households, and the 8,709 respondent households of RAPS 1 served as a sampling frame. The selection of PPCS households followed a one-stage cluster sampling design. Specifically, 641 Enumeration Areas (EA) were randomly selected from the 1,824 EAs of RAPS 1 with Probability Proportional to Size (PPS). All the RAPS 1 households contained in the selected EAs were included in the PPCS sample. To meet logistical needs of the data collection, variable X = “number of RAPS 1 respondent households per EA” was used as measure of size (MOS) for the PPS algorithm. This made the inclusion of EAs containing fewer RAPS 1 households less likely, thus avoiding logistical challenges and associated high data collection costs, but also fully preserving the probability sampling nature of the PPCS survey.
The individual questionnaire of the PPCS survey was administered to one selected adult member (aged 18 years or above) of each respondent household. For each respondent PPCS household, interviewers attempted to identify and re-interview the same adult individual who responded to the individual module of SEFSEC 2018. Only if the attempt was unsuccessful, the interviewer used a Kish grid (which accompanied the questionnaire) to randomly select, with equal probability, one adult from among all adult members of the household.
The individual questionnaire of the PPCS survey was administered to one selected adult member (aged 18 years or above) of each respondent household. For each respondent PPCS household, interviewers attempted to identify and re-interview the same adult individual who responded to the individual module of SEFSEC 2018. Only if the attempt was unsuccessful, the interviewer used a Kish grid (which accompanied the questionnaire) to randomly select, with equal probability, one adult from among all adult members of the household.
Response Rate
Of 7,057 planned households, 917 did not respond, yielding an overall household nonresponse rate of 13 percent. Of those who did not respond, in 84 percent of cases the household was not found or had moved abroad, in 12 percent of cases the household refused to be interviewed again, and the remaining households only partially completed the interview.
Of 6,138 adults, only 261 did not respond, yielding an overall nonresponse rate of 4 percent. We adjusted sampling weights for this nonresponse, as described in the next subsection. The re-interview attempt was mostly successful: among PPCS adult respondents, 91 percent had already responded to the individual module of SEFSEC 2018.
Of 6,138 adults, only 261 did not respond, yielding an overall nonresponse rate of 4 percent. We adjusted sampling weights for this nonresponse, as described in the next subsection. The re-interview attempt was mostly successful: among PPCS adult respondents, 91 percent had already responded to the individual module of SEFSEC 2018.
Weighting
The fundamental objectives of the weights calculation procedure were (i) mitigation of bias risks, and (ii) improvement of estimation efficiency. The main procedural steps taken were:
Step 1: Derive initial weights for PPCS households and roster individuals. This is obtained by multiplying the final household weights of RAPS 1 by the reciprocals of the inclusion probabilities generated by the PPS sampling algorithm.
Step 2: Adjust the weights of PPCS households and roster individuals for household-level nonresponse. Given the origin of the PPCS sample—that is, its provenance from the RAPS and SEFSEC panels—rich information was available on both respondent and non-respondent households to enable a propensity modeling approach to nonresponse adjustment.
Step 3: Calibrate the weights obtained at step 2, using as calibration benchmarks suitable household level and individual-level aggregates provided by PCBS as a result of a demographic estimation exercise (111 calibration constraints). Note that this step generated integrated household-level and individual level calibration weights. As a result, calibrated individual weights are constant within each household and are equal to the calibrated weight of the household.
Step 4: Suitably trim the calibration weights obtained at step 3 from unduly large weights that might lead to unstable estimates and inflated standard errors. In particular, we trimmed any calibration weights larger than the maximum nonresponse adjusted weight. Moreover, the trimming adjustment was performed consistently, by simultaneously preserving all the calibration constraints enforced in step 3 and not jeopardizing the individual-household integration property of the weights.
Step 5: Derive initial weights for the sample of adult individuals (one per household) who responded to the individual module of the questionnaire, based on the composition of each respondent household.
Step 6: Calibrate the weights obtained at step 5, using as calibration benchmarks suitable household level and individual-level aggregates provided by PCBS as a result of a demographic estimation exercise referred to the adult population (95 calibration constraints).
Step 1: Derive initial weights for PPCS households and roster individuals. This is obtained by multiplying the final household weights of RAPS 1 by the reciprocals of the inclusion probabilities generated by the PPS sampling algorithm.
Step 2: Adjust the weights of PPCS households and roster individuals for household-level nonresponse. Given the origin of the PPCS sample—that is, its provenance from the RAPS and SEFSEC panels—rich information was available on both respondent and non-respondent households to enable a propensity modeling approach to nonresponse adjustment.
Step 3: Calibrate the weights obtained at step 2, using as calibration benchmarks suitable household level and individual-level aggregates provided by PCBS as a result of a demographic estimation exercise (111 calibration constraints). Note that this step generated integrated household-level and individual level calibration weights. As a result, calibrated individual weights are constant within each household and are equal to the calibrated weight of the household.
Step 4: Suitably trim the calibration weights obtained at step 3 from unduly large weights that might lead to unstable estimates and inflated standard errors. In particular, we trimmed any calibration weights larger than the maximum nonresponse adjusted weight. Moreover, the trimming adjustment was performed consistently, by simultaneously preserving all the calibration constraints enforced in step 3 and not jeopardizing the individual-household integration property of the weights.
Step 5: Derive initial weights for the sample of adult individuals (one per household) who responded to the individual module of the questionnaire, based on the composition of each respondent household.
Step 6: Calibrate the weights obtained at step 5, using as calibration benchmarks suitable household level and individual-level aggregates provided by PCBS as a result of a demographic estimation exercise referred to the adult population (95 calibration constraints).
Data Collection
Dates of Data Collection
| Start | End |
|---|---|
| 2022-03-07 | 2022-05-09 |
Data Collection Mode
Computer Assisted Personal Interview [capi]
Data Collection Notes
Data collection started on March 7th and ended on May 9th of 2022, with approximately three interviews completed per interviewer per day. PCBS implemented strict quality control procedures to ensure the quality of the data as well as satisfactory effort in tracking households and individuals from the SEFSEC 2018. For example:
Interviewers could not start the questionnaire until they were in the vicinity of the respondent’s dwelling, as verified through GPS. This ensured that enumerators were interviewing the right household and not entering false information without consulting the respondent.
Inconsistent responses (since the last interview or within the interview) were flagged and the interviewer was asked to clarify any inconsistencies. For example, if the refugee status recorded in previous rounds was different to the response provided in the interview, the interviewer was asked to double check this answer.
Supervisors conducted random visits and a call-back survey with 5 percent of the interviewed households. The supervisory visits were particularly important during the first few weeks of data collection, as they were able to answer any questions the interviewers might have and suggest improvements to the interviewing technique.
Interviewers could not start the questionnaire until they were in the vicinity of the respondent’s dwelling, as verified through GPS. This ensured that enumerators were interviewing the right household and not entering false information without consulting the respondent.
Inconsistent responses (since the last interview or within the interview) were flagged and the interviewer was asked to clarify any inconsistencies. For example, if the refugee status recorded in previous rounds was different to the response provided in the interview, the interviewer was asked to double check this answer.
Supervisors conducted random visits and a call-back survey with 5 percent of the interviewed households. The supervisory visits were particularly important during the first few weeks of data collection, as they were able to answer any questions the interviewers might have and suggest improvements to the interviewing technique.
Access policy
Contacts
| Name | Affiliation | |
|---|---|---|
| Alia Aghajanian | World Bank | aaghajanian@worldbank.org |
| Arden Finn | World Bank | afinn1@worldbank.org |
Confidentiality
Citation requirements
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
Example:
World Bank. West Bank and Gaza - Palestinians' Psychological Conditions Survey 2022 (PPCS 2022). Ref: WBG_2022_PPCS_v01_M. Downloaded from [uri] on [date].
- 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
Example:
World Bank. West Bank and Gaza - Palestinians' Psychological Conditions Survey 2022 (PPCS 2022). Ref: WBG_2022_PPCS_v01_M. Downloaded from [uri] on [date].
Access authority
| Name | Affiliation |
|---|---|
| Alia Aghajanian | World Bank |
| Arden Finn | World Bank |
Disclaimer and copyrights
Disclaimer
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.
Metadata production
DDI Document ID
DDI_WBG_2022_PPCS_v01_M_WB
Producers
| Name | Abbreviation | Affiliation | Role |
|---|---|---|---|
| Development Data Group | DECDG | World Bank | Documentation of the study |
Date of Metadata Production
2023-04-03
DDI Document version
Version 01 (2023-04-03)