READ Me file

# Title: National Survey on the Effects of COVID-19 on the Wellbeing of Mexican Households (ENCOVID-19- OCTOBER)

# Título: Encuesta Nacional sobre los Efectos del COVID-19 en el Bienestar de los Hogares Mexicanos (ENCOVID-19- OCTOBER)

# How to cite
PŽrez Hern‡ndez, V’ctor Hugo; Teruel Belismelis, Graciela; Gaitan-Rossi, Pablo; Vilar-Compte, Mireya; Hern‡ndez Solano, Alan Mart’n; Triano Enr’quez, Manuel; L—pez Escobar, Emilio. (2020) Encuesta Nacional sobre los Efectos del COVID-19 en el Bienestar de los Hogares Mexicanos (ENCOVID-19 - OCTOBER) [Data set]. EQUIDE-UIA. 

# ENCOVID-19 project
Amid the COVID-19 outbreak, the ENCOVID-19 provides information on the well-being of Mexican households in four main domains: labor, income, mental health, and food insecurity. It offers timely information to understand the social consequences of the pandemic and the lockdown measures. It is a project consisting of a series of cross-sectional telephone surveys collected once a month for one year. In addition to the four main domains and a set of COVID19-related questions, the survey includes new key indicators every month to capture the impact of the pandemic on issues like education, social programs, and crime. This is the fifth dataset of the project, corresponding to October 2020, collected seven months after the lockdown began in Mexico. Data collection was performed between October 8 and October 22. 

# ENCOVID-19 project resources
Results and additional resources regarding the project are uploaded here: https://equide.org/pobreza/https-equide-org-pobreza-impactos-del-covid-19-en-mexico/

# Language
The survey was conducted in Spanish and documentation and dataset are in Spanish.

#Questionnaire
The questionnaire is available in Spanish in the file "Cuestionario Octubre 2020 Nacional.pdf".

# Dataset format
The same dataset is presented in two formats. The .dta dataset is fully labelled to facilitate the identification of variables. The .csv dataset eases the preservation in the repository and amplifies the number of statistical packages able to read it.

# Methodology (text in Spanish in the Metodologia.pdf)

Target population: persons 18 years or more who have a mobile phone. 

Sample frame: Randomly selected mobile telephone numbers taken from the publicly available Plan Nacional de Numeraci—n (National Numbering Plan) from the Instituto Federal de Telecomunicaciones (Mexico Federal Institute of Telecommunications, IFT), updated on October 1, 2020. As of the third quarter of 2019, mobile phone coverage in Mexico was 96%. National surveys suggest that about 90% of all Mexican households have at least one mobile phone user (94% in cities and 74% in rural areas).

Data Collection Technique: Telephone interviews based on a structured questionnaire using Computer Assisted Telephone Interviewing (CATI). Interviewers were previously trained and randomly supervised by the PI team to assess the quality of the interviews. Everyone worked from their own houses without risk of contagion.  The interview manual is included below the questionnaire. 

Sample design: One-stage probabilistic sample of mobile telephone numbers stratified at the state-level using Random Digit Dialing (RDD). The number selection was performed from the national numbering plan for each stratum using using the R package RDDQuantosIC v1.5 (developed by Quantos).

Stratification. The sample was stratified for the 32 states in Mexico.

Sampling weights: To avoid minor deviations in the demographic and socioeconomic characteristics of our sample, simple sampling weights were estimated based on Narain (1951) and Horvitz-Thompson (1952). Sampling weights adjust for differences in geographic location (state), sex, age, and socioeconomic status’ distributions as observed in census-like data of 2015 from INEGI (Encuesta Intercensal 2015, INEGI). We use SES categories as defined by the AMAI (Mexican Market Research Association). We provide appropriate sampling weights at both the individual and household levels, which may be used at discretion of the user according to the relevant unit of analysis of each question. The suffix “_n” indicates the normalized weights (ie, such that sum to the sample size).

# Institutional Review Board (IRB): The project was approved by the Ethics Committee from Universidad Iberoamericana with the registry: CONBIOÉTICA-09—CEI-008-2016060

# Results 
Every variable is labelled in the stata data set file (.dta) according to the items in the questionnaire. Frequency tables with percentages for every variable are provided in the file Descripcion BD.pdf. Variable names can be linked to the questionnaire using this file. 

#Contact 
If international researchers require additional information, please contact Victor Perez (victor.hernandez@ibero.mx) or Pablo Gaitan-Rossi (pablo.gaitan@ibero.mx).