README.Rmd

---
output: github_document
---

<!-- README.md is generated from README.Rmd. Please edit that file -->

```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)
```

# {rmsSampling}

<!-- badges: start -->
<!-- badges: end -->

The goal of {rmsSampling} is to ease the implementation of the RMS (Results Monitoring Survey) Sampling guidance, with a perpective of streamlining as many assumptions as possible. 

Building a sampling plan is not a mechanical process and needs back and forth review in order to be optimal. Using information already available from UNHCR Official Statistics for each country, and based on the confirmation of sampling assumptions, this  companion app  will help operations building an initial Methodological Approach Documentation that can then be reviewed by regional DIMA in order to inform the preparation of an RMS survey.

The companion app should help disseminating knowledge on sampling for a non-expert audience (_typically UNHCR Information Management officer in the field_).


This App is still under development - the work in progress version is available here: [https://rstudio.unhcr.org/rmsSampling/](https://rstudio.unhcr.org/rmsSampling/)  

## A guided workflow

User are guided through a workflow:  

 * Select the country so that the last population statistics aggregated by group are pulled  
 
     * RAS  - aka REF+ASY+OIP, Refugees, Asylum-seeker and other in Need of International Protection -  "Pillar-1-Refugee"  
     
     * STA: Stateless Persons --  "Pillar-2-Stateless" 
     
     * IDP: Internally displaced Persons   -  "Pillar-4-IDP"  
     
     * REP (RET+ROC): Returned Refugees & IDPs - "Pillar-3-Reintegration"  
     
     * OOC: Other of Concerns         
 
Then for each group a specific tab will be available with a series of conditional questions to confirm wether one the 7 potential sampling method can apply:   

 * Flag population groups with less than 5,000 individuals, to confirm if there is really budget to implement a sampling strategy  
 
 * Request user to  indicate if a reliable registration group list is available for each population group  
 
 * Check size of the country to confirm if the implementation area is small (allowing for __Simple Random Sampling without stratification__) 
 
 * Check if disaggregation by geographic location are available within the countries  (allowing for __Simple Random Sampling within strata__)  
 * Check breakdown by country of Origin to confirm if potential stratification by country of orgin would make sense for Refugee and Asylum Seekers?  (allowing for __multiple Stage Cluster Sampling__)  
 
 * Check if budget and time is available to conduct a list exercise (allowing for either __Probability Proportion to Size__ or __Adaptive Cluster Sampling__)  
 
 * Check if the population tend to gather to a certain location on a specific day time? (allowing for a __Location Time Sampling__) 
 
 * Check if the community if well connected and is there is adequate time to conduct a formative survey to identify the sees and prepare coupons? (allowing for a __Respondent Driven Sampling__)  
 
 * ask for expected margin of errors, level of confidence and sample power. 
 
 
 As an output they can generate a report to document the appropriate sampling approach by country.

 

## Installation

You can install the development version of rmsSampling like so:

``` r
install.packages("pak")
pak::pkg_install("unhcr-americas/rmsSampling")  
```