~dricottone/replications

59da47fa9c6a18cc9250376127ad0d2c071ec6fb — Dominic Ricottone a month ago 49ad383
ANES updates

Converted the generated Stata data preparation syntax into working
scripts.

Passed all files through `dos2unix`.

Added a comparison script.
5 files changed, 317 insertions(+), 180 deletions(-)

A data/anes/.gitignore
M data/anes/README.md
A data/anes/anes 2004 2012 2020 comparison.do
A data/anes/anes 2004 cumulative dictionary.do
R data/anes/{sub-stata.txt => anes 2004 cumulative infile.do}
A data/anes/.gitignore => data/anes/.gitignore +9 -0
@@ 0,0 1,9 @@
# 2004
sub-data.txt

# 2012
08475-0001-Data.dta

# 2020
anes_timeseries_cdf_stata_20220916.dta


M data/anes/README.md => data/anes/README.md +40 -27
@@ 1,27 1,40 @@
File origins.

## sub-data.txt

Sourced from ANES 2004 cumulative data file.
Raw data in delimited format.

## sub-stata.txt

Sourced from ANES 2004 cumulative data file.
Supplemental data preparation that is appropriate only for Stata.

## 08475-0001-Data.dta

Sourced from ANES 2012 cumulative data file.
Data prepared in Stata format.

## 08475-0001-Supplemental_syntax.do

Sourced from ANES 2012 cumulative data file.
Supplemental data preparation that is appropriate only for Stata.

## anes_timeseries_cdf_stata_20220916.dta

Sourced from ANES 2020 cumulative data file.
Data prepared in Stata format.

File origins.

## sub-data.txt

Sourced from ANES 2004 cumulative data file.
Raw data in delimited format.

## anes 2004 cumulative dictionary.do

Sourced from ANES 2004 cumulative data file.
Raw data import routine.
Prepared from original '' file.

## anes 2004 cumulative infile.do

Sourced from ANES 2004 cumulative data file.
Supplemental data preparation that is appropriate only for Stata.
Prepared from original '' file.

## 08475-0001-Data.dta

Sourced from ANES 2012 cumulative data file.
Data prepared in Stata format.

## 08475-0001-Supplemental_syntax.do

Sourced from ANES 2012 cumulative data file.
Supplemental data preparation that is appropriate only for Stata.

## anes_timeseries_cdf_stata_20220916.dta

Sourced from ANES 2020 cumulative data file.
Data prepared in Stata format.

## anes 2004 2012 2020 comparison.do

My work.
Compares the 2004, 2012, and 2020 cumulative data files in a simple way.
Demonstrates that data was adjusted in non-trivial ways over time.


A data/anes/anes 2004 2012 2020 comparison.do => data/anes/anes 2004 2012 2020 comparison.do +117 -0
@@ 0,0 1,117 @@
// ANES 2004 cumulative file
clear all

do "anes 2004 cumulative infile.do"
svyset VCF0006A [pweight=VCF0009A], vce(linearized)

keep if inlist(VCF0004,1982,1984,1998,1992,1996,2000,2002,2004)
recode VCF0803 (1/3=1 "Lib") (4 9=2 "Mod") (5/7=3 "Cons") (else=.), gen(conserv)
svy: tab conserv VCF0004, col
/*
(running tabulate on estimation sample)

Number of strata =      1                         Number of obs   =     12,528
Number of PSUs   = 11,368                         Population size = 12,501.836
                                                  Design df       =     11,367

-------------------------------------------------------------------------
RECODE of |
VCF0803   |
(Liberal- |
Conservat |
ive 7pt   |                         Year of Study                        
Scale)    |  1982   1984   1992   1996   1998   2000   2002   2004  Total
----------+--------------------------------------------------------------
      Lib | .1464  .1786  .2022  .1813  .1837  .1963  .2199  .2221  .1904
      Mod | .5829  .5357  .4975  .4929  .5134  .5016  .4362  .4591  .5046
     Cons | .2707  .2858  .3003  .3258  .3029  .3021  .3439  .3188   .305
          | 
    Total |     1      1      1      1      1      1      1      1      1
-------------------------------------------------------------------------
Key: Column proportion

  Pearson:
    Uncorrected   chi2(14)        =   90.6470
    Design-based  F(13.60,  1.5e+05)=    5.6565   P = 0.0000
*/

*******************************************************************************

// ANES 2012 cumulative file
clear all

use "08475-0001-Data.dta"
do "08475-0001-Supplemental_syntax.do"
svyset VCF0006A [pweight=VCF0009Z], vce(linearized)

keep if inlist(VCF0004,1982,1984,1998,1992,1996,2000,2002,2004)
recode VCF0803 (1/3=1 "Lib") (4 9=2 "Mod") (5/7=3 "Cons") (else=.), gen(conserv)
svy: tab conserv VCF0004, col

/*
(running tabulate on estimation sample)

Number of strata =      1                         Number of obs   =     12,675
Number of PSUs   = 11,515                         Population size = 12,653.846
                                                  Design df       =     11,514

-------------------------------------------------------------------------
RECODE of |
VCF0803   |
(IDEOLOGY |
:         |
Liberal-C |
onservati |                 STUDY VARIABLE: Year of Study                
ve Scale) |  1982   1984   1992   1996   1998   2000   2002   2004  Total
----------+--------------------------------------------------------------
      Lib | .1464  .1786  .2022  .1813  .1837  .1963  .2199  .1877  .1875
      Mod | .5829  .5357  .4975  .4929  .5134  .5016  .4362  .5017  .5081
     Cons | .2707  .2858  .3003  .3258  .3029  .3021  .3439  .3106  .3044
          | 
    Total |     1      1      1      1      1      1      1      1      1
-------------------------------------------------------------------------
Key: Column proportion

  Pearson:
    Uncorrected   chi2(14)        =   79.4757
    Design-based  F(13.60,  1.6e+05)=    4.9729   P = 0.0000
*/

*******************************************************************************

//ANES 2020 cumulative file
clear all
use "anes_timeseries_cdf_stata_20220916.dta"
svyset VCF0006a [pweight=VCF0009z], vce(linearized)

keep if inlist(VCF0004,1982,1984,1998,1992,1996,2000,2002,2004)
recode VCF0803 (1/3=1 "Lib") (4 9=2 "Mod") (5/7=3 "Cons") (else=.), gen(conserv)
svy: tab conserv VCF0004, col

/*
(running tabulate on estimation sample)

Number of strata =      1                         Number of obs   =     12,675
Number of PSUs   = 11,515                         Population size = 12,653.846
                                                  Design df       =     11,514

-------------------------------------------------------------------------
RECODE of |
VCF0803   |
(Liberal- |
Conservat |
ive       |                         Year of Study                        
Scale)    |  1982   1984   1992   1996   1998   2000   2002   2004  Total
----------+--------------------------------------------------------------
      Lib | .1464  .1786  .2022  .1813  .1837  .1963  .2199  .1877  .1875
      Mod | .5829  .5357  .4975  .4929  .5134  .5016  .4362  .5017  .5081
     Cons | .2707  .2858  .3003  .3258  .3029  .3021  .3439  .3106  .3044
          | 
    Total |     1      1      1      1      1      1      1      1      1
-------------------------------------------------------------------------
Key: Column proportion

  Pearson:
    Uncorrected   chi2(14)        =   79.4757
    Design-based  F(13.60,  1.6e+05)=    4.9729   P = 0.0000
*/

A data/anes/anes 2004 cumulative dictionary.do => data/anes/anes 2004 cumulative dictionary.do +44 -0
@@ 0,0 1,44 @@
dictionary using "sub-data.txt" {
*****************************************************************
*  Replace 'Y' with the name of the data file.
*  The default suffix is '.raw'.
*
*  Put this dictionary into a separate file (with the suffix .dct).
*  (The first line of the dictionary file must contain the
*  'dictionary' command.)
*
*  Then edit the name of that dictionary file into the 'do-file'
*  portion of the STATA definitions.
*******************************************************************
*  Records per case:
_lines(1)

_line(1)

_column(1)     long  CASEID             %8f    "Case ID"
_column(11)    str28 VERSION            %28s   "ANES Version Number"
_column(41)    int   VCF0004            %4f    "Year of Study"
_column(46)    int   VCF0006            %4f    "Study Respondent Number"
_column(51)    long  VCF0006A           %8f    "Unique Respondent Number"
_column(60)    float VCF0009            %6.4f  "Type 0 General Weight Variable"
_column(67)    float VCF0009A           %6.4f  "Type 0 Weight - Post-Stratified Pre"
_column(74)    float VCF0010            %6.4f  "Type 1 General Weight Variable"
_column(81)    float VCF0010A           %6.4f  "Type 1 Weight - Post-stratified Pre"
_column(88)    float VCF0011            %6.4f  "Type 2 General Weight Variable"
_column(95)    float VCF0011A           %6.4f  "Type 2 Weight - Post-Stratified Pre"
_column(102)   int   VCF0140A :VCF0140A %1f    "R Education 7-category"
_column(104)   int   VCF0302  :VCF0302  %1f    "Initial Party ID Response"
_column(106)   int   VCF0702  :VCF0702  %1f    "Did R Vote in Election"
_column(108)   int   VCF0717  :VCF0717  %1f    "Did R Try to Influence Others Vote"
_column(110)   int   VCF0718  :VCF0718  %1f    "Did R Attend Political Meetings"
_column(112)   int   VCF0719  :VCF0719  %1f    "Did R Work for Party or Candidate"
_column(114)   int   VCF0720  :VCF0720  %1f    "Did R Display Candidate Button/Stic"
_column(116)   int   VCF0721  :VCF0721  %1f    "Did R Donate Money to Party/Candida"
_column(118)   int   VCF0803  :VCF0803  %1f    "Liberal-Conservative 7pt Scale"
_column(120)   int   VCF0806  :VCF0806  %1f    "R Position 7pt Govt Health Insuranc"
_column(122)   int   VCF0809  :VCF0809  %1f    "R Position 7pt Govt Guaranteed Jobs"
_column(124)   int   VCF0830  :VCF0830  %1f    "R Position 7pt Aid to Blacks Scale"
_column(126)   int   VCF0838  :VCF0838  %1f    "When Should Abortion Be Allowed by"
_column(128)   int   VCF0839  :VCF0839  %1f    "R Position 7pt Govt Services/Spendi"
_column(130)   int   VCF0843  :VCF0843  %1f    "R Position 7pt Defense Spending Sca"
}

R data/anes/sub-stata.txt => data/anes/anes 2004 cumulative infile.do +107 -153
@@ 1,153 1,107 @@
*******************************************************************
*  Stata "do-file" file with labels and missing data specifications
*  Created by ddltox on Nov 15, 2024  (Fri 10:03 AM PST)
*  DDL source file: "/var/www/sda/tmpdir/ddl3640268546116940221.txt".
*
*  Note that the data dictionary is given at the end of this file.
*  Put the dictionary into a separate file (by editing this file).
*  Then specify below the name of the dictionary file.
*
*  DDL file gives the following dataset description:
*    Records per case: 1
*    Record length:    130
*******************************************************************


label data "American National Election Study 1948-2004 - Cumulative"

#delimit ;
label define VCF0140A  1 "8 grades or less ('grade school')" 
                       2 "9-12 grades ('high school'), no diploma/equivalency" 
                       3 "12 grades, diploma or equivalency" 
                       4 "12 grades, diploma or equivalency, plus non-academic" 
                       5 "Some college, no degree;junior/community college level" 
                       6 "BA level degree" 7 "Advanced degrees incl. LLB" 
                       8 "DK" 9 "NA;no Pre IW;short-form 'new' Cross Section" ;
label define VCF0302   1 "Republican" 2 "Independent" 
                       3 "No preference;none;neither" 4 "Other" 5 "Democrat" 
                       8 "DK" 9 "NA;refused" ;
label define VCF0702   0 "DK;NA;no Post IW;refused to say if voted;" 
                       1 "No, did not vote" 2 "Yes, voted" ;
label define VCF0717   0 "DK;NA;no Post IW;form III,IV (1972);abbrev." 1 "No" 
                       2 "Yes" ;
label define VCF0718   0 "DK;NA;no Post IW;form III,IV (1972);abbrev." 1 "No" 
                       2 "Yes" ;
label define VCF0719   0 "DK;NA;no Post IW;form III,IV (1972);abbrev." 1 "No" 
                       2 "Yes" ;
label define VCF0720   0 "DK;NA;no Post IW;form III,IV (1972);abbrev." 1 "No" 
                       2 "Yes" ;
label define VCF0721   0 "DK;NA;no Post IW;form III,IV (1972);abbrev." 
                       1 "No (includes 'not asked for money' in 1966,1968)" 
                       2 "Yes (includes 'tax check-off' in 1976)" ;
label define VCF0803   
                       0 "NA;no Post IW;form III,IV (1972);R not administered" 
                       1 "Extremely liberal" 2 "Liberal" 3 "Slightly liberal" 
                       4 "Moderate, middle of the road" 
                       5 "Slightly conservative" 6 "Conservative" 
                       7 "Extremely conservative" 
                       9 "DK;haven't thought much about it" ;
label define VCF0806   
                       0 "NA;form II (1972);no Post IW;telephone IW (1984: see" 
                       1 "Government insurance plan" 
                       7 "Private insurance plan" 
                       9 "DK;haven't thought much about it" ;
label define VCF0809   0 "NA;no Post IW;form A (1986);telephone IW (2000)" 
                       1 "Government see to job and good standard of living" 
                       7 "Government let each person get ahead on his own" 
                       9 "DK;haven't thought much about it" ;
label define VCF0830   0 "NA;telephone IW (2000);no Post IW" 
                       1 "Government should help minority groups/blacks" 
                       7 "Minority groups/blacks should help themselves" 
                       9 "DK;haven't thought much about it" ;
label define VCF0838   0 "NA;no Post IW" 
                       1 "By law, abortion should never be permitted" 
                       2 "The law should permit abortion only in case of rape," 
                       3 "The law should permit abortion for reasons other than" 
                       4 "By law, a woman should always be able to obtain an" 
                       9 "DK;other" ;
label define VCF0839   0 "NA;telephone IW (2000)" 
                       1 "Government should provide many fewer services: reduce" 
                       7 "Government should provide many more services: increase" 
                       9 "DK;haven't thought much about it" ;
label define VCF0843   0 "NA;telephone IW (2000)" 
                       1 "Greatly decrease defense spending" 
                       7 "Greatly increase defense spending" 
                       9 "DK;haven't thought much about it" ;


#delimit cr

*******************************************************************
infile using X
* Replace 'X' with the name of the dictionary file. 
*
* The contents of the dictionary are given at the end of this file.
* Put the dictionary into a separate file (by editing this file).
* Then specify here the name of the dictionary file.
*******************************************************************
* The md, min and max specifications were translated 
* into the following "REPLACE...IF" statements:

replace VCF0140A = . if (VCF0140A == 0)
replace VCF0140A = . if (VCF0140A >= 8 ) 
replace VCF0302 = . if (VCF0302 == 8)
replace VCF0302 = . if (VCF0302 >= 9 ) 
replace VCF0702 = . if (VCF0702 == 0)
replace VCF0717 = . if (VCF0717 == 0)
replace VCF0718 = . if (VCF0718 == 0)
replace VCF0719 = . if (VCF0719 == 0)
replace VCF0720 = . if (VCF0720 == 0)
replace VCF0721 = . if (VCF0721 == 0)
replace VCF0803 = . if (VCF0803 == 0)
replace VCF0806 = . if (VCF0806 == 0)
replace VCF0809 = . if (VCF0809 == 0)
replace VCF0830 = . if (VCF0830 == 0)
replace VCF0838 = . if (VCF0838 == 0)
replace VCF0839 = . if (VCF0839 == 0)
replace VCF0843 = . if (VCF0843 == 0)


dictionary using Y {
*****************************************************************
*  Replace 'Y' with the name of the data file.
*  The default suffix is '.raw'.
*
*  Put this dictionary into a separate file (with the suffix .dct).
*  (The first line of the dictionary file must contain the
*  'dictionary' command.)
*
*  Then edit the name of that dictionary file into the 'do-file'
*  portion of the STATA definitions.
*******************************************************************
*  Records per case:
_lines(1)

_line(1)

_column(1)     long  CASEID             %8f    "Case ID"
_column(11)    str28 VERSION            %28s   "ANES Version Number"
_column(41)    int   VCF0004            %4f    "Year of Study"
_column(46)    int   VCF0006            %4f    "Study Respondent Number"
_column(51)    long  VCF0006A           %8f    "Unique Respondent Number"
_column(60)    float VCF0009            %6.4f  "Type 0 General Weight Variable"
_column(67)    float VCF0009A           %6.4f  "Type 0 Weight - Post-Stratified Pre"
_column(74)    float VCF0010            %6.4f  "Type 1 General Weight Variable"
_column(81)    float VCF0010A           %6.4f  "Type 1 Weight - Post-stratified Pre"
_column(88)    float VCF0011            %6.4f  "Type 2 General Weight Variable"
_column(95)    float VCF0011A           %6.4f  "Type 2 Weight - Post-Stratified Pre"
_column(102)   int   VCF0140A :VCF0140A %1f    "R Education 7-category"
_column(104)   int   VCF0302  :VCF0302  %1f    "Initial Party ID Response"
_column(106)   int   VCF0702  :VCF0702  %1f    "Did R Vote in Election"
_column(108)   int   VCF0717  :VCF0717  %1f    "Did R Try to Influence Others Vote"
_column(110)   int   VCF0718  :VCF0718  %1f    "Did R Attend Political Meetings"
_column(112)   int   VCF0719  :VCF0719  %1f    "Did R Work for Party or Candidate"
_column(114)   int   VCF0720  :VCF0720  %1f    "Did R Display Candidate Button/Stic"
_column(116)   int   VCF0721  :VCF0721  %1f    "Did R Donate Money to Party/Candida"
_column(118)   int   VCF0803  :VCF0803  %1f    "Liberal-Conservative 7pt Scale"
_column(120)   int   VCF0806  :VCF0806  %1f    "R Position 7pt Govt Health Insuranc"
_column(122)   int   VCF0809  :VCF0809  %1f    "R Position 7pt Govt Guaranteed Jobs"
_column(124)   int   VCF0830  :VCF0830  %1f    "R Position 7pt Aid to Blacks Scale"
_column(126)   int   VCF0838  :VCF0838  %1f    "When Should Abortion Be Allowed by"
_column(128)   int   VCF0839  :VCF0839  %1f    "R Position 7pt Govt Services/Spendi"
_column(130)   int   VCF0843  :VCF0843  %1f    "R Position 7pt Defense Spending Sca"
}
*******************************************************************
*  Stata "do-file" file with labels and missing data specifications
*  Created by ddltox on Nov 13, 2024  (Wed 07:10 PM PST)
*  DDL source file: "/var/www/sda/tmpdir/ddl489112400220024910.txt".
*
*  Note that the data dictionary is given at the end of this file.
*  Put the dictionary into a separate file (by editing this file).
*  Then specify below the name of the dictionary file.
*
*  DDL file gives the following dataset description:
*    Records per case: 1
*    Record length:    130
*******************************************************************


label data "American National Election Study 1948-2004 - Cumulative"

#delimit ;
label define VCF0140A  1 "8 grades or less ('grade school')" 
                       2 "9-12 grades ('high school'), no diploma/equivalency" 
                       3 "12 grades, diploma or equivalency" 
                       4 "12 grades, diploma or equivalency, plus non-academic" 
                       5 "Some college, no degree;junior/community college level" 
                       6 "BA level degree" 7 "Advanced degrees incl. LLB" 
                       8 "DK" 9 "NA;no Pre IW;short-form 'new' Cross Section" ;
label define VCF0302   1 "Republican" 2 "Independent" 
                       3 "No preference;none;neither" 4 "Other" 5 "Democrat" 
                       8 "DK" 9 "NA;refused" ;
label define VCF0702   0 "DK;NA;no Post IW;refused to say if voted;" 
                       1 "No, did not vote" 2 "Yes, voted" ;
label define VCF0717   0 "DK;NA;no Post IW;form III,IV (1972);abbrev." 1 "No" 
                       2 "Yes" ;
label define VCF0718   0 "DK;NA;no Post IW;form III,IV (1972);abbrev." 1 "No" 
                       2 "Yes" ;
label define VCF0719   0 "DK;NA;no Post IW;form III,IV (1972);abbrev." 1 "No" 
                       2 "Yes" ;
label define VCF0720   0 "DK;NA;no Post IW;form III,IV (1972);abbrev." 1 "No" 
                       2 "Yes" ;
label define VCF0721   0 "DK;NA;no Post IW;form III,IV (1972);abbrev." 
                       1 "No (includes 'not asked for money' in 1966,1968)" 
                       2 "Yes (includes 'tax check-off' in 1976)" ;
label define VCF0803   
                       0 "NA;no Post IW;form III,IV (1972);R not administered" 
                       1 "Extremely liberal" 2 "Liberal" 3 "Slightly liberal" 
                       4 "Moderate, middle of the road" 
                       5 "Slightly conservative" 6 "Conservative" 
                       7 "Extremely conservative" 
                       9 "DK;haven't thought much about it" ;
label define VCF0806   
                       0 "NA;form II (1972);no Post IW;telephone IW (1984: see" 
                       1 "Government insurance plan" 
                       7 "Private insurance plan" 
                       9 "DK;haven't thought much about it" ;
label define VCF0809   0 "NA;no Post IW;form A (1986);telephone IW (2000)" 
                       1 "Government see to job and good standard of living" 
                       7 "Government let each person get ahead on his own" 
                       9 "DK;haven't thought much about it" ;
label define VCF0830   0 "NA;telephone IW (2000);no Post IW" 
                       1 "Government should help minority groups/blacks" 
                       7 "Minority groups/blacks should help themselves" 
                       9 "DK;haven't thought much about it" ;
label define VCF0838   0 "NA;no Post IW" 
                       1 "By law, abortion should never be permitted" 
                       2 "The law should permit abortion only in case of rape," 
                       3 "The law should permit abortion for reasons other than" 
                       4 "By law, a woman should always be able to obtain an" 
                       9 "DK;other" ;
label define VCF0839   0 "NA;telephone IW (2000)" 
                       1 "Government should provide many fewer services: reduce" 
                       7 "Government should provide many more services: increase" 
                       9 "DK;haven't thought much about it" ;
label define VCF0843   0 "NA;telephone IW (2000)" 
                       1 "Greatly decrease defense spending" 
                       7 "Greatly increase defense spending" 
                       9 "DK;haven't thought much about it" ;


#delimit cr

*******************************************************************
infile using "anes 2004 cumulative dictionary.do"
* Replace 'X' with the name of the dictionary file. 
*
* The contents of the dictionary are given at the end of this file.
* Put the dictionary into a separate file (by editing this file).
* Then specify here the name of the dictionary file.
*******************************************************************
* The md, min and max specifications were translated 
* into the following "REPLACE...IF" statements:

replace VCF0140A = . if (VCF0140A == 0)
replace VCF0140A = . if (VCF0140A >= 8 )
replace VCF0302 = . if (VCF0302 == 8)
replace VCF0302 = . if (VCF0302 >= 9 )
replace VCF0702 = . if (VCF0702 == 0)
replace VCF0717 = . if (VCF0717 == 0)
replace VCF0718 = . if (VCF0718 == 0)
replace VCF0719 = . if (VCF0719 == 0)
replace VCF0720 = . if (VCF0720 == 0)
replace VCF0721 = . if (VCF0721 == 0)
replace VCF0803 = . if (VCF0803 == 0)
replace VCF0806 = . if (VCF0806 == 0)
replace VCF0809 = . if (VCF0809 == 0)
replace VCF0830 = . if (VCF0830 == 0)
replace VCF0838 = . if (VCF0838 == 0)
replace VCF0839 = . if (VCF0839 == 0)
replace VCF0843 = . if (VCF0843 == 0)