National Social Science Association

National Social Science Association Home
NSSA History
Membership Form
Conferences and Seminars
Publications
Officers and Board Members
Newsletter
New Announcements
Contact NSSA
 
 
 

The Effect of the Marshall Hypothesis
on Attitudes toward the Death Penalty

Alvin D. Mitchell
Winston-SalemStateUniversity

Introduction
The Birth of the Marshall Hypothesis and Research Question
     As a result of the range of flaws seen by some Supreme Court justices inherent in the administration of the death penalty, the Marshall hypothesis was born. The Marshall Hypothesis was developed by Thurgood Marshall, an associate justice on the United States Supreme Court who served from 1967 to 1991. The Marshall Hypothesis “reasoned that public support for the death penalty is a function of lack of knowledge about the subject and if ‘the great mass of citizens’ were provided with information about the death penalty, it would conclude “that the death penalty is immoral and unconstitutional” (Bohm, Vogel, & Maisto, 1993, p.30). Or, in other words, most people support the death penalty because they are ignorant about the death penalty and its effects (Bohm & Aveni, 1985; Ellsworth & Ross, 1983). Therefore, Marshall’s research question is: To What Extent does Knowledge about the Death Penalty Affect Attitude toward the Death Penalty?
Theoretical Framework
     In 1972, Chief Justice Thurgood Marshall argued what he considered to be serious flaws with the issuance of the death penalty. Those flaws ranged from the application of the death penalty being immoral and racist to it being too expensive to administer. In addition, Justice Marshall stated that the death penalty was ineffective in deterring crime, and led to the execution innocent people (Furman v. Georgia, 1972). The following excerpt from Justice Marshall’s opinion presents the basis of his hypothesis:

While a public opinion poll obviously is of some assistance in indicating public acceptance or rejection of a specific penalty, its utility cannot be very great. This is because whether or not a punishment is cruel and unusual depends, not on whether its mere mention "shocks the conscience and sense of justice of the people," but on whether people who were fully informed as to the purposes of the penalty and its liabilities would find the penalty shocking, unjust, and unacceptable. In other words, the question with which we must deal is not whether a substantial proportion of American citizens would today, if polled, opine that capital punishment is barbarously cruel, but whether they would find it to be so in the light of all information presently available. This is not to suggest that with respect to this test of unconstitutionality people are required to act rationally; they are not. With respect to this judgment, a violation of the Eighth Amendment is totally dependent on the predictable subjective, emotional reactions of informed citizens. (408 U.S. 238, 362)

Justice Marshall emphatically believed that American citizens knew almost nothing about the death penalty. He felt that an informed citizenry would come to realize the following:

  • The death penalty did not deter crime and life imprisonment was a more effective deterrent.
  • Convicted murderers were rarely executed; convicted murderers were usually model prisoners, and once released from prison they rarely returned to prison.
  • The cost of executing someone is more expensive than incarcerating that person for life; a person sentenced to death does not perform the duties that “lifers” perform, and this may stimulate the person sentenced to death to commit more crimes (Furman v Georgia, 1972).

Justice Marshall felt that this information would change the opinion of the average citizen but the pivotal problem he would have had was in convincing a person that the death penalty was immoral. This would be difficult to do because society’s attitude toward killers is focused on retribution. Justice Marshall’s explanation for not supporting the death penalty was that the death penalty is barbaric, and serves no purpose in the United States. Instead, he believed Americans should focus on rehabilitation, not revenge.
     Justice Marshall felt it was necessary to let the public know certain facts concerning the death penalty. If the morality, ineffective deterrence, retribution, and other issues mentioned above did not change a persons mind, he surely felt that these facts concerning discrimination and executing innocent people would. In Furman v Georgia (1972), Justice Marshall states this proposition:

I believe that the following facts would serve to convince even the most hesitant of citizens to condemn death as a sanction: capital punishment is imposed discriminatorily against certain identifiable classes of people. There is evidence that innocent people have been executed before their innocence can be proved; and the death penalty wreaks havoc with our entire criminal justice system…Regarding discrimination, it has been said that it is usually the poor, the illiterate, the underprivileged, the member of the minority group - the man who, because he is without means, and is defended by a court-appointed attorney becomes society's sacrificial lamb . . . Indeed, a look at the bare statistics regarding executions is enough to betray much of the discrimination. A total of 3,859 persons have been executed since 1930, of whom 1,751 were white and 2,066 were Negro. Of the executions, 3,334 were for murder; 1,664 of the executed murderers were white and 1,630 were Negro; 455 persons, including 48 whites and 405 Negroes, were executed for rape. It is immediately apparent that Negroes were executed far more often than whites in proportion to their percentage of the population. Studies indicate that while the higher rate of execution among Negroes is partially due to a higher rate of crime, there is evidence of racial discrimination. (408 U.S. 238, 365)

     Marshall discussed in his opinion that discrimination also existed based on gender. His facts included only 32 women having being executed since 1930, while 3,827 men have been executed since 1930, although both men and women have been found guilty for similar murders. He also showed how the poor, the ignorant, and the underprivileged members of our society received the death penalty while the wealthier, better represented did not receive the death penalty. Furthermore, Marshall stressed that the criminal justice system is not fool-proof. Prosecutors live for getting convictions, and they sometimes take immoral measures to complete this task. In addition, some convictions are the result of perjured testimony, mistaken identity, racism, and intimidation. This scenario can possibly lead to innocent people being executed (Furman v Georgia). Lastly, Marshall was so sure that others would feel as he did about capital punishment that he stated the following:

Assuming knowledge of all the facts presently available regarding capital punishment, the average citizen would, in my opinion, find it shocking to his conscience and sense of justice. For this reason alone capital punishment cannot stand. (408 U.S. 238, 370)

Justice Marshall’s Hypothesis is straight to the point: Ignorance influences attitudes. Purpose and Significance of This Study
     The purpose of this study is very similar to Bohm’s purpose (see literature review) to test two of Marshall’s conjectures:

  • The public lacks knowledge of the death penalty and its effects
  • An informed public generally would not support the death penalty (Bohm, Clark, & Aveni 1991).

These conjectures were made in the form of hypotheses by two research teams in two
previous studies (Sarat & Vidmar, 1976; Vidmar, Ellsworth & Ross, 1983).
     Testing the Marshall hypothesis today is extremely important because the United States is currently and annually increasing the number of death row inmates and their executions, and yet ironically commuting more inmates off death row (Allen & Simonensen, 2001). Also, the cases include a disproportionate number of defendants who had white victims and indigent public defenders. This irony raises many questions because executing more inmates while commuting more off death row, and executing convicted murderers of whites at a rate which is disproportionate to executing the killers of blacks, and condemning those who had indigent defenders to death as opposed to those who had paid for lawyers give the appearance that something is wrong with the system. This is a cause for concern-not a determination of guilt. One can argue that there may be procedural or issuance controversy within the application of the death penalty, which is the basis for the Marshall hypothesis (Bohm, Clark, & Aveni 1991). Clearly, this would have an effect on current and future policy making in reference to the death punishment. Nevertheless, a high percentage of the American public still supports the death penalty (Senna & Siegel, 2002).
Literature Review
     There is not an enormous amount of literature on the Marshall hypothesis. However, there is some literature worth noting. Past literature indicated that the Marshall Hypothesis has been tested several times (Bohm, Vogel, & Maisto, 1993; Bohm, Clark, & Aveni, 1991; Bohm, 1989; Bohm, 1991; Ellsworth & Ross, 1983; Sarat & Vidmar, 1976). Three studies that tested the Marshall Hypothesis showed mixed results. All three found that the Marshall hypothesis did not affect opinions about the death penalty overall, but, in some situations, it did. Bohm,Vogel, & Meisto, as well as Lord et al., presented studies that provided little support for the Marshall Hypothesis.
Past Studies
     Sarat & Vidmar (1976) tested the Marshall Hypothesis with a randomly selected sample of 200 residents in Amherst, Massachusetts by using a pre-test survey. The pre-test survey measured their attitudes toward capital punishment. The respondents were given an eighteen statement questionnaire designed to measure their attitudes toward the death penalty both generally and specifically toward certain crimes. A second questionnaire was given to the respondents to test their knowledge about the death penalty. Then, a third questionnaire was given to the respondents to measure their support or opposition to the death penalty. Finally, two series of questionnaires were given to the respondents that asked them to evaluate various aspects of the death penalty and the effects of capital punishment. A seven point Likert scale, ranging from “very strongly favor” to “very strongly oppose” was used to measure the respondents’ attitudes. Not all respondents participated in the survey, which left the researchers with 181 respondents.
     Next, the researchers used their experimental manipulation which consisted of two 1500 word essays, one discussing the Utilitarian approach toward capital punishment, and the other discussing the Humanitarian approach toward capital punishment which measured “knowledge” about the death penalty. These essays were used because the information in them was the heart of Justice Marshall’s argument that an informed person would be inclined to reject capital punishment. The first essay or the Utilitarian approach discussed how according to Justice Marshall, the death penalty was not an effective deterrent to crime. It consisted of summaries of statistical studies, reports of personal experience, arguments on the psychology of deterrence, and data on the recidivism rate of convicted murderers who were released. The second essay, or Humanitarian approach, discussed the way that the death penalty has typically been applied and administered as well as the physical and psychological aspects of execution.
     The respondents were read instructions and told the purpose of the study. Then each respondent was asked to read the essays given to them vigilantly, meticulously, and thoroughly. Then the respondents were given a post-test in which they were asked to fill out another questionnaire measuring their attitudes toward the death penalty and its application and effects. The experimental manipulation was one hour in duration.
     Sarat & Vidmar (1976) found at least two interesting results. First, Sarat & Vidmar found what they assumed: support for the death penalty among their respondents was similar to the support in the American public as a whole. Fifty-four percent of their respondents favored the death penalty, while fifty-nine percent of the American public favored the death penalty. Second, the experimental manipulation or “knowledge” did alter opinions when respondents were given the post test survey. In other words, some respondents did change their attitudes toward the death penalty after they were exposed to information about the death penalty, but not significantly. The largest amount of change was found when the students were exposed to the utilitarian aspect of the death penalty. This was also expected by the researchers.
     Vidmar & Dittenhoffer (1981) revisited the Marshall hypothesis, and gave a Canadian perspective, yet the death penalty had been abolished in Canada since 1976. However, anytime there was a spectacular murder, the House of Commons received a deluge of letters suggesting that the citizens may want the death penalty to return.
     They used 39 undergraduates from the University of Ontario. All were Canadian. Their age ranged from 19 to 32. The female respondents outnumbered the male respondents 26 to 13. The researchers used a 2x2 factorial design with non-random experimental and control groups. These groups were compared at pre-and post-tests. Respondents in the experimental group received $12,000 each, while respondents in the control group received $1.00.
      Instead of using two 1500 words essays, as did the experimental manipulation used by Sarat and Vidmar, Vidmar (1976) and Dittenhoffer (1981) used a 3500 word essay that included much of the humanitarian and utilitarian information used in the earlier study, as well as a summary of some of the main issues in the death penalty debate. In addition, the essay included topics about the death penalty that were not previously covered—deterrence, judicial administration, economics, religion, and moral arguments. Lastly, two books on capital punishment were given: Capital Punishment, by T. Sellin (1967) and The Death Penalty in America, by H. Bedau (1979). The essay and additional readings were considered “knowledge” about the death penalty. The respondents were allowed almost 4 hours to review the material given by the researchers.
     Vidmar and Dittenhoffer (1981) confirmed the findings of previous research done by Sarat and Vidmar (1976) in regard to the Marshall Hypothesis. They, too, find that during the pre-test, respondents lacked knowledge about the death penalty, but once they became informed or knowledgeable, overall support for the death penalty did decrease among respondents. However, the decrease in support was not statistically significant partly due to methodological and generalizability issues concerning the data. First, extreme attitudes, especially those that favored the death penalty strongly, were most resistant to information to counter their position, regardless if the underlying factor for strong support is authoritative, moral or some combination of both. Second, the researchers concluded that the data did not reach significance because of the small sample size used in their study (39 respondents). They could not say with certainty that results from their sample could apply to the general population.
     Ellsworth and Ross (1983) tested the Marshall Hypothesis also. Their study differed from the other researchers (Sarat & Vidmar, 1976; Vidmar & Dittenhoffer, 1981) in that their fundamental purpose was to understand what people mean and what they feel when they say that they favor or oppose capital punishment. In addition, they wanted to determine if the American people were ignorant of factual questions about the use of the death penalty.
     Ellsworth and Ross (1983) randomly sampled 500 northern California residents. Each respondent was mailed a survey and informed by telephone a day prior to the day the researcher was to pick up the survey to insure a better sample size and more completed questionnaires. The researchers set out to measure two questions:

  1. What is the level of general support for the death penalty?
  2. What are the specific basis and implications of retentionists and abolitionists attitudes toward the death penalty based on public opinion polling?

The results of Ellsworth and Ross’ study are consistent with Marshall’s conjecture that most Americans favor the death penalty and are ignorant of the facts concerning the death penalty. Almost 60% of respondents favored the death penalty. When the category undecided was eliminated, support for the death penalty increased to 64%. Ten factual questions were asked. Most respondents, abolitionists and retentionists alike showed a high level of ignorance of facts about the death penalty.
      Bohm et al (1991) tested the Marshall Hypothesis. Their study consisted of 272 undergraduate students from a mid-sized university in northern Alabama. They measured the opinions of respondents using a 7 point Likert scale ranging from “very strongly opposed” to “very strongly in favor” to measure the effects of knowledge through the use of the before and after experimental design with both experimental and control groups in a classroom setting. The respondents were not randomly selected. The experimental group was exposed to stimuli consisting of 40 hours of lecture material including lectures by instructors, presentations by guess speakers, films, and classroom discussion, while the control group was not. Hugo Bedau’s book The Death Penalty in America (1982) was the assigned text. In addition, issues that Justice Marshall argued would affect attitudes were discussed such as deterrence, incapacitation, public opinion, religious and retribution arguments, and information on the administration of the death penalty in reference to cost, arbitrariness and discrimination, and execution of innocent persons.
     The results of the study yielded interesting and expected outcomes in these two hypotheses:

  1. The public lacks knowledge about the death penalty and its effects
  2. An informed general public would oppose the death penalty.

     The results of the first hypothesis (the public lacks knowledge of the death penalty and its effects) suggested what the researchers expected to find that an uninformed public is not knowledgeable of the death penalty. In the pre-test both groups of respondents averaged 52% percent of the knowledge items correctly, which in Bohm’s estimation was considered “relatively uninformed.” The experimental group answered 54% of these items correctly, while the control group answered 42% of the knowledge items correctly. A “Z” test was used to determine if the difference in percentage of right and wrong answers of both groups was significant. It revealed that the difference in percentage of the two groups did not reach significance. When race and gender were controlled for, the results of the “Z” test still did not reach significance; race and gender had no effect on knowledge. Nevertheless, the results of the first hypothesis revealed that most respondents were not knowledgeable of the death penalty.
     The researchers also tested whether changes in opinion were related to the initial amount of knowledge the respondents possessed. The results indicated that the experimental group gained knowledge or became “more informed” when given the post-test. In the pretest, respondents in the experimental group correctly answered 54% of the questions; however, when given the post test, the respondents answered 79% of the items correctly. When each item was examined separately, eleven of the 14 items reached significance at 0.05 level. This was expected because of the experimental manipulation that followed the pre-test. The control group yielded similar results in both pre and post tests. Both tests indicated that the respondents were uninformed. The pre-test of the control group revealed that respondents correctly answered 46% of the questions while the post test revealed 48% correct answers.
     The second hypothesis: an informed public would generally oppose the death penalty yielded interesting but expected results also. It showed that an informed public would oppose the death penalty. When both groups were compared, the experimental group showed a change in attitude toward the death penalty after the stimuli were applied. A chi-square test revealed that the experimental stimuli did indeed produce a statistically significant increase in respondents who opposed the death penalty, though the increase did not reach 50%. However, the control group revealed no change. Race and gender were controlled for. Neither males nor females reached significance at the 0.05 level in either group. However, when Blacks and Whites were compared, significance level was reached. Black respondents in the experimental group reached significance at 0.05 in 75% of the death penalty conditions in the post test while white respondents in the experimental group did not reach significance at any level when the post-test was given. To add strength to the results of their hypothesis, Bohm et al (1991) used the Guttman scale to measure their respondents’ attitudes under certain conditions.
     The results of the conditions compiled by the Guttman scale and tested by the McNemar Chi-Square revealed that the percentage of experimental group subjects opposed to the death penalty in the experimental group (56.3%) was significantly greater than the percentage opposed prior to the experimental manipulation (34.7%). The control group showed no statistical significance changes in neither pre-test (30.5%) nor post-test (32.9). The researchers suggested that these results added credence to their assumption that the more citizens are informed about the death penalty, the less likely they will be to support it. The overall results indicated that even with an informed public, the Marshall hypothesis did not significantly change opinions in every death penalty condition. Yet, when race and gender were controlled, there was less support for the death penalty among Blacks and females; White males were the least likely to change their opinions.
     Bohm et al (1993) in a two year panel study (1988-1989) tested the Marshall Hypothesis also and addressed the stability of informed students and their opinions on the death penalty. The study consisted of 120 undergraduates from a medium-sized university in northeastern Alabama who had participated in a death penalty course that met for two hours a day, five days a week, for four weeks. The experimental manipulation consisted of videos, guest lecturers, lectures by the instructor, and an assigned text entitled “The Death Penalty in America (1982) by Hugo Bedau. The respondents were given a pre-test and post-test to measure their attitude toward the death penalty before and after the experimental manipulation. The researchers employed four different death penalty questions in this study. The first two questions (which are the ones of interest) were abstract and asked which of the following statements best describes your position toward the death penalty for some people convicted of first degree murder and which of the following statements best describe your position toward the death penalty for all cases? A 7-point Likert scale ranging from “very strongly” in favor to “very strongly” opposed was used to measure attitudes. Again, the results of the panel study yielded similar results as Bohm et al (1991). Of all the variables used, race was the only significant variable on both of the death penalty opinion questions. The researchers suggested that this result is not striking because race was always the distinguishing factor in death penalty opinions in America. Whites have always been more supportive of the death penalty than Blacks. This study showed that regardless of the death penalty opinion question asked, white support for the death penalty always exceeded Black support when given the pre-test, post-test, and follow-up. The results of this study indicated that students’ attitudes toward the death penalty were not changed totally by the assumptions of the Marshall Hypothesis. However, their study did show a little substantiation for the Marshall Hypothesis in that an informed public would show less support for the death penalty.
Methodological Concerns
     Past literature also suggested that the Marshall Hypothesis encountered methodological and operational problems. Bohm et al (1991) suggested that the terms “knowledgeable” and “informed” about the death penalty would be problematic to operationalize. As stated by Thurgood Marshall, “[F]or the average citizen, except those who base their opinion on retribution, knowledge of some information listed below is ‘critical’ to an informed opinion and would almost surely convince [the average citizen] that the death penalty was unwise” (Furman v Georgia, 1972). Bohm et al (1991) argued that this point is problematic because Marshall is assuming that an informed or knowledgeable citizen would know some or all of the following: the death penalty is more no more of an effective deterrent than life imprisonment; convicted murderers are rarely put to death, but are usually sentenced to a term in prison; convicted murderers usually are model prisoners and they almost always become law-abiding citizens after their release from prison; the cost of executing a prisoner exceeds the cost of imprisoning him for life; while in prison, a convict under the sentence of death performs none of the useful functions that life prisoners perform; no attempt is made to ferret out likely recidivists for execution; the death penalty may actually stimulate criminal activity; capital punishment is imposed discriminatorily against certain identifiable classes of people; there is evidence that innocent people have been executed before their innocence can be proven; and, the death penalty wreaks havoc with our criminal justice system.
     An interesting point made by Bohm et al (1991), however, considered how Marshall operationalized “informed.” What percentage of right answers did Marshall suggest constitute an “informed” citizen? According to Bohm (1991), Marshall did not set a standard on the amount of right answers needed to be informed or knowledgeable; therefore, this question will never be answered without skepticism. As a result, we can only assume what Marshall meant by “informed” or “knowledgeable.”
     There is a suggestion and a possible way to combat the operationalization problem encountered by Bohm et al (1991). Use the normal grading standard used in the public education institutions in the United States. This is not only an answer to the problem, but also a standard that can possibly be used whenever the Marshall Hypothesis is tested.
     I will operationalize “informed” or “knowledgeable” with the normal grading standard used in most public colleges in the United States which is the score of at least seventy percent for a minimum passing score. Therefore, the higher the percentage of right answers, the more knowledgeable or informed the citizen is. Since 70% is the standard passing grade at most United States public educational institutions, it is not unreasonable to use this standard with the Marshall Hypothesis.
     Past research also addressed the issue of question wording. Durham, Elrod, and Kincade (1994) made an interesting point about the question wording of capital punishment. They note that “when citizens are asked generic questions about capital punishment, the results seem to indicate strong support”(p. 707). But, according to the Gallop Polls in the years 1936, 1938, and 1986, different versions influenced the responses. For example, the first version question that had been asked about the death penalty was, “Do you believe in the death penalty for murder?” The second version asked, “Do you favor or oppose capital punishment for murder?” The third version asked, “Do you favor or oppose the death penalty for persons convicted of murder?” Durham, Elrod, and Kincade (1994) also pointed out that there was no category offered for “don’t know” or “no opinion.” Although they concluded that there is no clear evidence that the question wording dramatically affected the responses of citizens, and that these were minor problems, the survey was still changed in later polls with an additional category for “don’t know” and “no opinion” (p 707). But what is interesting is that when an alternative to the death penalty (such as no parole ever plus restitution) is given, then support for the death penalty dropped drastically (Allen et al., 2003, p 419).
     The validity of the Marshall Hypothesis was addressed by Bohm et al (1991) as well. Bohm’s main area of concern was with the experimental stimulus itself. Bohm suggested that there were experimental deficiencies in two previous studies by Sarat and Vidmar (1976) and Vidmar and Dittenhoffer (1981). Sarat and Vidmar’s stimuli, by their own admission, limited the potential for students to develop a true opinion about the death penalty because the students were limited in their time to read the two 1,500 word essays described as “scientific and other information.” As stated:

Without question our information manipulations had limited potential for developing truly informed opinion about the death penalty –the issues are intricate and complex while the essays are short and simple; furthermore, exposure to the information took place in a brief interview session without time and for reflection, discussion, or clarification. (p.183).

To avoid the validity problems experienced by Sarat and Vidmar (1976) and Vidmar and Dittenhoffer (1981), their procedure will not be used. This will, in fact, guarantee that the validity problem they encountered will not exist. This study will focus on the effect of knowledge on attitudes toward the death penalty.
Statement of the Hypothesis
     The Marshall Hypothesis suggests that knowledge and application of the death penalty will have an effect on citizens’ attitudes toward the death penalty. Therefore, I hypothesize the following:
H1: Students who are exposed to knowledge of the death penalty will show less support for the death penalty than those students who are not exposed to knowledge of the death penalty.
The null hypothesis is as stated:
H01: There will be no significance in students who are exposed to knowledge of the death penalty and those who are not exposed to knowledge of the death penalty.
Thus, the model is as follows:
Support for the death penalty = a (constant) + bx1 (knowledge) + bx2 (race) + bx3 (gender) + bx4 (education) + bx5 (income)
Methods
Subjects and Experimental Design
     A consent and confidentiality form was given to and signed by all subjects who participated in the study. Subjects of the study were 110 undergraduates of a medium-sized community college in Southern Louisiana: 81 subjects in an experimental group and 29 in the control group. The experimental group was comprised of students enrolled in a Corrections Process class which focused on the death penalty, and the control group consisted of students enrolled in American Government courses offered at about the same time as the Corrections Process course which was during the spring 2004 semester. None of the subjects were randomly selected. The Corrections Process class met three hours per day, once a week, for five weeks (a total of 15 hours) during the spring semester of 2004. The students were assigned several readings which included chapters from books and articles on the death penalty (see appendix for names of books and articles). In addition to the reading assignments, coursework included lectures by the instructor (6 hours), video tapes (6 hours), classroom discussion (3 hours), and a research paper examining issues concerning the death penalty. Lecture topics included the history of the death penalty in America; relevant Supreme Court cases, public opinion, deterrence, both specific and general, morality, cost, executing the innocent, and discrimination. This information was used as the stimulus for the experimental group. Subjects were asked to complete questionnaires at the beginning and end of the semester. Each questionnaire contained 81 items divided into 3 sections. The first section sought information about opinions regarding the death penalty. The second section tested knowledge of factual information about the death penalty. The third section sought information about the demographic characteristics of the subjects.
Design/Sampling Procedure
     This study used a quasi-experimental design similar to the before and after with a control group or classic experimental design. The major difference in these two designs is that this quasi-experimental design is not randomly selected. The students were not randomly selected because the setting was not applicable. Surveys compared the criminal justice students (experimental group) to the other students (control group). A non-probability sampling procedure known as “purposive” or “judgmental sampling” was used to conduct this study. Judgmental or purposive sampling allows the researcher the opportunity to become familiar with the subjects because of his/her knowledge of the population (Champion 2000). Champion also suggested that purposive sampling is used in social sciences when the community is small and sentimental issues such as capital punishment are discussed.
Measurement Procedures
     As mentioned prior, the subjects completed questionnaires that sought information about the following 3 variables in the study: opinions toward the death penalty, knowledge about the death penalty, and questions on the demographics of the respondents. Opinions toward the death penalty were measured by the 5-point Likert scale, with response categories ranging from strongly agree to strongly disagree. Likert scales are used in most social science research. The underlying rationale or purpose of a Likert scale is to group respondents in terms of their agreement or disagreement with a particular attitude or opinion (Taylor, 1994, p. 122). The Likert scale assumes that to measure an attitude or opinion several different items can be used. It also assumes that the items are almost interchangeable with each other, and that each person responsive to an item reflects the position of that respondent on that particular issue (Taylor, 1994). There were twenty-two knowledge questions that were measured by true or false answers. True and false questions were appropriate for this study to test whether respondents were knowledgeable of certain facts in reference to the death penalty. Either they knew the answers or they did not. The demographic questions were standard measures ranging from nominal to interval.
Assumptions
     Previous research indicates that most people are not cognizant about the death penalty; Blacks are less likely than Whites to support the death penalty; women are less likely than men to support the death penalty; the poor are less likely than the rich to support the death penalty; and the informed are less likely to support the death penalty than the uninformed (Bohm et al, 1993; Bohm et al, 1991; Bohm & Aveni; Ellsworth & Ross, 1983; Sarat &Vidmar, 1976). My research should indicate similar results. At the beginning of the semester, subjects were not cognizant of the death penalty. However, when the post-test was given at the end of the semester, the subjects were cognizant. This assumption was validated with the extent of knowledge among the subjects by asking them to answer 22 factual questions about the death penalty on both pre-tests and post-tests (see appendix). The results of this study, using the demographic variables discussed above, will be discussed later in this paper.
Measurement
Dependent Variables
     The dependent variable measured was attitude toward the death penalty. More specifically, the statement for support of the death penalty reads: I support the death penalty as it is presently applied in the United States. A five-point Likert scale ranging from “strongly disagree” to “strongly agree” will be used to measure these items.
     Support is measured by using a series of questions from a structured questionnaire (See Appendix). They are all attitudinal and therefore are used to tap into attitudes. These scales were developed by Robert Bohm, Susan Howell (University of New Orleans), and the researcher (myself) and will contain five demographic items including age, race, gender, religious preference, and political preference. The reliability of the additive scales will be evaluated by alpha or omega.
     A pilot test was conducted on students who did not participate in the experiment to insure that the subjects can and will respond according to instructions, to uncover and decide how to handle unanticipated problems and to use and check the accuracy of the instrument. In addition, the subjects were asked to give any feedback on questions they felt were difficult or unclear to answer. Pilot testing allows for improving the questionnaire before the actual study by correcting any unforeseen errors in questionnaire construction.
     The content validity of the instrument is evident. The instrument contains many items that cover a wide range of domain concept items which gives it face validity. The face validity of these items suggests reliability. Twenty-two items measure the construct knowledge; sixty-one items measure the construct attitude toward the death penalty. In addition, these are straight-forward measures of concrete phenomena. The measures of knowledge and attitude toward the death penalty should be related to each other; that is, a respondent who scores on a particular measure showed favorable attitudes toward the death penalty and would be expected to score low on a measure of knowledge. A respondent whose scores on a particular measure showed unfavorable attitudes toward the death penalty would be expected to score high on a measure of knowledge. If the measures of the constructs are not related, construct validity will be questioned.
Findings
Descriptive Statistics
     There were a total of 110 subjects within two groups. Eighty-one were in the experimental group and 29 in the control group. Age was broken down into four categories (16-25; 26-35; 36-45; over 45). Seventy-one subjects were in the 16-25 year-old category; 23 were in the 26-35 year old category; 11were in the 36-45 year old category; two were in the over 45 year old category; and two subjects did not complete the age category. As for gender, 59 (54.9%) subjects classified themselves as female and 50 (45.1%) classified themselves as male. As for race, 55 (50.5%) subjects classified themselves Black; 40 (36.7%) classified themselves White; 8 (7.3%) classified themselves Hispanic; and 6 (5.5%) classified themselves as other. As for education, 31 (28.4%) considered themselves freshmen; 44 (40.4%) considered themselves sophomores; 21 (19.3%) considered themselves juniors; 6 (5.5%) consider themselves seniors; and 7 (6.4%) consider themselves other. As for annual income, 38 (30.5%) had an annual income under $10,000; 33 (30.8%) had an annual income between $10,000 and $20,000; 24 (22.4%) had an annual income between $20,001 and $30,000; 7 (6.5%) had an annual income between $30,001 and $40,000; and 5 (4.7%) had an annual income over $40,000. As for religious denomination, 36 (33.6%) considered themselves Catholic; 31 ( 29%) considered themselves Baptist; 15 (14%) considered themselves Christian, 2 (1.9%) considered themselves Jewish; 2 (1.9%) considered themselves Methodist; 1 (0.9%) considered themselves Protestant; 2 (0.9%) considered themselves Muslims; and 19 (17.8%) considered themselves other. As for political affiliation, 42 (49.3%) considered themselves Democrat; 26 (24.3%) consider themselves Independent; 21 (19.6%) considered themselves Republican; and 18 (16.8) considered themselves other.
Inferential Statistics
     A battery of tests was conducted to measure disparity in the respondents’ attitude toward the death penalty based on race, gender, and political affiliation. Income and religion were not measured because of measurement errors that were detected after the experiment was done (see appendix). A comparison of means test was conducted in the pre-seminar and post-seminar period to examine the difference in the level of support for the death penalty between the experimental and the control group. Also conducted was a comparison of means test on the knowledge of death penalty between the experimental and control groups in the pre-seminar and post-seminar periods.
     First, an analysis of variance (ANOVA) test was conducted to examine if there were significant differences in the level of support for the death penalty based on the racial composition of the respondents. The results from this test revealed a significant difference between all racial groups in the pre-seminar and the post-seminar period. The mean opposition to the death penalty pre-seminar for Blacks was 3.2, Whites 1.9, Hispanics 2.4, and others 2.5. Also, the mean opposition to the death penalty post seminar for Blacks was 4.0, Whites 2.3, Hispanics 3.2, and other 2.6.
     To further explore these differences, a comparison of means test was conducted, selecting only two racial groups for each test. In the first test, the mean level of support for the death penalty between black and white respondents was conducted. Results from this test showed significant differences in the average level of support for the death penalty between black and white respondents. On the average, black respondents were more likely to be less supportive of the death penalty than white respondents (Table 2b). This result was not surprising because the previous literature supports this finding. When Black and Hispanic respondents were compared, the difference was statistically insignificant. In other words, Black respondents did not show a higher or lower level of support for the death penalty compared to Hispanics. It is important to mention here that this insignificance may be due to sampling errors. There are only 8 Hispanic respondents in this survey, compared to 55 Black respondents. One issue of concern in this case will be the problem of the homogeneity of variance of responses.
     To examine the difference in the level of support for the death penalty based on political affiliation, an ANOVA test was conducted on the respondents’ level of support using their political affiliation as the independent variable. The results were quite similar to those obtained from the ANOVA test analyzing races. The results showed a statistically significant difference in the average level of support for the death penalty, when respondents are classified by political affiliations. A more in-depth exploration of this difference using a comparison of means test revealed that the average level of support for the death penalty was not the same for democrats and republicans. Democrats are less supportive of the death penalty than republicans. This, too, was expected because past literature reveals this finding.
     Pre-seminar republicans scored the mean of 1.9 while democrats scored a mean of 3.1. This difference is statistically significant. In the pre-seminar survey Republicans appeared to be more supportive of the death penalty, while on the other hand, Democrats appeared to be less supportive of the death penalty. Examining the results from the post-seminar test, republicans scored a mean of 2.5 while democrats scored a mean of 3.75. This difference is also significant. While the level of support for death penalty reduced for both republicans and democrats, the opposition to the death penalty appears to increase more significantly amongst democrats than amongst republicans. The support for the death penalty among republican respondents is expected because the death penalty is one of the political issues of the Republican platform. Republicans are widely regarded for being more supportive of the death penalty than are democrats.
      A comparison of means test was also administered to see if the average level of support for the death penalty was the same for male and female respondents. Pre-seminar males scored a mean of 2.2, while females scored a mean of 3.0. Post-seminar males scored a mean of 3.2 and females scored a mean of 3.6. The p-value revealed that there is a difference in the level of support for the death penalty in the pre-seminar response between male and female respondents. This was expected as well because women generally are less supportive of harsh punishments.
     Another interesting trend examined was to see which race (if any) was most influenced by the seminar. The mean opposition to the death penalty pre-seminar for Blacks was 3.2, Whites 1.9, Hispanics 2.4, and others 2.5. While the mean opposition to the death penalty post seminar for Blacks was 4.0, Whites 2.3, Hispanics 3.2 and other 2.6, only Blacks had a significant level of difference between the pre-seminar and post-seminar period. The result is consistent with the fact that blacks were most influenced by the seminar, while whites and Hispanics were not influenced. Only black respondents showed a significant change in the level of support for the death penalty. After undergoing the seminar, blacks increased their opposition toward the death penalty. This was expected because past research indicates that blacks have been victims of the application of the death penalty disproportionately when compared to other racial groups.
     Finally, I conducted a comparison of means test to examine if there are significant differences in the knowledge of death penalty between the experimental group and the control group in the pre-seminar and the post-seminar period. The knowledge score of the experimental group (criminal justice class) was a mean of 61.6% pre-seminar, while the knowledge score of the control group (political science class) was a mean of 63.3% pre-seminar. There is no significant difference in the pre-seminar scores, between the experimental and the control group. This result indicates that prior to undergoing the seminar, it appeared that respondents in both the experimental group and the control group were not knowledgeable about the death penalty. However, the mean of the post-seminar knowledge score of the experimental group was 81.7% while the mean of the post seminar knowledge score of the control group was 56.0%. This difference is statistically significance and appears to be consistent with the fact that the seminar did improve the respondent’s knowledge of the death penalty.
Correlations
     Correlations were also examined to see if there was any correlation between the level of support for the death penalty and the following explanatory variables (age, education, and annual income). The results of the correlations indicated that in the pre-seminar test there was no correlation among those variables. However, post-seminar correlation existed between education and support for the death penalty: the more educated a respondent is the higher, his/her level of support for the death penalty.
Discussion & Conclusion
Summary
     This study was designed to test Supreme Court Justice Thurgood Marshall’s Hypothesis that an informed public would show less support for the death penalty. His hypothesis was based on variables that he assumed would change a person’s mind after that person was exposed to the facts concerning those variables in relation to the death penalty. Those variables included morality, deterrence effect, cost to execute, type of lawyer, error, and discrimination. Marshall assumed that if a person was exposed to enough information about those variables and their relationship to the death penalty, that person would not support the death penalty. Furthermore, he assumed that if people knew how those variables were infested with flaws, a person would have second thoughts in supporting the death penalty. Those issues were discussed thoroughly enough for the students to make an objective and sensible decision on whether they would affect their attitude toward the death penalty; they were discussed thoroughly enough for students to become knowledgeable and aware of how the death penalty really works in America; they showed flaws that existed in the application of the death penalty. Students were exposed to literature, video tapes, and lecture material that discussed all of those variables in detail. In addition, students were assigned to write a research paper discussing those issues. The literature, lecture material, video tapes, and research paper were the tools used as the stimuli to increase the students’ knowledge and make them better informed about the death penalty. Marshall felt that this type of exposure would decrease a person’s willingness to support the death penalty.
     When testing the Marshall hypothesis, two groups were used, the control group and the experimental group. Both groups were given a pre-test in the form of a questionnaire that measured their attitude and their knowledge toward the death penalty. The stimuli included facts concerning issues on morality, deterrence, cost, type of lawyer, errors, and discrimination; video tapes casting doubt on conviction; lecture material; and an assigned term paper all given to the experimental group. Then both groups were given the post-test in the form of a questionnaire to see if their knowledge about and attitudes toward the death penalty had changed. Past literature supported the Marshall Hypothesis at least partially because after the stimulus was applied respondents became more aware of the death penalty and their attitudes did change, sometimes significantly. This study indicated similar results to that of past literature.
     The literature review of past studies that tested the Marshall Hypothesis indicated that most students were going to support the death penalty and most students were not going to know much about important issues and facts on issues concerning the death penalty. The literature review also indicated that two trends emerged in support of the death penalty. From 1936-1986, Whites, males, republicans, Protestants, wealthier people, and Westerners supported the death penalty, while Blacks, democrats, females, and Southerners tended not to support the death penalty. As of 1991, Whites, males, and republicans still supported the death penalty, while Blacks, democrats, and women tended not to support the death penalty. Income, religion, age, education level, and region were not factors. This study mirrors the results of past literature.
     A battery of tests was run to see if any patterns emerged between the experimental and control groups. Some were descriptive and others were inferential. They were conducted to describe the data and measure disparity in the respondents’ attitude towards the death penalty based on race, gender, and political affiliation. These tests included correlations, several comparison of means tests, and analysis of variance (ANOVA). The results of the tests partially supported Marshall’s hypothesis in the following situations. Knowledge decreased support for the death penalty in the experimental group while it had no effect in the control group. Both comparison of means test and correlation tests supported the findings in the regression models. They showed a significant variation in the level of support for the death penalty between races, political affiliation, and gender. As mentioned prior, blacks, democrats, and women are less supportive of the death penalty, while whites, republicans, and men are more supportive. Interestingly, the models indicated several other findings. Older respondents and the more educated respondents tended to support the death penalty while younger and less educated respondents tended not to support the death penalty. These findings were not revealed in past literature on the Marshall hypothesis. However, an explanation could be that older people tend to become more conservative. They may have developed a “get tough on crime” attitude that is part of the conservative political platform. As far as education is concerned, one explanation could be that more uneducated people are being put to death, and that the more educated do not feel like it would happen to them. Clearly, much more research should be done in this area in the future. The following issues should be addressed before any generalizations could be made. First, students’ participation needs to be addressed. Some students were asleep; some were absent on certain days; and some were inattentive. Surely this could have affected the results of the surveys. Second, a random sample should be given, although this is impossible when dealing with college students in a particular program. Then, a more concrete generalization can be made. And, thirdly, more up to date and exhaustive information about facts concerning the death penalty, and other stimuli to affect death penalty opinions should be used.

Table 1: Descriptive Statistics of Sample


A: Groups

 

 

 

 

Frequency

Valid Percent

Cumulative Percent

Experimental

81

73.6

73.6

Control

29

26.4

100

Total

110

100

 

 

 

 

 

B: Age

 

 

 

 

Frequency

Valid Percent

Cumulative Percent

16 to 25 years

71

66.4

66.4

26 to 35 years

23

21.5

87.9

36 to 45 years

11

10.3

98.1

Over 45 years

2

1.9

100

Total

107

100

 

 

C: Sex

 

 

 

 

Frequency

Valid Percent

Cumulative Percent

Female

59

54.1

54.1

Male

50

45.9

100

Total

109

100

 

 

 

 

 

D: Race

 

 

 

 

Frequency

Valid Percent

Cumulative Percent

Black

55

50.5

50.5

White

40

36.7

87.2

Hispanic

8

7.3

94.5

Others

6

5.5

100

Total

109

100

 

E: Education

 

 

 

 

Frequency

Valid Percent

Cumulative Percent

Freshmen

31

28.4

28.4

Sophomore

44

40.4

68.8

Junior

21

19.3

88.1

Senior

6

5.5

93.6

Other

7

6.4

100

Total

109

100

 

 

 

 

 

F: Annual Income

 

 

 

 

Frequency

Valid Percent

Cumulative Percent

Under $10,000

38

35.5

35.5

$10,000 to $20,000

33

30.8

66.4

$20,001 to $30,000

24

22.4

88.8

$30,001 to $40,000

7

6.5

95.3

Above $40,000

5

4.7

100

Total

107

100

 

 

 

 

 

G: Religious Denomination

 

 

 

Frequency

Valid Percent

Cumulative Percent

Protestant

1

0.9

0.9

Catholic

36

33.6

34.6

Baptist

31

29

63.6

Muslim

1

0.9

64.5

Jewish

2

1.9

66.4

Methodist

2

1.9

68.2

Christian

15

14

82.2

Other

19

17.8

100

Total

107

100

 

 

 

 

 

H: Political Affiliation

 

 

 

 

Frequency

Valid Percent

Cumulative Percent

Republican

21

19.6

19.6

Democrat

42

39.3

58.9

Independent

26

24.3

83.2

Other

18

16.8

100

Total

107

100

 

Table 2: Analysis of Variance and Comparison of Means Tables.

A: ANOVA

 

 

 

 

 

 

Dependent Variable: Mean Opposition to the death penalty

 

 

 

 

TEST PERIOD

 

Sum of Squares

Df

Mean Square

F

Sig.

Pre seminar

Between Groups

22.679

3

7.56

4.186

0.008

 

Within Groups

189.633

105

1.806

 

 

 

Total

212.312

108

 

 

 

Post seminar

Between Groups

31.75

3

10.583

6.616

0.000

 

Within Groups

164.755

103

1.6

 

 

 

Total

196.505

106

 

 

 

Comparison of Means Tables


B: Comparison of Means test on Opposition to the death penalty between Blacks and White

 

 

 

 

TEST PERIOD

Race

N

Mean

Std. Deviation

Std. Error Mean

 

Pre seminar

Black

55

3.2

1.238

0.167

 

 

White

40

2.25

1.565

0.247

 

Post seminar

Black

60

3.82

1.255

0.162

 

 

White

35

2.71

1.274

0.215

 

 

 

 

 

 

 

 

 

 

 

 

 

T-test for Equality of Means

 

TEST PERIOD

 

 

T

Sig. (2-tailed)

Mean Difference

Std. Error Difference

 

 

 

 

 

 

 

Pre Seminar

Mean Support for the death penalty

3.302

0.001

0.95

0.288

Post Seminar

Mean Support for the death penalty

4.107

0

1.1

0.268

 

 

 

 

 

 

 

C: Comparison of Means test on Opposition to the death penalty between Blacks and Hispanics

 

 

 

 

TEST PERIOD

Race

N

Mean

Std. Deviation

Std. Error Mean

 

Pre seminar

Black

55

3.2

1.238

0.167

 

 

Hispanic

8

2.5

1.069

0.378

 

Post seminar

Black

60

3.817

1.255

0.162

 

 

Hispanic

7

3.286

1.113

0.421

 

 

 

 

 

 

 

 

 

 

 

 

 

T-test for Equality of Means

 

TEST PERIOD

 

 

T

Sig. (2-tailed)

Mean Difference

Std. Error Difference

 

 

 

 

 

 

 

Pre Seminar

Mean Support for the death penalty

1.516

0.135

0.700

0.462

Post Seminar

Mean Support for the death penalty

1.070

0.289

0.531

0.496

 

 

 

 

 

 

 

D: Comparison of Means test on Opposition to the death penalty between the experimental group and the control group

 

 

 

 

 

TEST PERIOD

Groups

N

Mean

Std. Deviation

Std. Error Mean

 

 

Pre seminar

Experimental

81

2.716

1.485

0.165

 

 

 

Control

29

2.828

1.136

0.211

 

 

Post seminar

Experimental

81

3.481

1.388

0.154

 

 

 

Control

30

2.933

1.230

0.225

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

T-test for Equality of Means

 

 

TEST PERIOD

 

 

T

Sig. (2-tailed)

Mean Difference

Std. Error Difference

 

 

 

 

 

 

 

 

 

Pre Seminar

Mean Support for the death penalty

-0.367

0.714

-0.112

0.304

 

Post Seminar

Mean Support for the death penalty

1.902

0.060

0.548

0.288

 

 

 

 

 

 

 

 

 

E: Comparison of Means test on Opposition to the death penalty between Males and Females

 

 

 

 

TEST PERIOD

Sex

N

Mean

Std. Deviation

Std. Error Mean

 

Pre seminar

Male

35

2.286

1.178

0.199

 

 

Female

45

3.067

1.629

0.243

 

Post seminar

Male

31

3.226

1.407

0.253

 

 

Female

50

3.640

1.367

0.193

 

 

 

 

 

 

 

 

 

 

 

 

 

T-test for Equality of Means

 

TEST PERIOD

 

 

T

Sig. (2-tailed)

Mean Difference

Std. Error Difference

 

 

 

 

 

 

 

Pre Seminar

Mean Support for the death penalty

-2.487

0.015

-0.781

0.314

Post Seminar

Mean Support for the death penalty

-1.311

0.194

-0.414

0.316

 

 

 

 

 

 

 

 

 

 

 

 

 

 

F: ANOVA Test on the mean opposition to the death penalty across political affiliations

 

 

 

 

 

 

 

 

 

TEST PERIOD

 

Sum of Squares

Df

Mean Square

F

Sig.

Pre seminar

Between Groups

20.69940548

3

6.900

3.812

0.012

 

Within Groups

186.4407814

103

1.810

 

 

 

Total

207.1401869

106

 

 

 

Post seminar

Between Groups

22.44212636

3

7.481

4.352

0.006

 

Within Groups

166.7261905

97

1.719

 

 

 

Total

189.1683168

100

 

 

 

 

 

 

 

 

 

 

G: Comparison of Means test on Opposition to the death penalty between Republicans and Democrat

 

 

 

 

 

TEST PERIOD

Political Affiliations

N

Mean

Std. Deviation

Std. Error Mean

 

 

Pre seminar

Republican

35

1.9

0.831

0.181

 

 

 

Democrat

45

3.12

1.533

0.237

 

 

Post seminar

Republican

31

2.5

1.318

0.295

 

 

 

Democrat

50

3.75

1.246

0.180

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

T-test for Equality of Means

 

 

TEST PERIOD

 

 

T

Sig. (2-tailed)

Mean Difference

Std. Error Difference

 

 

 

 

 

 

 

 

 

Pre Seminar

Mean Support for the death penalty

-4.073

0.000

-1.214

0.298

 

Post Seminar

Mean Support for the death penalty

-3.621

0.001

-1.250

0.345

 

 

 

 

 

 

 

 

 

H: Comparison of Means test on Knowledge of the death penalty between the experimental group and the control group

 

 

 

 

TEST PERIOD

GROUPS

N

Mean

Std. Deviation

Std. Error Mean

 

Pre seminar

Experimental

81

61.670

15.318

1.702

 

 

Control

29

63.383

34.128

6.337

 

Post seminar

Experimental

81

81.717

16.348

1.816

 

 

Control

30

56.032

17.086

3.120

 

 

 

 

 

 

t-test for Equality of Means

 

TEST PERIOD

 

 

T

Sig. (2-tailed)

Mean Difference

Std. Error Difference

 

 

 

 

 

 

 

Pre Seminar

Knowledge Score

 

-0.261

0.796

-1.713

6.562

Post Seminar

Knowledge Score

 

7.262

0.000

25.685

3.537

 

 

 

 

 

 

 

I: Comparison of Means test on Opposition to the death penalty amongst all races between the pre- and post seminar response.

 

 

 

 

Race

TEST PERIOD

N

Mean

Std. Deviation

Std. Error Mean

 

Black

Pre Seminar

44

3.273

1.301

0.196

 

 

Post Seminar

49

4.041

1.154

0.165

 

White

Pre Seminar

27

1.926

1.567

0.302

 

 

Post Seminar

23

2.391

1.118

0.233

 

Hispanic

Pre Seminar

5

2.400

1.140

0.510

 

 

Post Seminar

5

3.200

1.304

0.583

 

Others

Pre Seminar

4

2.500

1.000

0.500

 

 

Post Seminar

3

2.667

2.082

1.202

 

 

 

 

 

 

 

 

Independent Samples Test

 

 

 

 

 

 

 

 

 

 

T-test for Equality of Means

 

Race

 

 

T

Sig. (2-tailed)

Mean Difference

Std. Error Difference

Black

Mean Support for the death penalty

-3.018

0.003

-0.768

0.255

White

Mean Support for the death penalty

-1.189

0.240

-0.465

0.391

Hispanic

Mean Support for the death penalty

-1.033

0.332

-0.800

0.775

Others

Mean Support for the death penalty

-0.143

0.892

-0.167

1.167

 

Table 3: Pearson Correlations

Test Period

 

 

Support for the death penalty

Age

Education

 

Pre Seminar

Support for the death penalty

Pearson Correlation

 

 

 

 

Sig. (2-tailed)

 

 

 

 

 

 

 

 

 

Age

Pearson Correlation

0.087

 

 

 

Sig. (2-tailed)

0.374

 

 

 

 

 

 

 

 

Education

Pearson Correlation

0.004

0.126

 

 

Sig. (2-tailed)

0.966

0.198

 

 

 

 

 

 

 

Annual Income

Pearson Correlation

-0.008

0.566

0.240

 

Sig. (2-tailed)

0.934

0.000

0.013

 

 

 

 

 

 

Post Seminar

Support for the death penalty

Pearson Correlation

 

 

 

 

Sig. (2-tailed)

 

 

 

 

 

 

 

 

 

Age

Pearson Correlation

0.175

 

 

 

Sig. (2-tailed)

0.066

 

 

 

 

 

 

 

 

Education

Pearson Correlation

-0.190*

0.178

 

 

Sig. (2-tailed)

0.048

0.066

 

 

 

 

 

 

 

Annual Income

Pearson Correlation

-0.144

0.558

0.245

 

Sig. (2-tailed)

0.140

0.000

0.011

 

 

 

 

 

 

***. Correlation is significant at the 0.01 level (2-tailed).

**. Correlation is significant at the 0.05 level (2-tailed).

Bibliography

Acker, J., Bohm, R. M. & Lanier, C (1998). America’s experiment with capital punishment: Reflections on the past, present,
       and future of the ultimate sanction
. Carolina Academic Press: North Carolina.
Allen, H E. & Simonsen, C (2001). Corrections in America: An introduction. 9th edition. Prentice Hall: America.
Allen H.E., Simonsen, C.E., & Latessa, E. (2004). Corrections in America: An introduction. 10th Edition. USA: Prentice Hall.
Bedau, Hugo A. (1982). The death penalty in America. 3rd edition. Oxford University Press: New York.
Bedau, H.A. (1982). The death penalty in America. Oxford University Press: New York
Bohm, R.M. (1989a). The affects of classroom instructions and discussions on death penalty opinions: a teaching note.
       Journal of criminal justice
. Vol. 17: 123-131.
Bohm, R.M. (1989b). Humanism and the death penalty, with special emphasis on the post-Furman experience.
       Justice Quarterly 6:173-95.
Bohm, R.M. (1990). Death penalty opinions: Effects of a classroom experience and public commitment. Social Inq 60:285-97.
Bohm, R.M., Clark, L.J. & Aveni, A.F. (1990). The influence of knowledge on reasons for death penalty opinions:
       An experimental test. Justice Quarterly 7:175-88.
Bohm, R.M. (1991). Knowledge and death penalty opinion: A test of the Marshall hypotheses. J. Res Crime 28:360-97.
Bohm, R.M., & Vogel, R.E., (1991). Educational experiences and death penalty opinions: Stimuli that produce changes.
       Journal of Criminal Justice Education 2:69-80.
Bohm, R .M., Vogel, R.E. & Maisto, A.A. (1993). Knowledge and death penalty opinion: A panel study.
       J Criminal Justice
21:29-45.
Champion D (2000). Research methods for criminal justice and criminology. 2nd edition. Prentice Hall: New Jersey
Death Penalty Information Center (Internet: www.deathpenaltyinfo.org)
Durham, A.M., Elrod, H.P.& Kinkade, P.T. (1994). Images of crime and justice: murder and the “true crime” genre.
       Unpublished Manuscript. Tampa, Fla: University of Tampa: 9.
Durham, A, Elrod, P. H., & Kinkade, P. T. (December 1996). Public support for the death penalty: beyond gallop.
       Justice Quarterly, Volume 13.No.4.
Durham, M.A. (1996). Public support for the death penalty: Beyond Gallup. Justice Quarterly, Vol. 13 No. 4,
       December 1996

Ellsworth, P (1978). “Attitudes Toward Capital Punishment: From Application to Theory”. Paper presented at
       the Society for Experimental Social Psychology.
Ellsworth, P & Ross, L (1975) “Empirical Data and Judical Decision Making”. Paper presented at the Annual
       Convention of the American Psychological Association.
Ellsworth, P.C., & Ross, L. (1983). Public opinion and capital punishment: a close of examination of the views
       of abolitionist and retentionist. Crime and Delinquency. Vol. 29: 116-169.
Ellsworth, P.C. & Gross, S.L. “Hardening of the attitudes: Americans’ views on the death penalty.”
       Journal of Social Issues. L, pp. 19-52.
Furman V. Georgia (1976). 408 U.s. 238, 92 S.Ct. 2726, 33 L.Ed. 2d 346.
Lord, C. G. Ross, L., & Lepper, M. R. (1979). Biased assimilation and attitude polarization: the effects of
       prior theories on subsequently considered evidence. J. Perspectives Social Psychology. Volume 37:2098-2109.
Sarat, A. & Vidmar, N. (1976). Public opinion, the death penalty, and the 8th amendment: testing the Marshall hypothesis.
       Wisconsin law review. Vol. 17: 171-206.
Senna, J. & Siegel, L. Introduction to Criminal Justice.2002.
Taylor, R.B., (1994). Research Methods in Criminal Justice. McGraw Hill: America.
Sellen Thorsten (1967). Capital punishment
Vidmar, N. & Dittenhoffer, T. (1981). Informed public opinion and death penalty attitudes. Canadian
       Journal of Criminology. Vol. 23: 43-56
.
Vidmar, N., & Ellsworth, P. (1974). Public opinion and the death penalty. Stanford L Rev 26:1245-70.

Home | About NSSA | Membership Form | Conferences & Seminars | Publications | Officers & Board | Newsletter | Announcements | Contact Us
Site Map | Terms and Conditions | Privacy Policy
Designed by Dreamwirkz Web Designs 2007 All Rights Reserved