Bryce York​
​Hypothesis and motivation
We believe that there is some variability in sentencing and believe in a bias possessed by judges, lawyers, and other actors in the court room. In this process we are looking specifically at the role race plays when a defendant is sentenced. As most Americans may be led to believe that crimes committed by anyone of any social status or race are punished equally, we are here to find disparity in that claim. In our case a null hypothesis would prove a lack of statistical significance in variation when comparing the length of a murder sentence and race is different cases. Studying how our legal system operates and being informed of the processes helps citizens understand the legal proceedings that they are subject to under United States federal law. Finding disparity and problems with fair sentencing in US Courts is the first step at correcting and repairing this system. 
 
Description of Concepts, Data, and Measurement
The independent variable for this study is race which by the US Census Bureau includes: White American, Black or African American, Native American and Alaska Native, Asian American, Native Hawaiian and Pacific Islander, and Other. The dependent variable for this study is the length to which the defendants are sentenced for murder. Prior charges also play a role in sentencing and often result in longer sentences for people with multiple sentences. To accurately portray sentencing in the US for murder charges the cases studied should consist of only first-time offenders with no significant prior criminal history. 
 
 Presentation of Analysis and Results 
After analysis the correlation between race and sentencing shows to be statistically significant while gender and sentencing as seen in Morgan’s analysis shows a much higher correlation. This means that we would have to accept our null hypothesis that race and sentencing for federal crimes has little correlation to the length of an individual’s criminal sentence. 
 
ANOVAa                                                                      
Model             Sum of Squares           df         Mean Square  F          Sig.
1          Regression      24282.744       1          24282.744       2.539   .111b
            Residual          320135731.793           33468  9565.428                     
            Total    320160014.537           33469                          
2          Regression      1549612.774   2          774806.387     81.386 .000c
            Residual          318610401.763           33467  9520.136                     
            Total    320160014.537           33469                          
a Dependent Variable: SentenceInYears                                                                    
b Predictors: (Constant), Race1                                                                      
c Predictors: (Constant), Race1, Murder1                                                                  
 
 
Coefficientsa                                                               
Model             Unstandardized Coefficients              Standardized Coefficients       t           Sig.
                        B          Std. Error         Beta                 
1          (Constant)       15.835 .671                 23.608 .000
            Race1  1.770   1.111   .009     1.593   .111
2          (Constant)       14.564 .677                 21.524 .000
            Race1  1.820   1.108   .009     1.642   .101
            Murder1          37.635 2.973   .069     12.658 .000
a Dependent Variable: SentenceInYears                                                                    
 
 
Works cited 
Carlin, Jerome E., and Howard, Jan. “Legal Representation and Class Justice” UCLA Law Review, vol. 12, no. 2, January 1965, p.381-437. HeinOnline https://heinonline.org/HOL/Page?handle=hein.journals/uclalr12&div=34&g_sent=1&casa_token=&collection=journals1  
Kennedy, Edward, M. “Toward a System of Criminal Sentencing: Law with Order.” The American Criminal Law Review, vol. 6, no. 4, 1979, p. 353-382. https://heinonline.org/HOL/Page?handle=hein.journals/amcrimlr16&id=367&collection=journals&index=
US Census Bureau. “About Race.” About, United States Census Bureau, 23 Jan. 2018 
https://www.census.gov/topics/population/race/about.html.

​Hypotheses and motivation
We are studying the criminal justice system’s sentencing because we want to find out if the race of the criminal has an effect on the sentence length involving murder cases within the federal jurisdiction of the United States, in order to help the readers understand if fair treatment is being given under the Sixth Amendment. Does prior history and race effect the severity of a criminals federal murder sentence in the United States of America?  The Null Hypothesis (Ho) is prior history does not play a role in sentencing when involving murder cases within the federal jurisdiction of the United States. While the Alternative Hypothesis (H1) is prior history does not play a role in sentencing when involving murder cases within the federal jurisdiction of the United States. Abramm, Bertan and Mullainathan in their study of “Do Judges Vary in Their Treatment of Race” found that race played a role in conviction rate but were sentenced similarly to the white offenders.  Everate and Wojtkiewics found that race does play a role in sentencing and that minorities are sentenced at a higher rate than the other white offenders in their study “Difference, Disparity, and Race/Ethnic Bias in Federal Sentencing”.
Description of concepts, data, and measurement (25 points)
Our independent variables from Harvard’s DataVerse included, race (White vs People of Color as defined by the US Census Bureau, murder 1 classification, and the prior history of each defendent. Our dependent variable was the sentencing length in years.  Using the initial data that Lorrie presented, I added the variable of prior history and it showed that the prior history did increase the sentence by an average of 5.796 years but is not significantly significant. With the concepts of race with the description of all 5 categories, the federal murder definition, and the severity of punishment being defined in Lorries post below.  I added the prior history to see if the more crimes an individual committed prior to a Murder 1 charge added more time. The prior history is broken up into the number of charges, ranging from 0-10 prior charges. What the prior charges consisted of were not included in the data set which could also play a role in the addition of years to a sentence or play no role at all as well as the margin of error increased the larger the number of prior charges.
Presentation of analysis and results (25 points):
As Lorrie presented below, race (white individuals with a value of 1, and all other races with a 2 value), murder 1 (murder run as a value of 1, and other crimes with a value of 0) , and the sentencing in years had significant significance of 0.111 and 0.101.  Prior history has a significance of 0.269 when run with race, murder 1, and sentencing in years meaning that it plays a role but that is not significantly significant. With this data we would have to accept the null hypothesis.

Works Cited:
David S. Abrams, Marianne Bertrand, & Sendhil Mullainathan. (2012). Do Judges Vary in Their Treatment of Race? The Journal of Legal Studies, 41(2), 347. https://doi.org/10.1086/666006

​Ronald S. Everett, & Roger A. Wojtkiewicz. (2002). Difference, Disparity, and Race/Ethnic Bias in Federal Sentencing. Journal of Quantitative Criminology, 18(2), 189. Retrieved from https://search.ebscohost.com/login.aspx?direct=true&db=edsjsr&AN=edsjsr.23366801&site=eds-live&scope=site

​United States Sentencing Commission. “PDF.” 2010.
https://www.ussc.gov/sites/default/files/pdf/research-and-publications/federal-sentencing-statistics/quarterly-sentencing-updates/USSC_2010_Quarter_Report_Final.pdf
 
US Census Bureau. “About Race.” About, United States Census Bureau, 23 Jan. 2018, 
https://www.census.gov/topics/population/race/about.html.
 
“18 U.S. Code § 1111 - Murder.” Legal Information Institute, Cornell Law School, 
https://www.law.cornell.edu/uscode/text/18/1111.

- Morgan Rutkowski
Hypotheses and Motivation
As mentioned in Lorrie’s post, we see the original study we would be conducting. However, within every study, there is always the possibility of a confounding variable. Therefore, I added sex as another independent layer that could impact sentencing duration for criminals. For the null hypothesis, we thought that there would be no racial or sexual bias in judicial decisions involving murder cases within the federal jurisdiction of the United States. Moreover, for our alternative hypothesis, we would say that there is some sort of racial and/or sexual bias in judicial decisions involving murder cases within the federal jurisdiction of the United States.
Two studies to examine that are similar include Travis Franklin’s study on if there is racial disparity in criminal sentencing and if race influences the duration of the sentence (Franklin, 2018) and Kathleen Daly’s study on if their gender (sex) influences the sentencing duration of the offender, whether they are female or a male. Ultimately, although both have an impact, gender has a stronger statistical significance on sentencing duration.
Overall, we are studying the criminal justice’s sentencing system because we want to find out if the race, from the original study, and now the sex of the criminal have an effect on the severity of the punishment given by the judicial decisions involving murder cases within the federal jurisdiction of the United States, in order to help the readers understand if fair treatment is being given under the Sixth Amendment by judicial actors for biological factors they cannot control. For this project, the following variables are listed below:
 
Description of Concepts, Data, and Measurement
In our study, our operationalized independent variables from Harvard’s DataVerse include race (White vs People of Color) (US Census Bureau), murder type (Murder vs Not Murder) (“18 U.S. Code § 1111 - Murder.”), and Sex (either Male or Female) (Holmes, 2007). For our operationalized dependent variable, we looked at severity of punishment (in months) (United States Sentencing Commission). Our independent variable of race and murder, with the new addition of sex, causes our dependent variable of duration of the punishment to increase, especially with a stronger significance to if the criminal is a male possibility due to the fact that women could be seen as docile and “submissive” in comparison to a male who can seem more aggressive and “dangerous”.
Several concepts have been listed in Lorries post, including Murder (at the federal level), Race (and a description of all 5 categories), and Severity of Punishment. However, unique to my version, would be describing what “Sex” is vs what “Gender” is. Although there has been some recent debate on there being a difference between “Gender” and “Sex” of an individual, for the purpose of this study, “Sex” will be the biological difference and not the socially produced differences between being “masculine” and “feminine” (Holmes, 2007).
​
Presentation of Analysis and Results
Figure 1:




​









​






​
​​​Figure 2:

From the hypothesis test run in the figures 1-2 above, we see that while race alone, race with murder, and race with both murder and gender were all shown, we see the strongest statistically significant difference come from gender impacting the duration in months of the imprisonment of the criminal. Figure 3 represents this difference between the two sexes.
For this study, I ran a linear regression model examining Race (White =1 and People of Color=2), Murder (Murder =1 and Not Murder=2), and Sex (Male =1 and Female =2) in conversation with sentencing in months (DV). After running this test, I found that race only averaged a difference in a couple of months, about 1.785 months, when comparing a white person vs a person of color, resulting in a significance level of 0.117, which is not significant enough to express any real meaning to follow. However, when you add murder to race, the number of months dramatically increases, about 37.8 months, which makes sense because murder should result in a bigger difference in months when comparing it to someone who did not commit murder, resulting in a 0.000 significance. Finally, when adding gender/sex to the test, we saw that females were less likely to have a longer sentence than if the offender was a male, with about -4.470 months, favoring women, with a significance level of 0.015.
With all of this in mind, we see that race did not have a large enough significant difference on the outcome of severity on punishment. Instead, I found that the crime type (murder or not murder) and if the offender was female vs male did show a statistically significant difference. Ultimately, committing murder and being a male offender would increase the number of months in prison than if they did not commit murder and/or they were female.

Work Cited
Daly, Kathleen. Gender, Crime, and Punishment. Yale University Press, 1994. EBSCOhost,
search.ebscohost.com/login.aspx?direct=true&db=nlebk&AN=52930&site=eds-live&scope=site.

​Franklin, Travis W. “The State of Race and Punishment in America: Is Justice Really Blind?”
Journal of Criminal Justice, vol. 59, Nov. 2018, pp. 18–28. EBSCOhost, doi:10.1016/j.jcrimjus.2017.05.011.

Holmes, Mary. “What Is Gender?” Google Books, Google, 2007,
https://books.google.com/books?id=agLzMDaeYB4C&lpg=PP2&ots=5W9kaz6R9e&dq=what is gender&lr&pg=PA2#v=onepage&q&f=false.
 
United States Sentencing Commission. “PDF.” 2010.
https://www.ussc.gov/sites/default/files/pdf/research-and-publications/federal-sentencing-statistics/quarterly-sentencing-updates/USSC_2010_Quarter_Report_Final.pdf
 
US Census Bureau. “About Race.” About, United States Census Bureau, 23 Jan. 2018,
https://www.census.gov/topics/population/race/about.html.
 
“18 U.S. Code § 1111 - Murder.” Legal Information Institute, Cornell Law School, 
https://www.law.cornell.edu/uscode/text/18/1111.

Hypotheses and Motivation:
The hypothesis for the experiment here is meant to answer if race has an effect on the severity of a criminal federal murder sentence in the United States. The variables we look to answer this question are Race as defined by the United States Census Bureau and Severity of the punishment measured by the United States Sentencing Commission (US Census Bureau 2018; United States Sentencing Commission, 2016). To ensure that the test is as far away from a conformation bias is to set up the framing in two types of hypotheses. The null hypothesis would be there is no racial bias in judicial decisions involving murder cases within the federal jurisdiction of the United States. The alternative hypothesis is there is racial bias in judicial decisions involving murder cases within the federal jurisdiction of the United States. The study we have conducted is like that of the United States Sentencing Commission has conducted three separate times. Once in 2010 to follow up on guideline changes, a second time in 2012 and most recently in 2016. The test they conducted was to see if changes in standards of sentencing would lower disproportionate sentencing against Blacks (United States Sentencing Commission, 2018). They were able to see a change in the sentencing of women becoming equal but not enough change among men (United States Sentencing Commission, 2018). This study along with ours allows for judicial actors to adequately assess the role that race may play within a courtroom. If the findings show support for the null it can help these actors focus more in other aspects of a case for a defendant. If the null is found to be false, then it can show that race is something to watched when bringing a defendant to court. 
Description of Concepts, Data, and Measurement:
The variables as mentioned above, are clearly defined by the United States federal government. Race, which is our independent variable, for example is defined into 6 sperate categories of White, Black or African American, American Indian and Alaska Native, Asian, Native Hawaiian and other Pacific Islander and finally two or more races (US Census Bureau 2016). All these categories are defined by the Census Bureau in order to figure out the demographics of the nation as a whole (US Census Bureau 2016). This definition allows to be as accurate within the data we are examining. The second, which is our dependent variable, variable of Severity of punishment is provided by the Sentencing Commission. This definition comes from a department that creates the baseline of sentencing given for all crimes. This allows us to discuss the specifics of sentencing for the federal government when looking at murder cases (United States Sentencing Commission, 2018). By using this scale, we are able to make the timeline of sentencing as uniform when looking at the data (United States Sentencing Commission, 2018). The final factor that is not necessarily a variable but is crucial to the study is murder. The definition that we are looking at comes from the Penal Codes that all states utilize to decide how to charge defendants based on the crime they have committed18 (U.S. Code § 1111 – Murder, 2018). The independent variable of severity of punishment can be affected by the race because people of color can subject to implicit and explicit bias. This would mean that even if a defendant had all other factors of bias which is attempted to be controlled for by excluding jurors or judges from any relation of a case. By having this procedure in place, the defendant is protected by outward discrimination while at trial. While race is a factor that while at trial it is not necessarily possible for a court to control for a defendant’s race.
Presentation of Analysis and Results:
The method of our test had to do with looking at data that come from Oregon over 21-year period. This data allowed for a representative sample size of the population due to the fact that we had both genders, a wide range of ages and mix of races. From our set of data, we were able to compare races by assigning the value of 1 to white prisoners and a value of 2 to all other races. This allowed us to create a baseline of white people who are sentenced as the standard situation of being sentenced and the rest being people of color who may or may not be affected by their race. The next part of the was to look at sentencing in years for each case which just allowed us to look at all of the sentences of the defendants and compare them. Once we had both of these values, we were able to compare the data in a linear regression model. This allowed for the data to give an accurate average of the data when comparing the variables. The data after running a linear regression model showed that we would have to accept the null hypothesis. The model was able to show that there was only .11 when looking to race and sentencing and .10 when factoring in murder. This data while is not considered to be statistically significant enough does provide an avenue for judicial actors to suggest that race is a factor that should also be controlled for when adjudicating cases. When adjudicating a matter as serious as murder it is important that bias (implicit or otherwise) be checked when determining the guilt of a defendant. 
 

Works Cited 
Berdejo, Carlos, and Yuchtman, Noam, "Crime, Punishment, and Politics: An Analysis of Political Cycles in Criminal Sentencing." Review of Economics and Statistics 95:3, 741-756. (2013). https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/TEW68X.
Editors  of the Legal Information Institute. “18 U.S. Code § 1111 - Murder.” Legal Information Institute, Cornell Law School, https://www.law.cornell.edu/uscode/text/18/1111. 
United States Census Bureau. “About Race,” United States Census Bureau, Jan 23, 2018. https://www.census.gov/topics/population/race/about.html.  
United States Sentencing Commission “Demographic Differences in Sentencing.” United States Sentencing Commission, Accessed Jan 24,  2018. https://www.ussc.gov/research/research-reports/demographic-differences-sentencing. 
United States Sentencing Commission. “Offense Conduct,” United States Sentencing Commission, 54-55,Accessed Nov 23, 2016. https://www.ussc.gov/sites/default/files/pdf/guidelines-manual/2016/CHAPTER_2.pdf.