In context-dependent/fsctemplates:

knitr::opts_chunk$set(echo = TRUE, dev = "CairoPNG", warning = FALSE, message = FALSE)
knitr::opts_knit$set(root.dir = here::here())

pacman::p_load(
  "tidyverse"
)

source("R/pipeline/load-data.R")

Introduction

Purpose

This notebook presents the results of initial exploratory analysis designed to support decision-making around data cleaning. Use it to explore the data produced by the load_data function and record cleaning decisions. These cleaning decisions will be formalized in the clean_data function.

Important questions to answer include:

• Samples: which sets of records will be used in the analysis?
• Joins: where are you combining data from different datasets and why?
• Filters: if any records are dropped from the raw data, why? What impact does each filter have on the sample size?
• Transformations: if any new variables are created, or raw variables are transformed in any way, how and why?
• Selections: which variables will be included in the final analysis dataset(s)

Support your decisions with analysis where appropriate.

Load data

Load the cached output from the load_data function.

dat <- load_data(use_cache = TRUE)

Summary of decisions

Summarize your decisions here, with a special focus on the samples.

Analysis

Put any analysis outputs here, and describe the results in text.

context-dependent/fsctemplates documentation built on April 5, 2020, 10:07 p.m.