2008 Time Series Study

Updates & Errata

Current Dataset Release: May 19, 2015 Version
October 30, 2018
The defense spending 7-point scale (variable V083112 ) was correctly administered to respondents although the labels for the endpoint categories in the codebook documentation represent an error in the instrument. Endpoint labels in the codebook represent what appeared onscreen to interviewers during administration of the survey. However, although incorrect, they do not represent the endpoint labels which were correct in the Respondent Booklet used by respondents to make a number selection from the scale. The 2008 data for this variable should be considered representing questions administered comparably to other years.
May 19, 2015
We have been notified of a problem with the AMP module in the Advance Release of the 2008 ANES Time Series study (Variables V085310 to V085358a). The problem is due to an error in one of the computer programs that was written to implement the AMP on the ANES. We are currently researching the matter. We expect to have a detailed explanation of the problem and its implications for the usage of AMP variables posted within the next two weeks. If you are interested in analyzing the AMP data, we advise that you wait until we post a full explanation of the error. This information will be posted on this website as soon as possible.
March 20, 2009
We have been notified of a problem with the AMP module in the Advance Release of the 2008 ANES Time Series study (Variables V085310 to V085358a). The problem is due to an error in one of the computer programs that was written to implement the AMP on the ANES. We are currently researching the matter. We expect to have a detailed explanation of the problem and its implications for the usage of AMP variables posted within the next two weeks. If you are interested in analyzing the AMP data, we advise that you wait until we post a full explanation of the error. This information will be posted on this website as soon as possible.
January 25, 2015
Release adding category identifying black interviewer in vars V082254 and V084254
July 16, 2013
Summary variable V083136x (N4h2ax. [NEW] SUMMARY: Rep Pres cand on illeg immgr citznshp) was released without data. This summary variable may be created from existing variables as follows (SPSS):

if (V083136=1) V083136x = V083136a .
if (V083136=2 and any(V083136a,1,2,3)) V083136x = abs(V083136a-8) .
if (any(V083136a,-8,-9)) V083136x = V083136a .
if (any(V083136,-1,-8,-9)) V083136x = V083136 .
if (V083136 = 3) V083136x = 4 .

We thank Justin Wedeking from the University of Kentucky for noting this error.

December 19, 2012
These corrections affect the following variables in the Redacted Open-Ended Responses file: MIPpers1, MIPpers2, MIPpolit1, MIPpolit2.For each of the variables mentioned above from the CSES module of the 2008 ANES Time Series study, beginning with caseID number 583, the redacted open-ended responses were not aligned with the correct caseID number in their respective excel worksheets.Each of the worksheets have now been reordered so that each response corresponds to the correct caseID number as found in the original, unredacted file. Download the revised file here.

September 27, 2011
In the post-election media module, for respondents assigned to administration of codes in reverse order, corrections are necessary for both OLD and NEW interest in campaign variables (V085001a/V085001b), and for all attention variables (OLD and NEW). For these cases, valid responses were coded in reverse order. The variables may be corrected using the following recode:

if V084401 eq 2 then do;
 if V085001a eq 1 then V085001a=5; if V085001a eq 5 then V085001a=1;
 if V085001b eq 5 then V085001b=1; if V085001b eq 4 then V085001b=2; if V085001b eq 2 then V085001b=4;if V085001b eq 1 then V085001b=5;
 if V085003  eq 5 then V085003 =1; if V085003  eq 4 then V085003 =2; if V085003  eq 2 then V085003 =4;if V085003  eq 1 then V085003 =5;
 if V085005  eq 5 then V085005 =1; if V085005  eq 4 then V085005 =2; if V085005  eq 2 then V085005 =4;if V085005  eq 1 then V085005 =5;
 if V085007  eq 5 then V085007 =1; if V085007  eq 4 then V085007 =2; if V085007  eq 2 then V085007 =4;if V085007  eq 1 then V085007 =5;
 if V085009  eq 5 then V085009 =1; if V085009  eq 4 then V085009 =2; if V085009  eq 2 then V085009 =4;if V085009  eq 1 then V085009 =5;
 if V085011  eq 5 then V085011 =1; if V085011  eq 4 then V085011 =2; if V085011  eq 2 then V085011 =4;if V085011  eq 1 then V085011 =5;
 if V085012  eq 5 then V085012 =1; if V085012  eq 4 then V085012 =2; if V085012  eq 2 then V085012 =4;if V085012  eq 1 then V085012 =5;
 if V085014  eq 5 then V085014 =1; if V085014  eq 4 then V085014 =2; if V085014  eq 2 then V085014 =4;if V085014  eq 1 then V085014 =5;
 if V085016  eq 5 then V085016 =1; if V085016  eq 4 then V085016 =2; if V085016  eq 2 then V085016 =4;if V085016  eq 1 then V085016 =5;
 if V085018  eq 5 then V085018 =1; if V085018  eq 4 then V085018 =2; if V085018  eq 2 then V085018 =4;if V085018  eq 1 then V085018 =5;
 if V085020  eq 5 then V085020 =1; if V085020  eq 4 then V085020 =2; if V085020  eq 2 then V085020 =4;if V085020  eq 1 then V085020 =5;
 if V085022  eq 5 then V085022 =1; if V085022  eq 4 then V085022 =2; if V085022  eq 2 then V085022 =4;if V085022  eq 1 then V085022 =5;
 if V085023  eq 5 then V085023 =1; if V085023  eq 4 then V085023 =2; if V085023  eq 2 then V085023 =4;if V085023  eq 1 then V085023 =5;
 if V085024  eq 5 then V085024 =1; if V085024  eq 4 then V085024 =2; if V085024  eq 2 then V085024 =4;if V085024  eq 1 then V085024 =5;
end;

however corrected data are also available in this file .

Apri 12, 2011
Correction to V085024 data.

March 2, 2011
CODEBOOK ONLY:

V082415 - cross-references ("REFERENCE") should read:
     PreRandom14. (VERSION E/F economy)
     F2b1. - F2b1a. (VERSION F economy questions - Democratic candidate)
     F2b2. - F2b2a. (VERSION F economy questions - Republican candidate)
V081210, V081214a, V081214b - the confidential missing value is blank [not -3]
V082003b - a note should appear:
        "The IW for case 0124 began late on the day before the 2008 elections
         and finished a little after midnight; it thus has a value of 0."
V083267 - the note should read: "2008 State and Country Master Code"
V083271a - the noted range should be: 0-7 [not 1-7]
V085024 - a note should appear indicating that VOL code 5 ("5. None of the time [VOL]")
          did not appear onscreen in the instrument, resulting in no cases coded 5.
V085201e - var label should be: "R4e. DHS: Has R ever: given money to religious organization"

DATA AND CODEBOOK:

V082415 - code labels should be:
    1. Democratic Presidential cand is 1st - F2b1,F2b1a
    2. Republican Presidential cand is 1st - F2b2,F2b2a
V083094b - var label should be: "G2b. [VERSION Q] Party more likely to involve US in war"
V083108x,3110x,3111x,3115x,3117x,3118x - code 5 label should be: "5. Slightly fewer" [not "..Slight.."]
V083136 - var label should be: "N4h2. [NEW] Rep pres cand favor/oppose citizenship illegals" [not "..Dem.."]
V083137 - var label should be: "N5a. Govt assistance to blacks scale: self-placement"
V083139a - var label should be: "N5c2. Aid to blacks Placement: Dem Pres cand"
V083139b - var label should be: "N5c2. Aid to blacks Placement: Rep Pres cand"
V083265b - age range in the code labels should read: "11-17" [not "10-17"]
V084408 - var label should be: "PostRandom.8. Order House cand liberal-conservative"
V084502a - code 34 label should be:  "34. Republican House incumbent" [not "..challenger"]
V085101c,5102c - valid codes should be: 1,2,3 [not 1,3,5]
V085117 - var label should be: "H10. Hope the U.S. has a woman president in R's lifetime" [not "..woman present.."]
V085166 - var label should be: "N2e. Not a big problem if some have more chance in life" [not "..problem is.."]

DATA ONLY:

V083184 - code 3 label is missing:  "3. The Bible is a book written by men and is not the word"
V083257 - code labels are missing :
  "1. Both parents born in the U. S."
  "2. One parent born in the U. S."
  "3. Neither parent born in the U. S."
V083258 - code 1 label should be: "1. Mainland U. S."
V083260 - code 23 label is missing: "23. U. S. A."
V085088-V085091b - code 1 label is missing: "1. By law, abortion should never be permitted."
V085291 - code 23 label is missing: "23. U.S.A."
February 24, 2011
This file  provides the correct data for variable V081120 (Screener IWR ID). The data in the uncorrected variable do not represent interviewer ID for any case.

June 3, 2010
Code labels for summary variable V085103x have been reversed. Valid codes should be:

	1. Should discourage a great deal
	2. Should discourage only a little
	3. Should stay out of this matter
	4. Should encourage only a little
	5. Should encourage a great deal
February 3, 2010
Case number 0262 has been dropped as an incomplete. Case 0262 was pre-only (no post-election reinterview). The total number of interviews for the ANES 2008 Time Series Study is now 2,322.
December 9, 2009
For Stata users, the raw (ASCII) data file (anes2008TS_dat.txt) should have a hard return character added to the end of the file. This file is used with downloaded Stata setup (.do and .dct) files to create a Stata system file.
October 20, 2009
The strata variable  (V081206) for the 2008 Time Series sample is provided with actual values (1-9) scrambled.

September 16, 2009

  1. Corrected data  are available for the randomized Pre-election federal spending items V083141-V083151x. This data may be merged using case id variable V080001.
  2. In the Post-election media modules (OLD media module: V085001a through V085012; NEW media module: V085013 through V085024), for cases administered in ‘reverse’ order the coded data must be reversed, as follows (SAS code):
       if V082401 eq 2 then do ;
         if V085003 gt 0 then V085003 = abs(V085003 -6) ;
         if V085005 gt 0 then V085005 = abs(V085005 -6) ;
         if V085007 gt 0 then V085007 = abs(V085007 -6) ;
         if V085009 gt 0 then V085009 = abs(V085009 -6) ;
         if V085011 gt 0 then V085011 = abs(V085011 -6) ;
         if V085012 gt 0 then V085012 = abs(V085012 -6) ;
         if V085014 gt 0 then V085014 = abs(V085014 -6) ;
         if V085016 gt 0 then V085016 = abs(V085016 -6) ;
         if V085018 gt 0 then V085018 = abs(V085018 -6) ;
         if V085020 gt 0 then V085020 = abs(V085020 -6) ;
         if V085022 gt 0 then V085022 = abs(V085022 -6) ;
         if V085023 gt 0 then V085023 = abs(V085023 -6) ;
         if V085024 gt 0 then V085024 = abs(V085024 -6) ;
      end ;

July 8, 2009
Coded data  and documentation  for respondent religious identification summary variables are now available for:

V083185x Xx1. SUMMARY: R religious identification 
V083185a Xx1a. SUMMARY: R denomination 
V083185b Xx1b. SUMMARY: R major religion group 

These variables summarize response data from the religious identification module X1-X6bx. Code labels for SAS, STATA, and SPSS are also provided here .

June 23, 2009
It has been discovered that the Version K pre-election traits variables V083100a-V083100g, V083102a-V083102g, and V083179a-V083179g require correction. Corrected data are available here  for:

      V083100a  K1b1. Pres Dem cand trait moral [VERSION K]
      V083100b  K1b2. Pres Dem cand trait strong leadership [VERSION K]
      V083100c  K1b3. Pres Dem cand trait really cares[VERSION K]
      V083100d  K1b4. Pres Dem cand trait knowledgeable [VERSION K]
      V083100e  K1b5. Pres Dem cand trait intelligent [VERSION K]
      V083100f  K1b6. Pres Dem cand trait honest [VERSION K]
      V083100g  K1b7. Pres Dem cand trait optimistic [VERSION K]
      V083102a  K2b1. Pres Rep cand trait moral [VERSION K]
      V083102b  K2b2. Pres Rep cand trait strong leadership [VERSION K]
      V083102c  K2b3. Pres Rep cand trait really cares [VERSION K]
      V083102d  K2b4. Pres Rep cand trait knowledgeable [VERSION K]
      V083102e  K2b5. Pres Rep cand trait intelligent [VERSION K]
      V083102f  K2b6. Pres Rep cand trait honest [VERSION K]
      V083102g  K2b7. Pres Rep cand trait optimistic [VERSION K]
      V083179a  R2b1. President trait: moral [VERSION K]
      V083179b  R2b2. President trait: strong leadership [VERSION K]
      V083179c  R2b3. President trait: really cares [VERSION K]
      V083179d  R2b4. President trait: knowledgeable [VERSION K]
      V083179e  R2b5. President trait: intelligent [VERSION K]
      V083179f  R2b6. President trait: honest [VERSION K]
      V083179g  R2b7. President trait: optimistic [VERSION K]

June 11, 2009
Coded data  and documentation  for respondent (self-identification) ethnicity mentions from the Pre-election survey are now available for the following variables:

 V083252a  Y25a. Main ethnic or nationality group - Mention 1
      V083252b  Y25b. Main ethnic or nationality group - Mention 2
      V083252c  Y25c. Main ethnic or nationality group - Mention 3
      V083252d  Y25d. Main ethnic or nationality group - Mention 4
      V083252x  Y25x. SUMMARY COUNT for: Main ethnic or nationality group
      V083253   Y26. IWR CHECKPOINT: number of ethnic mentions
      V083254   Y26a. Ethnic group most close
      V083255   Y27. IWR CHECKPOINT: MENTION ETHNIC HISPANIC
      V083256   Y28a. No ethnic mention Hispanic: is R of Hispanic
      V083256a  Y28b. Hispanic: type Hispanic

As a result of the ethnicity coding and review, errors in interviewer checkpoints (Y26,Y27) have been identified, and the Post-election Hispanic status summary variable (V085027) has been updated. It should be noted that code labels for some values have been revised as indicated in the documentation file.

May 29, 2009

For districts LA01 and OH03, House candidate codes were incorrectly identified as follows: In LA01, the incorrect name was given for the House incumbent in V084504 and the incumbent status (34) was incorrectly coded as 42 in V084504a; the Republican candidate code (34) was incorrectly coded as 32 in V084502a and the Democratic candidate code (35) was incorrectly coded as 31 in V084501a. In addition, the TYPE RACE variable, V084513, for LA01 should be coded 21. Because the incorrect incumbent name was preloaded in LA01, the incumbent evaluations in variables V085068, V085068x, and V085069 should be recoded to -4 “NA”. Followups in variables V085068a and V085068b should be recoded to -1 “INAP”. In OH03, the incumbent (34) was coded as 33 in V084504a; the Republican candidate (34) was incorrectly coded as 36 in V084502a, and the Democratic challenger (35) was incorrectly coded as 33 in V084501a. Furthermore, the TYPE RACE variable, V084513, for OH03 should be coded 21. Corrections (SAS code) should be made as follows:

      DISTRICT LA01
	   V084501a = 35;
	   V084502a = 34;
	   V084504 = "Stephen Scalise";
	   V084504a = 34;
	   V084513 = 21;
	   if V081202a eq "LA01" and V085050x eq 31 then V08050x = 35;
	   if V081202a eq "LA01" and V085050x eq 32 then V08050x = 34;
	   if V081202a eq "LA01" V085068 then do;
	   	V085068 = -4;
	   	V085068a = -1;
	   	V085068b = -1;
	   	V085068x = -4;
	   	V085069 = -4;
	   end;

	  DISTRICT OH03
	   V084501a = 35;
	   V084502a = 34;
	   V084504a = 34;
	   V084513 = 21;
	   if V081202a eq "OH03" and V085050x eq 33 then V08050x = 35;
	   if V081202a eq "OH03" and V085050x eq 36 then V08050x = 34;
	   end;
May 24, 2009
The codes for John McCain and Barack Obama in variables V085197a-V085197c (candidates for whom the respondent would never vote) have been reversed in the documentation. John McCain should be code 1 and Barack Obama should be code 5.
May 22, 2009
The post-election codebook documentation for variable pairs V085188a/V085188b, and V085190a/V085190b has reversed the documentation for the 2 candidates. V085188a ‘Like-Dislike’ and V085190a ‘Left-Right’ scale data should correspond with respondent evaluation of McCain; V085188b and V085190b ‘Left-Right’ scale data should correspond with respondent evaluation of Obama.
May 14, 2009
Data  and documentation  for the summary party identification variable V083098x are now available.

pavement-enterprise