Chapter 1 Purpose & How to use this Guide

This guide presents standardized tools for collection, processing, analysis and visualization of food security direct evidence indicators and technical variables used in the Cadre Harmonisé process.

1.0.1 A brief overview of the chapters in this guide:

Chapter 2 Setting up R & Rstudio how to install R & Rstudio software, how to add required packages and an overview on Rmarkdown (used to generate reports).

Chapter 3 Gecodes explains how/why to use geo-codes in data collection (xlsforms) to speed up mapping and data processing.

Chapter 4 Household Hunger Scale paper and electronic versions of standardized questionnaire and SPSS/R syntax for calculating the Household Hunger Scale indicator.

Chapter 5 reduced Coping Strategy Index paper and electronic versions of standardized questionnaire and SPSS/R syntax for calculating the reduced Coping Strategy Index indicator.

Chapter 6 Food Consumption Score & Household Dietary Diversity Score paper and electronic versions of standardized questionnaire and SPSS/R syntax for calculating the Food Consumption Score/Groups and Household Dietary Diversity Score indicators.

Chapter 7 Livelihood Coping Strategies paper and electronic versions of standardized questionnaire and SPSS/R syntax for calculating the maximum coping strategy used for CARI light version and ENA version of the Livelihood Coping Strategies module.

Chapter 8 Combined Questionnaire & Syntaxes for all 5 indicators combines the questionnaires and syntaxes from chapter 4 - 7.

Chapter 9 Data Quality Monitoring Report provides instructions and R syntax for a data quality monitoring report, using raw survey data.

Chapter 10 Calculating Matrice Intermediare provides instructions and a generic R syntax to produce from a processed spss file the direct food security indicators and geocodes which can be copied into the matrice intermediare.

Chapter 11 Visualizing CH Direct Food security Indicators provides instructions and generic R syntax for visualizing maps and tables of the phasing of CH direct evidence indicators from the matrice intermediare or a processed data set.

1.0.2 About software

While use of the software R & Rstudio is necessary to produce the outputs in chapters 9, 10 and 11, users who aren’t ready to fully replace SPPS with R for their survey processes do not need to disrupt their normal data processing steps (cleaning, calculating, indicators, adding weights, etc) - this can remain in SPSS. Syntax for calculating indicators in SPSS is provided in chapters 4 - 8 .

The R syntax used for the data quality monitoring report is compatible with the “raw data” that is downloaded from data collection servers (Kobo/MoDA etc.) and the R syntax used to produce the matrice intermediare is compatible with the “processed data” in spss format.

1.0.3 About the example datasets

For this guide, two “semi-fictitious” data sets were created by randomly selecting and de-identifying several surveys. The results are “safe” data sets that serve as examples for the syntax and processes illustrated in this guide:

exampledataEnglish_raw - is what a “raw” data set typically should look like when collecting with standardized modules and downloaded from a ODK server in spss format.

exampledataEnglish_processed - is what a “processed data” typically should look like when applying standardized syntax to calculate indicators and adding survey weights.

1.0.4 About the questionnaires and variable names

The standardized questionnaires were developed from official guidelines/references listed in the chapters and were adapted reflecting after consultation with several food security analysts. While many questions in this guide were taken verbatim from the official references, the wording of some question modules follow the opinion that instructing enumerators to repeat the content of a question such as recall period (i.e. in the last 7 days…) and qualifier (…because lack of food or lack of money to buy food) explicitly within the question text is better practice than having enumerators read introductory text (i.e. In the last 7 days, because of lack of food did you have to use the following strategies…) and then read off a list of food items/strategies.

The variable names in this guide were largely copied from a draft master codebook which is forthcoming.

1.0.5 About coding style

Most of the code in this guide uses the tidyverse, but it might not be elegant or efficient. It gets the job done but future work will be to try to make it comply with the tidyverse style code: https://style.tidyverse.org/.

For now, one key thing to note. Most of the time, the code refers to the label (i.e. “Yes”) rather than the value (i.e. “1”). There are trade-offs with this approach. Using a label makes it easier to read what the code is trying to do and troubleshoot a problem. The disadvantage is it makes the code more long winded and it makes it more fussy because the labels (and sometimes they can be long) must exactly match the code.

1.0.6 One last about - About proposing fixes and suggestions

Finally, this guide is just a start. If you spot an error, find something confusing, have a suggestion or a better way of doing something - don’t be a stranger. You can create an issue https://docs.github.com/en/github/managing-your-work-on-github/creating-an-issue on the github repository of this guide https://github.com/WFP-VAM/RBD_FS_CH_guide_EN