Full Set of SPSS Syntax and Data

Several SPSS programs were used to modify and merge a wide range of data for this article. Links to these SPSS syntax files appear below, along with an indication of the datafiles read and written by each syntax file, and a brief mention of the operations performed. These files provide the fullest possible description of the data processing involved in the creation of the final analysis dataset. In order to replicate the entire process leading to the results in the SPPQ paper, these syntax files must be edited to reflect the directory structure on your computer, and the original input files must be downloaded for processing. The syntax files cannot simply be executed sequentially without some manual intervention during or after execution. In particular: ext_auburn requires the manual addition of two justices after extracting the Auburn Multi User Database of Apellate Justices.

If you wish to simply replicate the results in Table 1 without recreating the entire dataset from scratch, return to the replication page and use the fullags.sav and agcases.sav datasets, and accompanying SPSS syntax.

SPSS syntax: ext_songer.sps
- Input Data: cta96.por = Songer Courts of Appeals data available from The Program for Law and Judicial Politics of the Michigan State University Department of Political Science.
- Output Data: CoAdata.sav = SPSS file with case level data.
- Processes:
  - Cutting out data from before 1970.
  - Culling cases which fail to meet threshold issues.
  - Correcting erroneous page length data.
  - Writing out reduced version of Songer data.
SPSS syntax: ext_auburn.sps *manual intervention required* see note below
- Input Data: auburn.dat = text file of MULTI-USER DATABASE ON THE ATTRIBUTES OF U.S. APPEALS COURT JUDGES 1801-1994 available as study s6796 from ICPSR.
- Output Data: auburn.sav = SPSS format version of updated judicial data
- Processes:
  - Reads in text file with auburn data on appelate judges.
  - Updates retirement data for 1996.
  - Note: user must manually pause this program for the manual addition of judges appointed in 1996. Simply running the file does not add those two judges.
  - Writes spss file auburn.sav.
SPSS syntax: panels.sps
- Input Data: CoAdata.sav = data output from ext_songer.sps
  auburn.sav = data from ext_auburn.sps
- Temp data: Note: This syntax produces numerous temporary datafiles.
- Output Data: panels.sav = SPSS data on ideology of panels
- Processes:
  - Extract panel level variables.
  - Remove superfluous records.
  - Remove district judges who sat on appellate panels.
  - Reassign ID numbers to judges who changed circuit. and have 2 id numbers in the songer data, but one in the auburn data.
  - Merge judicial characterisitcs (Auburn) with panel composition (Songer data).
  - Assign W-nominate scores used for judicial ideology.
  - Write out data on ideology of panels.
SPSS syntax: circuits.sps
- Input Data: auburn.sav = data from ext_auburn.sps
- Temp data: Note: This syntax produces numerous temporary datafiles
- Output Data: circuits.sav = SPSS data on ideology of circuits
- Processes:
  - Inserts W-nominate scores used for ideology.
  - Writes out data on circuit ideology
SPSS syntax: SPPQ data merge.sps (A multi step program)

Step 1.
Input Data:
ags.sav Background on attorneys general
naagbudgt.sav AG office budgets
naagstaffandcpi.sav NAAG staff and salary data
(These datasets entered manually from primary sources)
Output Data:
fullags.sav = full set of data on Attorneys General.
Step 2.
Input Data:
CoAdata.sav = case characteristics (from ext_songer.sps).
PANELS.sav = panel characterisitcs
circuits.sav = panel characteristics
Temp Data:
temp.sav = a temporary dataset of case + judge data
AGappel.sav = cases where AG is petitioner.
AGresp.sav = cases where AG is respondent.
Output Data:
AGCASES.sav = full set of case data.
Step 3.
Input Data:
fullags.sav = full set of data on Attorneys General.
AGCASES.sav = full set of case data.
Output Data: analysis.sps = final dataset for statistical analysis.
Processes:
- Merge various datasets about Attorneys general.
- linear interpolation to fill in missing data on staff and salary.
- Merge judicial characteristic data (Auburn) with case data (Songer).
- Create Duplicate records if AG's are on both sides of case.
- create dummy for petitioner or respondent victory.
- create dummy for AG victory or loss.
- create dummy for Criminal cases.
- create circuit dummies.
- recoding state for cases with missing state id.
- Calculate cummulative appearances before the circuit bench.
- Create marker for largest state per circuit.
- Create strong opponent variable
- Create Case complexity variable
- Computing two measures of USSC pro state orientation.
- Create dummy variable for economic cases.
- Create criminal petitioner variable.
- Creating prisoner appelant cases term.
- Create dummy for private party economic appeals.
- Create indicator for policy cases.
- Create measure for first term AG.
- Compress scales to make parameters visible.
- Determine directionality of the petitioner and respondent in each case.
- Identify ideological agreement of state brief and panel.
- Write out data used for LOGIT and Clarify analyses of AG success.