| agg_mit |
An additive scale of aggravating
and mitigating circumstances. Each of 16 possible aggravating circumstances
that was present in the federal habeas case was assigned a value of
1. Each of 12 possible mitigating circumstances that was present in
the case was assigned a value of -1. This variable is the sum of those
aggravating and mitigating circumstance values. |
|
bltot, bltotst
|
The number of death verdicts
imposed in the state in the relevant year that were awaiting court
review (i.e., were backlogged) as of the end of the study period.
For Analyses 16 and 17, bltotst is the state's mean
capital backlog during the study period. |
|
blkpctst
|
The average percent of the
state's population comprised of African-Americans during the study
period. |
| claimno |
The number of claims (grounds for relief) presented
by the habeas petitioner at the final stage of federal habeas review
at which there was a decision on the merits.
|
| cntydor2 |
The county's number of death verdicts per
1000 homicides. Homicides are taken as a two-year average of the
year of the death sentence and the prior year. When there are no
death sentences in the year of the death sentence, or the prior
year, the average is lagged by a year. For Analyses 16 and
17, cntydor2 is the county's mean number of death verdicts
per 1000 homicides (calculated as described above) during the study
period.
|
|
dir_exrt
|
The state's direct expenditures
on its court system per 100,000 population. |
|
dlos_ffd
|
A binary measure, coded 1 for
federal habeas petitioners represented by a lawyer from out of state
at the final stage of federal habeas review at which there was a decision
on the merits, and coded 0 for petitioners not represented by an out-of-state
lawyer at that stage. |
| dor, dorst |
The state's number of death
verdicts per 1000 homicides. For Analyses 16 and 17,
dorst is each state's mean number of death verdicts per 1000 homicides
during the study period. |
| dswvrt |
The state's number of death
verdicts imposed for offenses against at least one white victim per
100 white homicide victims. |
| fac_csld |
A principal component factor (combined
measure) of the state's rate of court case backlogs (civil cases,
criminal cases, felony cases, total cases (rgrtbklg)) awaiting decision
in the state's courts. Each component of the factor is a static
measure estimated as filings minus dispositions averaged over the
10-year period (1985-1995) in which data are available. This
variable measures court backlogs caused by all, not simply capital
or criminal, cases.
|
| facvic2l |
&#A principal component factor (combined
measure) of the log of the state's number of death verdicts imposed
for offenses against at least one white victim per 100 white homicide
victims (dswvrt), and the log of the percent of death verdicts imposed
for homicides against at least one white victim divided by the percentage
of all homicide victims in the state who are white (wvdsst).
|
| fac_welf |
&#A principal component factor (combined
measure) of the number of welfare recipients per 100,000 population,
and the amount of expenditures on welfare per 100,000 population.
|
| feh2 |
A binary measure coded 1 for federal
habeas cases in which a federal evidentiary hearing was held prior
to final federal habeas review, and coded 0 for federal habeas cases
in which no federal evidentiary hearing was held.
|
| hrst |
&#The state's number of homicides per
100,000 population.
|
| hr2 |
The county's number of homicides per
100,000 population.
|
| pcbl |
The number of death verdicts imposed
in the state in the relevant year that had not been finally reviewed
at the state post-conviction stage as of the end of the study period.
For the reasons discussed in Appendix C, pp. C-1 to C-2, this variable
is calculated as: (death verdicts that were not finally reviewed
at any of the three review stages + death verdicts that had been
reviewed and had been approved on state direct appeal and were available
for state post-conviction review) ! (death verdicts reversed at
the state post-conviction stage).
|
| ofvcindx |
An additive scale of seven elements relating
to the character of the defendant and traits of the victim in a
capital federal habeas caseCthat the defendant had a prior criminal
record; the defendant had a history of alcohol abuse; the defendant
had a history of drug abuse; the defendant was intoxicated at the
time the crime was committed; the defendant had a connection to
the community; the victim had one or more of a list of prominent
statuses within the community (police officer, fire fighter, elected
official, etc.); the victim was female. The value assigned each
case is the number of the seven elements present in the case.
|
| pajid |
This party-adjusted judge ideology
score, used in one of our analyses, is a combined measure of state
supreme court justices' liberal versus conservative decision making
that was developed by Paul Brace, Laura Langer & Melinda Hall.
See Measuring the Preferences of State Supreme Court Judges,
62 J. of Politics 387 (2000). The variable compares states based
on the party affiliations of their state supreme court justices
and based on measures of the ideological disposition of the electorate
(for states where judges are elected) and of elite portions of the
population (where judges are appointed). The measure is a state
mean from 1970 to 1993, scored from conservative to liberal.
|
| pctbdiffnew |
&#The percent of defendants sentenced to
death by the state who are African-American minus the percent of
the state's population that is comprised of African-Americans (plus
1, to assure there are no negative values).
|
| pctblack, pctblkst |
&#The percent of the state's population comprised
of African-Americans.
|
| pctblk |
&#The percent of the county's population
comprised of African-Americans. For Analyses 16 and 17,
pctblk is the mean of that percent during the study period.
|
| pnindx |
&#The ratio of the number of inmates incarcerated
in prison in the state in the relevant year per 100 FBI Index Crimes
committed in the state in that year. This is measure of the rate
at which serious criminals are apprehended, convicted and incarcerated
in the state. For Analyses 16 and 17, pnindx is the
mean of this ratio during the study period.
|
|
ppindx, ppindx2
|
Political pressure indexes
are two similar measures of the extent to which state judges are subject
to political discipline for their rulings through political or electoral
politics. The indexes include measures of the way in which judges
are selected and retained, and the length of the first term. Two indices
were created. The first index combines the way in which judges are
selected, the way they are retained, and the length of the first term.
Selection method consists of a scale of 1 to 4, with 1 being the least
political method and 4 being the most political, with scores based
on the appointing authority (legislature, governor), whether the appointment
is subject to retention elections, or whether elections without appointments
are used. Retention is a binary nominal category coded 1 for appointed
judges who face constitutionally mandated retention votes, and zero
for all other judges. Length of first term scales years from 1 to
4 with the assumption that longer terms diminish political pressure.
Years are categorized based on frequency distributions of term lengths.
Terms of 10 to 15 years are categorized as 1; 8 years is categorized
as 2; 6 years is categorized as 3; and from 1 to 4 years is categorized
as 4. In order to account for the short duration of many appointments,
a second scale was used based on the length of judges= first elected
term, or the longer of retention terms. For example, an appointed
first term of 1 year followed by an election term of 15 years is considered
a 15 year first term, and scaled as a 1 to reflect lower political
pressure. |
| prednew |
A predicted value for each state and year that
is assigned to each county and year in two county-level analyses.
The predicted value is derived from a parallel state-level analysis:
Each county in the relevant state and year is assigned a value representing
the amount of reversible error predicted for that state and year
by the various explanatory factors in the state-level analysis.
Prednew measures how much variance in the amount of error found
in death verdicts from one county and year to the next can
be explained by the amount of error that the state-level analysis
predicts for the relevant state and year.
|
| psst |
&#A principal component factor of the
log of the state's population and the log of the state's population
density. This is a measure of the state's population structure.
For Analyses 16 and 17, psst is the mean of this factor
during the study period.
|
| ps |
A principal component factor of the
log of the county's population and the log of the county's population
density. This is a measure of the county's population structure.
For Analyses 16 and 17, ps is the mean of this factor
during the study period.
|
| repmaj |
A binary variable coded 1 for federal
habeas cases in which a majority of the reviewing judges at the
final stage of federal habeas review were appointed by Republican
Presidents, and 0 otherwise.
|
| rgrtbklg |
&#The rate per 1000 population of cases
awaiting decision in the state=s court system. This is a static
measure, estimated as filings minus the dispositions averaged over
the 10-year period (1985-1995) in which data are available. This
is one of the four measures of general court backlogs in fac_csld.
|
| seh2 |
A binary measure coded 1 for federal
habeas cases in which a state evidentiary hearing was held on a
claim later raised in the federal habeas proceeding, and coded 0
for federal habeas cases in which no such state evidentiary hearing
was previously held.
|
| sentyr |
The year the death verdict was imposed.
|
|
stblk1
|
The mean of the percent of
the state's population comprised of African Americans during the study
period. |
| wbrtst |
The state's rate of white homicide
victims per 100,000 white residents divided by its rate of black victims
per 100,000 black residents. This is a measure of how the threat of
homicides in the state is distributed between whites and blacks. |
| wbrt |
The county's rate of white homicide
victims per 100,000 white residents divided by its rate of black
victims per 100,000 black residents. For Analyses 16 and
17, wbrt is the mean of this rate during the study period.
|
| whblrts |
The average number of white
victims in the state per 100,000 white population divided by the average
number of black victims in the state per 100,000 black population
during the study period. |
| wvdsst |
The percent of death verdicts
imposed by the state for homicides against at least one white victim
divided by the percentage of all homicide victims in the state who
are white. |
| wvrtst |
&#The state's rate of white homicide victims
per 100,000 white residents.
|
| year |
The year in which the death verdict
being studied was imposed.
|
| yearn |
The 23 years in the study period, taken
as a linear trend from the beginning to the end of the study period.
|
