-
Notifications
You must be signed in to change notification settings - Fork 56
/
Copy pathREADME.Rmd
120 lines (99 loc) · 10.6 KB
/
README.Rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
---
title: "README"
author: "CoronaNet Project Team"
date: "May 5th, 2021"
output: github_document
---
```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = FALSE,warning=FALSE,message = FALSE)
require(gghighlight)
require(dplyr)
require(tidyr)
require(stringr)
require(lubridate)
```
## About This Repository
This repository contains the raw data from the [CoronaNet data collection project](http://coronanet-project.org). This data is in a policy record format, in which one row equals one specific policy with end and beginning dates. While it is a very compact format, it is not necessarily ideal for data analysis projects. We also have 6 indices (social distancing, business restrictions, school restrictions, health monitoring, health resources and masks), and a wide data format with 156 of our indicators for January 1st to January 15th, 2021. Both the indices and the wide data format have one row per policy type per day per country, which is much easier to merge with our data sources. For more info, please see this [Github repository](https://github.com/saudiwin/corona_index).
If you do want to see the data in policy record format, which includes more information than what is released in our wide data, please see below for more information about what is in this repository.
# CoronaNet Raw Data
When using our raw data, we recommend checking the [CoronaNet Update Tracker](https://docs.google.com/spreadsheets/d/1h5dqVxLghvXr2wwl74ZUeSIWRQLySrxWCjD65_MevUM/edit#gid=0), so you can track our policy updates by country and subnational unit. We have hundreds of RAs working to keep the data up to date, but there will inevitably be issues in the data in terms of being up to date.
First, CoronaNet data releases:
**Please note that while we make every effort to validate this data, the speed and scale with which it was collected means that we cannot validate all of it. If you find an error in the data, please file an issue on this Github page.**
The format of the data is in country-day-`record_id` format. Some `record_id` values have letters appended to indicate that the general policy category `type` also has a value for `type_sub_cat`, which contains more detail about the policy, such as whether health resources refers to masks, ventilators, or hospitals. Some entries are marked as `new_entry` in the `entry_type` field for when a policy of that type was first implemented in the country. Later updates to those policies are marked as updates in `entry_type`. To see how policies are connected, look at the `policy_id` field for all policies from the first entry through updates for a given country/province/city. If an entry was corrected after initial data collection, it will read corrected in the `entry_type` field (the original incorrect data has already been replaced with the corrected data).
1. **`data/CoronaNet/data_bulk/coronanet_release[.rds/csv.gz]`** These files contain variables from the CoronaNet government response project, representing national and sub-national policy event data from more than 140 countries since January 1st, 2020. The data include source links, descriptions, targets (i.e. other countries), the type and level of enforcement, and a comprehensive set of policy types. For more detail on this data, you can see our [codebook here](https://docs.google.com/document/d/1zvNMpwj0onFvUZ_gLl4RRjqS-clbHv3TIX6EOHofsME).
2. **`data/CoronaNet/data_bulk/coronanet_release_allvars[.rds/csv.gz]`** These files contains the government response information from `coronanet_release.csv` along with the following datasets:
a. Tests from the CoronaNet testing database (see http://coronanet-project.org for more info);
b. Cases/deaths/recovered from the JHU data repository (https://github.com/CSSEGISandData/COVID-19);
c. Country-level covariates including GDP, V-DEM democracy scores, human rights indices, power-sharing indices, and press freedom indices from the Niehaus World Economics and Politics Dataverse (https://niehaus.princeton.edu/news/world-economics-and-politics-dataverse)
3. **`data/CoronaNet/data_country/coronanet_release_[country].csv`** For each country in `coronanet_release`, we have generated a separate data file in a .csv format.
4. **`data/CoronaNet/data_country/coronanet_release_allvars_[country].csv`** For each country in `coronanet_release_allvars`, we have generated a separate data file in a .csv format.
## `coronanet_release.csv` Field Dictionary
1. `record_id` Unique identifier for each unique policy record
2. `policy_id` Identifier linking new policies with subsequent updates to policies
3. `recorded_date` When the record was entered into our data
4. `date_updated` When we can confirm the country - policy type was last checked/updated (we can only confirm policy type for a given country is up to date as of this date)
5. `date_announced` When the policy is announced
6. `date_start` When the policy goes into effect
7. `date_end` When the policy ends (if it has an explicit end date)
8. `entry_type` Whether the record is new, meaning no restriction had been in place before, or an update (restriction was in place but changed). Corrections are corrections to previous entries.
9. `event_description` A short description of the policy change
10. `domestic_policy` Indicates where policy targets an area within the initiating country (i.e. is domestic in nature)
11. `type` The category of the policy
12. `type_sub_cat` The sub-category of the policy (if one exists)
13. `type_text` Any additional information about the policy type (such as the number of ventilators/days of quarantine/etc.)
14. `index_high_est` The high (95% posterior density) estimate of the country policy activity score (0-100)
15. `index_med_est` The median (most likely) estimate of the country policy activity score (0-100)
16. `index_low_est` The low (95% posterior density) estimate of the country policy activity score (0-100)
17. `index_country_rank` The relative rank by each day for each country on the policy activity score
18. `country` The country initiating the policy
19. `init_country_level` Whether the policy came from the national level or a sub-national unit
20. `province` Name of sub-national unit
21. `target_country` Which foreign country a policy is targeted at (i.e. travel policies)
22. `target_geog_level` Whether the target of the policy is a country as a whole or a sub-national unit of that country
23. `target_region` The name of a regional grouping (like ASEAN) that is a target of the policy (if any)
24. `target_province` The name of a province targeted by the policy (if any)
25. `target_city` The name of a city targeted by the policy (if any)
26. `target_other` Any geographical entity that does not fit into the targeted categories mentioned above
27. `target_who_what` Who the policy is targeted at
28. `target_direction` Whether a travel-related policy affects people coming in (Inbound) or leaving (Outbound)
29. `travel_mechanism` If a travel policy, what kind of transportation it affects
30. `compliance` Whether the policy is voluntary or mandatory
31. `enforcer` What unit in the country is responsible for enforcement
32. `link` A link to at least one source for the policy
33. `ISO_A3` 3-digit ISO country codes
34. `ISO_A2` 2-digit ISO country codes
<!-- 22. `severity_index_5perc` 5% posterior low estimate (i.e. lower bound of uncertainty interval) for severity index -->
<!-- 23. `severity_index_median` posterior median estimate (point estimate) for severity index, which comes from a Bayesian latent variable model aggregating across policy types to measure country-level policy severity (see paper on our website) -->
<!-- 24. `severity_index_5perc` 95% posterior high estimate (i.e. upper bound of uncertainty interval) for severity index -->
## `coronanet_release_allvars.csv` Field Dictionary
1. All of the fields listed above, plus
2. `tests_daily_or_total` Whether a country reports the daily count of tests a cumulative total
3. `tests_raw` The number of reported tests collected from host country websites or media reports
4. `deaths` The number of COVID-19 deaths, aggregated to the country-day level (JHU CSSE data)
5. `confirmed_cases` The number of confirmed cases of COVID-19, aggregated to the country-day level (JHU CSSE data)
6. `recovered` The number of recoveries from COVID-19, aggregated to the country-day level (JHU CSSE data)
7. `ccode` The Correlates of War country code
8. `ifs` IMF IFS country code
9. `Rank_FP` (most recent year available from Niehaus dataset) Reporters without Borders Press Freedom Annual Ranking
10. `Score_FP` (most recent year available from Niehaus dataset) Reporters with Borders Press Freedom Score
11. `state_IDC` (most recent year available from Niehaus dataset) State/Provincial Governments Locally Elected
12. `muni_IDC` (most recent year available from Niehaus dataset) Municipal Governments Locally Elected
13. `dispersive_IDC` (most recent year available from Niehaus dataset) Dispersive Powersharing
14. `constraining_IDC` (most recent year available from Niehaus dataset) Constraining Powersharing
15. `inclusive_IDC` (most recent year available from Niehaus dataset) Inclusive powersharing
16. `sfi_SFI` (most recent year available from Niehaus dataset) State fragility index
17. `ti_cpi_TI` (most recent year available from Niehaus dataset) Corruption perceptions index
18. `pop_WDI_PW` (most recent year available from Niehaus dataset) World Bank population
19. `gdp_WDI_PW` (most recent year available from Niehaus dataset) World Bank GDP (total)
20. `gdppc_WDI_PW` (most recent year available from Niehaus dataset) World Bank GDP per capita
21. `growth_WDI_PW` (most recent year available from Niehaus dataset) World Bank GDP growth percent
22. `lnpop_WDI_PW` (most recent year available from Niehaus dataset) Log of World Bank population
23. `lngdp_WDI_PW` (most recent year available from Niehaus dataset) Log of World Bank GDP
24. `lngdppc_WDI_PW` (most recent year available from Niehaus dataset) Log of World Bank GDP per capita
25. `disap_FA` (most recent year available from Niehaus dataset) 3 category, ordered variable for disappearances index
26. `polpris_FA` (most recent year available from Niehaus dataset) 3 category, ordered variable for political imprisonment index
27. `latentmean_FA` (most recent year available from Niehaus dataset) the posterior mean of the latent variable index for human rights protection)
28. `transparencyindex_HR` (most recent year available from Niehaus dataset) Transparency Index
29. `EmigrantStock_EMS` (most recent year available from Niehaus dataset) Total emmigrant stock from
30. `v2x_polyarchy_VDEM` (most recent year available from Niehaus dataset) Electoral democracy index
31. `news_WB` (most recent year available from Niehaus dataset) Daily newspapers (per 1,000 people)