INTRODUCTION

          The death penalty continues to be a major social debate in the United States.  The issue was recently brought to light again in a rather peculiar way just a few weeks ago with the daring escape of death row inmate, Martin Gurule, from the Huntsville prison in Texas.  The nation waited for days as the authorities searched for Gurule, who eventually turned up dead in a near by pond. The tactics of allowing death row inmates to engage in prison work programs was called into question.  This very incident also raised from the depths the over all issue of the death penalty.
           The United States has, by and large, always been a pro-death penalty country.  U.S. citizens have usually wanted justice, at least in regards to murder, to be swift and stiff.  The question then arises;  Who supports the death penalty?  What is the key ingredient to the kind of people who support the death penalty compared to those who do not?  The debate over the morality of the death penalty is most visibly seen in two arenas.  The first is the political arena and the second is the religious arena.
          Political affiliation seems to have a lot to do with whether one supports the death penalty or not.  Logically speaking from a general, socially observable point of view those who are more conservative seem to support the death penalty more than those who are more liberal.  The terms liberal and conservative, at least in the last half of the twentieth century in U.S. politics, have been applied to the Democrats and Republicans in that order.  Meaning that Democrats are viewed as a more liberal, thus by conventional logic would be less likely to support the death penalty in regards to murder.  The Republicans are viewed as being more conservative, thus by conventional logic would be more likely to support the death penalty in regards to murder.  This general political party view of the death penalty has generally held up in the past few decades.
         
One of the best ways to see party affiliation is usually through presidential support during a presidential election.  The country had one of its most interesting elections in 1992 with the running of three major candidates.  Clinton ran for the Democrats, usually thought to be a more liberal party.  Bush ran for the Republicans, usually thought to be a more conservative party.  Therefore, it seems reasonable to say that the differences in support for the death penalty in comparison of Clinton voters to Bush voters would be noticeable.  Logically, Clinton voters should be more opposed to the death penalty than are Bush voters.
         
However, in 1992 H. Ross Perot was also able to mount a significant campaign for the presidency.  The conservative/liberal spectrum for Perot voters was a hazy one.  In hindsight, analysis of the election results showed that the Perot voters took away many of Bush’s more conservative voters.  It is predicted that Perot’s taking of Bush’s more conservative supporters cost Bush the election.  In this regard Perot voters can be seen as even more conservative than Bush’s voters, thus more likely than the other two groups to strongly support the death penalty.
         
The second area often talked about with regard to people’s views on the death penalty is religion.  It is true that people’s religious preferences often determine their views on social issues.  This “social fact” appears to be prevalent throughout the world, from Amish views on the use of electricity to Catholic and Southern Baptist views on drinking.  The death penalty should be no exception.  
          There are three main religious groups in the U.S. that can be looked at for views on the death penalty.  These groups are Protestants, Catholics, and Jews.  In general, Protestants seem to be more conservative in their religious views.  Often this group throughout U.S. history is related to conservative causes such as the puritan beginnings of America, on through the Temperance movement, and on up to political activist groups like the Christian Coalition that speak for large numbers of Protestants.  The South itself is often seen as a place where Bible beating, fire and brimstone Protestant preachers expound their sermons on Sundays.  In fact, the Bible Belt is defined by various Protestant denominations located within the southern region. Protestants, much more than Catholics, seem to be involved in fundamentalist thinking that leads to stricter more conservative views.  Thus, given this heritage of American Protestant conservatism, it would be assumed that a large number of Protestants would support the death penalty for convicted murder.
          On the other hand, the Roman Catholic Church officially is opposed to the death penalty in all circumstances.  Interestingly enough the Missouri Supreme Court delayed the execution of Darrell Mease due to the Pope’s visit to St. Louis in January for a youth rally.  The execution was moved from January 27 to February 10 (John Bacon, USA Today, Thursday, Dec. 3, 1998, p3A EXECUTION DELAYED).  The Catholic Church has a strong pro-life stance on all social issues in regard to abortion, euthanasia, and even the death penalty.  Thus it would be assumed, with much logic, that Catholics would be less likely to support the death penalty.  The American Jewish community appears to be more fickle in regards to a definite response to the death penalty.  Jewish people are often seen as either split into very conservative Orthodox Jews to very liberal Reform Jews.  It would then be assumed that the stance on the death penalty by the Jewish community would be equally fickle and possibly divided pretty evenly among those who oppose and those who favor.

Literature Review

          No relevant sociological literature could be found on the topic of the 1992 voting behavior in relation to views on capital punishment.  However, the topic of religious preference and views on capital punishment bore a bit more fruit.  Articles “Religious Orientation, Race and Support for the Death Penalty” and “Unlikely Alliances: The Changing Contours of American Religious Faith” provided little relevant material to the specific defined topics that are dealt with in this research.
          In 1992 nearly 80% of the American public supported the death penalty.  This same year Robert L. Young published his findings of religious preference and views on capital punishment in the article “Religious Orientation, Race and Support for the Death Penalty” for the Journal for the Scientific Study of Religion.  His study was limited to Fundamentalists and Evangelicals and their support of the death penalty.  Fundamentalists were defined as those who had a literal interpretation of the Bible.  They were regarded as having an absolutist point of view and thus were considered to be more conservative.  This all pointed to the belief that they would be more in favor of the death penalty.  The evangelicals were defined as those who proselytized to others.  They were regarded as having compassion, since they had to reach out to others more frequently.  This all pointed to the belief that the evangelicals would be more liberal and less likely to favor the death penalty.  Young’s research showed these two basic assumptions to be true.  Fundamentalist’s had a positive coefficient of 0.1742 with regards to supporting the death penalty.  Evangelicals had a negative coefficient of -0.2028 with regards to supporting the death penalty.
         
The second research article entitled “Unlikely Alliances: The Changing Contours of American Religious Faith” by James Davison Hunter and John Steadman Rice for the book America at Century’s End by editor Alan Wolfe contained very little relevant information on the topic of religious preference and views on the death penalty.  It focused more on divisions between liberal and conservative members of religious denominations and less about the denominations themselves.  However, the piece did seem to suggest that Catholics remain more liberal on a variety of social issues than Protestants.  All in all, the material found on either presidential voting in 1992 or religious preference in regard to views on the death penalty were hard to come by.  That is why I propose the two following hypothesis to check within my analysis:

  1. The choice of a respondent’s political party presidential candidate will help determine the respondent’s views on the death penalty for convicted murder.  I theorize that those who voted for Clinton will be less in favor of the death penalty.  Those who voted for Bush will be more in favor of the death penalty than Clinton voters.  While those who voted for Perot will highly favor the death penalty.
  2. The second hypothesis is that the religious preference of the respondent will determine the respondent’s views on the death penalty for convicted murder.  I theorize that Protestants will be highly in favor of the death penalty.  Catholics will be highly opposed to the death penalty. While Jews will be fairly evenly divided on the issue.

Methods

          This analysis utilizes interview data collected by the National Opinion Research Center (NORC) in the 1994 General Social Survey (hereafter GSS).  The GSS, a nationwide annual survey, offers the advantage of multi-stage probability sampling and can be considered representative of English-speaking, non-institutionalized adults (18 years of age and older) living in U.S. households.  (For more detailed information on the GSS, see Babbie and Halley [1994].)  This examination of the relationships between Favor Or Oppose Death Penally for Murder (CAPPUN), Vote For Clinton, Bush, Perot (PRES92), and RS Religious Preference (RELIG) relies on a subset of 514 of the 2992 original respondents.  The data extract includes only questions asked on both interview ballots B and C for Version 2 of the 1994 GSS.  This provides the researcher with a continuous set of questions with a lower number of missing cases; however, the trade-off is the lower the number of total cases.  Following is a brief description of the variables considered and of the frequency distributions for these variables.
         
CAPPUN is the dependent variable that looks at the favorability or opposition to the death penalty for murder.  The variable’s exact wording is as such, “Do you favor or oppose the death penalty for persons convicted of murder?” The responses vary from 1. Favor, 2. Oppose, 8. Don’t Know (DK), or 9. No Answer (NA).  The missing statements of DK and NA answers were withdrawn from the computations through the use of missing values so that only 1. Favor and 2. Oppose remained.  This created a total of 892 valid responses with 79.0% (705) in favor and 2 1.0% (187) opposed to the death penalty for convicted murder.
          PRES92 is an independent variable that looks at whom the respondents voted for President of the United States in the 1992 elections, if they voted in that election.  The variable’s exact wording is as such, “If voted in 1992: Did you vote for Clinton, Bush, or Perot?” The responses vary from 1. Clinton, 2. Bush, 3. Perot, 4. Other, 6. No Pres. Vote, 8. Don’t Know (DK), or 9. No Answer (NA).  The responses 4. Other and 6. No Pres. Vote were recoded into the response 4. Other.  Then the missing statements of DK and NA were withdrawn from the computations through the use of missing values along with the recoded 4. Other so that only the three main presidential candidates would remain.  This created a total of 617 valid responses in which 49.9% (308) of the 1992 voting respondents voted for Clinton, 34.7% (214) voted for Bush, and 15.4% (95) voted for Perot.
         
RELIG is an independent variable that looks at the respondent’s religious preference. The variables exact wording is as such, “What is your religious preference? Is it Protestant, Catholic, Jewish, some other religion, or no religion?”  The responses vary from 1. Protestant, 2. Catholic, 3. Jewish, 4. None, 5. Other, or 6. No Answer (NA).  The statements 4. None and 5. Other were withdrawn from computations through the use of missing values along with 6. NA so that the three main U.S. religious denominations remained.  This created a total of 833 valid responses in which 65.7% (547) respondents claimed to be Protestant, 31.1% (259) claimed to be Catholic, and 3.2% (27) claimed to be Jewish.
         
The dependent variable CAPPUN will be crossed through crosstabulation with the independent variable PRES92 in order to determine the relationship between respondent’s presidential vote in 1992 and their views on capital punishment.  The dependent variable CAPPUN will then be crossed through crosstabulation with the independent variable RELIG in order to determine the relationship between religious preference and views on capital punishment of the respondents.  Then all three, CAPPUN (dependent), PRES92 (independent), and RELIG (independent) will be crossed through crosstabulations.  This will be done in order to determine weather or not the relationship between CAPPUN and PRES92 is spurious and if the religious preference of the respondent was in fact more of a cause for their presidential vote and thus their views on capital punishment. RELIG will be used as the control variable.  Finally a regression analysis will be run in order to further check the relation between PRES92, RELIG, a new variable OWNGUN, and the dependent variable CAPPUN.

Findings/Results

          The results for the bivariate cross between CAPPUN Favor Or Oppose Death Penally for Murder and PRES92 Vote For Clinton, Bush, Perot show that my first hypothesis may have statistical support.  Table 1 shows that 71.9% of those who voted for Clinton support the death penalty.  However, those who voted for Bush are shown to support the death penalty even more so by 85.1%.  Even those who supported the death penalty among Perot’s voters seem to confirm my hypothesis at 88.8%.  The relationship between respondent’s selection in the 1992 presidential race and their views on the death penalty seem to be significant with a Chi-Square of 18.250 and with an Asymp. Sig. 0.000.  The strength of this relationship, however, could not be determined due to the weakness of the lambda statistic.

Table 1.  Bivariate Analysis of Capital Punishment Attitude (CAPPUN) by 1992 Presidential Voting Behavior (PRES92).


PRES92
Voted for Clinton, Bush, or Perot in 1992?
   
CLINTON
BUSH
PEROT
TOTAL
CAPPUN
FAVOR
205
171
79
455
Favor or oppose death penalty for murder
71.9%
85.1%
88.8%

79.1%
OPPOSE
80
28.1%
30
14.9%
10
11.2%
120
20.9%
TOTAL
285
100.0%
201
100.0%
89
100.0%
575
100.0%
Chi Sq=
18.250
       
Asymp. Sig.=
0.000
       
Lambda
0.000
       

          The results for the bivariate cross between CAPPUN Favor Or Oppose Death Penalty for Murder and RELIG R'S Religious Preference show that my second hypothesis may not have statistical support. Table II shows that 79.9% of Protestants support the death penalty.  However, my second hypothesis seems to sink when practically the same amount of Catholics, 79.3%, also support the death penalty.  The Jewish respondents seemed less ambiguous than I had hypothesized with an overwhelming 8 1.5% supporting the death penalty.  The Chi-Square at 0.09 1 and Asymp. Sig. at 0.956 also seem to blow my second hypothesis out of the water.  Of course lambda showed nothing since there was no significance between the variables.

Table 2.  Bivariate Analysis of Capital Punishment Attitude (CAPPUN) by Religion (RELIG).


RELIG
    Respondent's Religion
   
PROTESTANT
CATHOLIC
JEWISH
TOTAL
CAPPUN
FAVOR
405
195
22
622
Favor or oppose death penalty for murder
 
79.9%
79.3%
81.5%
79.7%
OPPOSE
102
20.1%
51
20.7%
5
18.5%
158
20.3%
TOTAL
507
100.0%
246
100.0%
27
100.0%
780
100.0%
Chi Sq=
0.091
       
Asymp. Sig.=
0.956
       
Lambda
0.000
       

          The results for the cross between the dependent variable CAPPUN and the independent variable PRES92 while being controlled for by RELIG were very interesting.  The results of Table III seem to suggest that only the relationship between who the respondent voted for in 1992 and the respondent’s views on the death penalty is significantly affected by their religion when the respondent is Protestant.  The Protestant subgroup for the control variable RELIG shows a Chi-Square of 16.336 and a Asymp. Sig. of 0.000.  The other two subgroups, Catholic and Jewish, did not show such a significant relationship among the bivariate CAPPUN and PRES92 cross. Catholics had a Chi-Square of 0.279 and a Asymp. Sig. of 0.870.  Jews had a Chi-Square of 2.503 and a Asymp. Sig. of 0.286.  Lambda showed nothing for all three religious preferences. This is probably due to Lambda’s weak calculating power.

Table 3.  Bivariate Analysis of Capital Punishment Attitude (CAPPUN) by 1992 Presidential Voting Behavior (PRES92) controlling for Religion (RELIG).


PRES92
PROTESTANT Voted for Clinton, Bush, or Perot in 1992?
   
CLINTON
BUSH
PEROT
TOTAL
CAPPUN
FAVOR
112
105
44
261
Favor or oppose death penalty for murder
70.9%
86.1%
93.6%

79.8%
OPPOSE
46
29.1%
17
13.9%
3
6.4%
66
20.2%
TOTAL
158
100.0%
122
100.0%
47
100.0%
327
100.0%
CATHOLIC
   
CLINTON
BUSH
PEROT
TOTAL
CAPPUN
FAVOR
58
49
21
128
Favor or oppose death penalty for murder
78.4%
81.7%
77.8%

79.5%
OPPOSE
16
21.6%
11
18.3%
6
22.2%
33
20.5%
TOTAL
74
100.0%
60
100.0%
27
100.0%
161
100.0%
JEWISH
   
CLINTON
BUSH
PEROT
TOTAL
CAPPUN
FAVOR
13
7
2
22
Favor or oppose death penalty for murder
76.5%
100.0%
100.0%

84.6%

OPPOSE
4
23.5%
0
0.0%
0
0.0%
4
15.4%
TOTAL
17
100.0%
7
100.0%
2
100.0%
26
100.0%
PROTESTANT CATHOLIC JEWISH
Chi Sq=
16.336
   0.279    2.503
Asymp. Sig.=
0.000
   0.870    0.286
Lambda
0.000
   0.000    0.000

          In order to further analyze the three variables I decided to run a regression analysis.  In order to do so I had to compute PRES92 into two variables titled CLINTON and BUSH.  The subgroup Perot will be considered the Coefficient Constant.  The variable RELIG also had to be computed into two variables titled PROTEST (for Protestants) and CATHOLIC.  The subgroup Jewish will be considered the Coefficient Constant.  The variable OWNGUN Have Gun in Home is also added in order to provide control for the regression analysis.
         
The results of the regression analysis are seen in Table IV. It is suggested that the regression model explains very little since its R Square is merely 0.0 17.  The F Test is 1.542, but the regression model is not significant at a Sig. of 0.175.  The Constant Coefficient for Perot voters and Jewish respondents is 1.001 with favoring the death penalty numbered as 1 and opposing the death penalty numbered as 2.  The B coefficients suggest that Clinton voters are more likely to be opposed to the death penalty with a positive coefficient of 0.04795 (Beta 0.057).  Bush voters are more likely to be in favor of the death penalty then both the Clinton and Perot voters by a negative coefficient of -0.0 1577 (Beta = -0.017).  Protestants are more likely to be opposed to the death penalty than Jews by a positive coefficient of 0.03999 (Beta 0.051).  Catholics are more likely to be opposed to the death penalty than Jews by a positive coefficient of 0.03 897 (Beta 0.659).  Finally, the variable OWNGUN has a positive relationship with the variable CAPPUN.  The positive B coefficient, 0.08488, suggests that if a respondent owns a gun then they are more likely to support the death penalty, just as those who do not own a gun are more likely to oppose the death penalty.  The model, with a Beta of 0.108, also seems to suggest that whether a respondent owns a gun or not has the strongest affect on whether that respondent will support or oppose the death penalty.  This relationship is also significant at 0.025.

Table 4.  Multivariate Regression Analysis of Capital Punishment Attitude on 1992 Presidential Voting Behavior, Religion, and Gun Ownership.


Capital Punishment Attitude

B Beta Significance
Clinton 0.04795 0.057 n.s.
Bush -0.01577 -0.017 n.s.
Protestant 0.03999 0.051 n.s.
Catholic 0.03897 0.044 n.s.
OWNGUN 0.08488 0.108 p=0.025
(Constant) 1.011
R Square 0.017
F Test 1.542 (n.s.)

Discussion and Conclusion

          This study has found that my original first hypothesis seems to have some statistical backing.  The first hypothesis stated that:

  1. The choice of a respondent’s political party presidential candidate will help determine the respondent’s views on the death penalty for convicted murder. I theorize that those who voted for Clinton will be less in favor of the death penalty. Those who voted for Bush will be more in favor of the death penalty than Clinton voters. While those who voted for Perot will highly favor the death penalty.

          The first bivariate cross seemed to confirm.  This statement through the use of Chi-Square and Asymp. Sig. numbers.  I theorize that the reason for this relationship is due to the polarization of political parties in the United States within the last half of the twentieth century.  Each party seems to take the opposite view of the other party on almost every issue.  Democrats have been defined as liberal, thus it is reasonable that those who voted for a Democratic President would hold such similar liberal views on the death penalty.  This is the same reason that those who voted for a Republican President would hold such conservative views on the death penalty.  Also, it is reasonable to say that those who left Bush’s side for a more conservative agenda (Perot voters) would be even more in favor of the death penalty. Ultimately this hypothesis seems to hold some validity to it.
          However, this study also found that my second hypothesis did not have much substance to it. My second hypothesis was as follows:

  1. The religious preference of the respondent will determine the respondent’s views on the death penalty for convicted murder. I theorize that Protestants will be highly in favor of the death penalty. Catholics will be highly opposed to the death penalty. While Jews will be fairly evenly divided on the issue.

          The second bivariate cross seemed to shred this hypothesis up.  I believe that there are two reasons why my hypothesis turned out to be wrong and that the results ultimately showed that the members from all three religions supported the death penalty equally, with Jewish support only slightly higher.
          The first reason why my hypothesis seems to be false is due to the very nature of my initial speculations.  The assumptions that Catholics are less likely to support the death penalty than Protestants while Jews would be ambiguous on the subject are assumptions based primarily on stereotyping.  The only actual factual bases that was used for such speculation was that of the Catholic Church’s open stance against the death penalty.  Protestants are such a large and diverse group of people that it is hard to categorize them into one consensus.  The stereotype is that of a conservative Bible beating preacher, thus there emerges an idea of Protestants having a more conservative stance on social issues over all.  The Jewish community is always shown by the media as either very orthodox, dressed in black, or very liberal.  Thus emerges the idea of Jews as having a split opinion on support of the death penalty because both sides are apparent within our culture.   However, the statistics seem to suggest that these vague stereotypes do not necessarily apply.
          The second reason that the second hypothesis was not supported comes from an idea that was stressed in the articles “Religious Orientation, Race and Support for the Death Penalty” and “Unlikely Alliances: The Changing Contours of American Religious Faith.”  The idea states that every religion is split up into both conservative and liberal camps.  If a member of a certain religion leans more toward the liberal side of that religion than they will more than likely be more liberal in the political arena as well.  This view states that the idea of liberal and conservative is a mindframe that crosses religions and ultimately ends up at the door steps of politics.  In this way a liberal Catholic and liberal Baptist might agree more on social issues than they would with a more conservative member of their own religion.
          Ultimately this analysis has shown that CAPPUN and PRES92 have a significant relationship while CAPPUN and RELIG do not have a significant relationship, though the subgroup Protestant does have a significant control relationship between CAPPUN and PRES92.  With the complex orientation of religious preference among respondents, further studies should be done in order to help determine exactly what is the relationship between religious preference and views on capital punishment.

BIBLIOGRAPHY

Bacon, John, “Execution Delayed,” USA Today, Thursday, Dec. 3, 1998, p3A.

Hunter, James Davison and Rice, John Steadman, “Unlikely Alliances: The Changing Contours of American Religious Faith,” in America at Century's End , Alan Wolfe (ed.), Berkeley: University of California Press, 1991.

Young, Robert L., “Religious Orientation, Race, and Support for the Death Penalty,”  Journal for the Scientific Study of Religion, 31 (March, 1992), 76-87.