Update Freedom House 2016, Prefix variable labels to avoid collision,…

… Strip old dataset labels.
mpearce · Aug 3, 2016 · 9cd1a15 · 9cd1a15
1 parent 427fb9e
commit 9cd1a15
Show file tree

Hide file tree

Showing 7 changed files with 62 additions and 11 deletions.
diff --git a/README.md b/README.md
@@ -24,17 +24,19 @@ UCIMerge requires STATA 13. The .csv files which link countries across datasets
 ## Currently Supported Datasets
 
 * [Norris 2009](https://sites.google.com/site/pippanorris3/research/data#TOC-Democracy-Time-series-Data-Release-3.0-January-2009)
-* [Freedom House 2015](https://freedomhouse.org/report/freedom-world/freedom-world-2015)
+* [Freedom House 2016](https://freedomhouse.org/report/freedom-world/freedom-world-2015)
 * [Polity IV](http://www.systemicpeace.org/polityproject.html)
 * Polity IV Coups
 * [World Development Indicators](http://data.worldbank.org)
 * [KOF Index of Globalization](http://globalization.kof.ethz.ch)
 * [The Lexical Index of Electoral Democracy (LIED)](http://ps.au.dk/forskning/forskningsprojekter/dedere/datasets/)
 * [CIRI Human Rights Dataset](http://www.humanrightsdata.com)
 * [Quality of Government Standard dataset](http://qog.pol.gu.se/data/datadownloads/qogstandarddata)
-* [Cross National Time Series](http://www.databanksinternational.com)
+* [Cross National Time Series](http://www.cntsdata.com)
 * [Penn World Table version 8.1](http://www.rug.nl/research/ggdc/data/pwt/pwt-8.1)
 
+You can add your own datasets by using one of these examples as a template.
+
 ## Structure and Philosophy
 
 Intuitive directory structures and naming conventions means writing less code!
@@ -54,7 +56,7 @@ Intuitive directory structures and naming conventions means writing less code!
 Please contribute! If you discover an error, please submit an issue on github or send a fix.
 
 To contribute a new dataset:
-* create a new merge .do file in the /lib directory. The fh.do is a well documented template.
+* create a new merge .do file in the /lib directory. The polity.do is a well documented template.
 * include the file in Master.do file
 * update the UCIMergeList.csv with the new country code entries. Sort the merge list by the Source and UCINumeric fields.
 

diff --git a/lib/begin.do b/lib/begin.do
@@ -15,6 +15,22 @@ program define prefix
 	}
 end
 
+* Rename value labels
+capture program drop prefix_var_labels
+program define prefix_var_labels
+	display as text "`0'"
+	ds, has(vallabel)
+	local vars `r(varlist)'
+	foreach var of local vars {
+		local labname : value label `var'
+		label copy `labname' `0'_`labname', replace
+		label value `var' `0'_`labname'
+	}
+	labelbook, problems
+	la drop `r(notused)'
+end
+
+
 * Convenience method to create value lables from string variables
 capture program drop encode_value_labels
 program define encode_value_labels
@@ -59,6 +75,23 @@ program define use_or_import_source_excel
 	}
 end
 
+
+capture program drop use_or_import_source_csv
+program define use_or_import_source_csv
+	args PREFIX URL
+	capture confirm file "source/`PREFIX'.dta"
+	if _rc!=0 {
+		noisily display "Importing `PREFIX' from web."
+		import delimited "`URL'", case(preserve) clear
+		compress
+		save "source/`PREFIX'.dta"
+	}
+	else {
+		use "source/`PREFIX'.dta", clear
+	}
+end
+
+
 capture program drop use_or_import_source
 program define use_or_import_source
 	args PREFIX URL

diff --git a/lib/end.do b/lib/end.do
@@ -1,5 +1,6 @@
 drop if year == .
 notes drop _dta
+label data
 
 rename UCINumeric country
 gen UCINumeric = country

diff --git a/lib/fh.do b/lib/fh.do
@@ -1,36 +1,42 @@
-* 2015 Freedomhouse Freedom in the World Report
-* https://freedomhouse.org/report/freedom-world/freedom-world-2015
+* 2016 Freedomhouse Freedom in the World Report
+* https://freedomhouse.org/report-types/freedom-world
 
 * If no cached file in source/ go here:
-local URL "http://mattpearce.name/files/fh2015.dta"
+local URL "https://raw.githubusercontent.com/mpearce/FH/master/FH2016.csv"
 * Prefix this source's variables with:
 local PREFIX "fh"
 * What is the year variable called in the source:
 local YEARVAR "year"
 * What is the country variable called in the source:
-local COUNTRYVAR "country"
+local COUNTRYVAR "UCINumeric"
 * Which set of ids are used, matching "source" field in UCIMergeList:
-local MERGEWITH "fh"
+local MERGEWITH "UCI"
+* In this example, the source already has a UCINumeric column.
+* The "UCI" source is a replication of the UCICountries codes.
+
 * Which field in UCIMergelist matches the country variable above (Name, Alpha or Numeric):
-local MERGEUSING "Name"
+local MERGEUSING "Numeric"
 
 * Create the merge file to use with this source
 tempfile merge
 prep_mergefile `MERGEWITH' `MERGEUSING' `merge'
 
 * If the file source file doesn't already exist then download and create it.
-use_or_import_source `PREFIX' `URL'
+use_or_import_source_csv `PREFIX' `URL'
 
 * Do any transformations/recodes/corrections here
-
 gen sum = cl + pr
 label variable sum "Combined Civil Liberties and Political Rights score."
 
 * Prefix the variables
 prefix `PREFIX'
 
+encode_value_labels fh_status
+
 * Combine the merge file IDs with the current source
 merge_ids `PREFIX' `YEARVAR' `COUNTRYVAR' `MERGEUSING' `merge'
 
+drop `PREFIX'_UCINumeric
+
 * Merge with the Dataset
 merge_with_master
diff --git a/lib/norris.do b/lib/norris.do
@@ -26,6 +26,7 @@ This merge uses Refno accounting for the inconsistencies below. Refno may be not
 */
 
 prefix `PREFIX'
+prefix_var_labels `PREFIX'
 merge_ids `PREFIX' `YEARVAR' `COUNTRYVAR' `MERGEUSING' `merge'
 
 duplicates drop UCINumeric year, force

diff --git a/lib/polity.do b/lib/polity.do
@@ -1,11 +1,17 @@
 * 2015 Polity 4
 * http://www.systemicpeace.org/polity/polity4.htm
 
+* If no cached file in source/ go here:
 local URL "http://www.systemicpeace.org/inscr/p4v2015.xls"
+* Prefix this source's variables with:
 local PREFIX "polity"
+* What is the year variable called in the source:
 local YEARVAR "year"
+* What is the country variable called in the source:
 local COUNTRYVAR "ccode"
+* Which set of ids are used, matching "source" field in UCIMergeList:
 local MERGEWITH "Polity"
+* Which field in UCIMergelist matches the country variable above (Name, Alpha or Numeric):
 local MERGEUSING "Numeric"
 
 tempfile merge

diff --git a/lib/qog.do b/lib/qog.do
@@ -20,8 +20,10 @@ ds ccode cname year ccodealp cname_year ccodealp_year ccodecow ccodewb version,
 egen _tmp = rownonmiss(`r(varlist)'), s
 drop if _tmp == 0
 drop _tmp
+label data
 
 prefix `PREFIX'
+prefix_var_labels `PREFIX'
 merge_ids `PREFIX' `YEARVAR' `COUNTRYVAR' `MERGEUSING' `merge'
 
 * Merge with the Dataset