Once children are prosecuted as adults, they become subject to the same prison sentences that can be imposed on adults, including in forty-two states, the sentence of life without parole. Only Kentucky, New York, Oregon, and the District of Columbia specifically exclude anyone under the age of eighteen who is tried as an adult from life without parole sentencing. In twenty-seven of the forty-two states in which youth can be sentenced to life without parole, the sentence is mandatory for anyone, child or adult, found guilty of certain enumerated crimes.44
As of 2004, there were at least 2,225 youth offenders serving life without parole in U.S prisons.45 Because of the absence of any national database tracking the sentencing of youth to life without parole (or indeed any data tracking the presence of child offenders in adult prisons), Human Rights Watch and Amnesty International compiled this figure from data obtained directly from individual state departments of corrections and other sources. This figure includes youth offenders from forty of the forty-two states in which youth offenders may be sentenced to life without parole and from the federal bureau of prisons (see Appendix B for methods).
We have data on age at offense for 1,291 of the child offenders sentenced to life without parole. As shown in Table 2, the youngest children (six in total) were thirteen years old at the time of offense, and the average age was sixteen. Sixteen percent were imprisoned for crimes committed when they were fifteen or younger. Applying this proportion to the total number of youth offenders serving life without parole suggests that some 354 youth offenders nationwide currently face a lifetime behind bars for crimes they committed before their sixteenth birthdays.
Table 2: Age of Child Offenders Sentenced to Life without Parole at Time of Offense
Source: Data provided from thirty-eight state correctional departments and additional other sources for the states of Alabama and Virginia.
Life without parole is imposed for a variety of crimes, as shown in Table 3.46 However, it is most often imposed on child offenders who have been convicted of crimes of homicide, as shown in Figure 2.
Table 3: Crime Categories
Source: see Table 2, above.
Source: see Table 2, above.
Almost 93 percent of the youth sentenced to life without parole were convicted of homicide. It is a misconception, however, that the sentence is reserved only for the most calculated and heinous of murderers. As already emphasized, it is often imposed, for example, on children convicted of felony murderthat is, on teens who participated in a felony such as robbery during which another participant in the crime killed someone without the child offender having intended the murder to occur and sometimes without even knowing the other participant was armed. In the cases examined by Human Rights Watch and Amnesty International, many of these felony murder crimes were robberies that went awry, often involving a group of offenders, at least one of whom was an adult. Unfortunately, data are not available to enable us to determine the nationwide number of child offenders convicted of felony murder who are serving life without parole.
However, we do know that 26 percent, or 45 of the 172 youth offenders across the nation who self-reported to us on this question, were sentenced to life without parole for felony murder.47 We also know that 33 percent of the 24 youth offenders investigated in depth by Human Rights Watch in 2005 in Colorado are serving life without parole for felony murder offenses;48 and that nearly half of the 146 youth surveyed by the American Civil Liberties Union in Michigan in 2004 were sentenced to life without parole for felony murder or for aiding and abetting a murder in which another person pulled the trigger.49
In terms of gender, all but a tiny fraction (2.6 percent) of the child offenders serving life without parole are male.50 This is not surprising considering both the marked differences in violent juvenile crime rates (especially homicide offenses) by gender51 and that boys are much more likely to be transferred to adult court than girls.52 Although their total numbers were very small, the girls in the sample were more likely to have been convicted of homicide. Eight percent more female than male youth offenders serving life without parole were convicted of homicide.53 The actual discrepancy may be even greater, since crime data on 18 percent of the female offenders is not available.
The specter of super predators created much of the national furor over youth violence. Politicians and the public thought their communities were (or would be) besieged by vicious teenagers with long records of crime. Yet few of the child offenders sentenced to life without parole fit this super predator profile. Our research suggests that 59 percent of youth offenders received a life without parole sentence for their first-ever criminal conviction of any sort. These youth had neither an adult criminal record nor a juvenile adjudication.54 The other 39 percent had prior criminal records that ranged from convictions as adults for serious crimes such as robbery, to juvenile offenses such as getting into fights with other teenagers.55
T. was age thirteen in this photo and committed his crime at age fourteen.
Stacey T. was about to enter the tenth grade at a Pennsylvania high school, under a magnet program for students who excelled in school, at the time of his crime. He lived at home with his mother, a single parent. When Stacey T. was fourteen years old, he was arrested for the murder of Alexander Porter, a young man who was his girlfriends brother. Stacey was convicted of second-degree murder (felony murder in Pennsylvania) and sentenced to life without parole. He had no juvenile record, and this crime was his first. Stacey was charged directly in adult court; he never had a juvenile transfer hearing.
Court documents and an interview with Staceys attorney for this report indicate that Stacey T. agreed to participate with two adults (Henry Daniels, who was Staceys cousin, and Kevin Pelzer) in a robbery of Alexander Porter, who was assumed to be wealthy, because it was common knowledge that his family was involved in drug dealing. 56 The plan involved coercing Porter to give over the keys to his apartment so that the two adults (Daniels and Pelzer) could rob it.
The courts decision in the case of Staceys cousin establishes that the three perpetrators: [S]et up a purported drug transaction with Porter, in order to lure him to a meeting, whereupon they bound and gagged him, confiscated his keys, and stuffed him in the trunk of his car. One of the conspirators [Stacey], a friend of Porters, allowed himself to be tied up in front of Porter. . . . He was released after Porter was locked in the trunk, then taken home so that
Porter would later believe that he had been murdered. The remaining conspirators drove Porters car, with Porter in the trunk, to the garage of one of the individuals and parked it there.57
Stacey explained that Daniels and Pelzer pretended to murder him in order to coerce Porter to give over the keys or face the same fate. In Pelzers appeal, the court described what Pelzer and Daniels did in the twenty-four hours after Stacey left the scene: Twice during the next twenty-four hours while Porter was kept in the car trunk, the kidnappers used Porter's car on excursions. . . . First, they used the vehicle to get to Porter's parents' apartment to commit burglaries. . . . [Pelzer] and Daniels went home, slept for a few hours, then took Porter to a park, shot him four times in the neck and back with a .25 caliber handgun, and left him by the roadside where his body was discovered the following day. . . . While Porter was being bound, [Stacey] was led outside, supposedly to be punished, but actually to be released.58 [According to Pelzer] [M]e and [Daniels] got into the boy's car, the black shiny one to drive [Stacey] home. I drove the car. We dropped [Stacey] off and returned to my house.59
In short, while Stacey T. had agreed to participate in a robbery scheme, he was not present for the murder of Porter, nor was there evidence presented at trial that suggested he knew Daniels and Pelzer were going to murder Porter. Indeed, it would appear that the murder itself was never planned as part of the scheme. Stacey was convicted of second degree murderPennsylvanias equivalent of a felony murder conviction in other states, and which has put him behind bars for the rest of his life.
Stacey wrote: Convinced that I could make some money, I agreed with my cousin to rob this guy of his keys so that [my cousin] and his friend could rob the guys and his fathers apartment. . . . but I had no idea that this guy would end up dead. . . . Yes, I made a mistake. I associated with the wrong crowd. I engaged in committing a crime with them. However, is it fair that I spend the rest of my life in prison for a crime which was committed by someone else without my knowledge or without me being present? I feel sorry for the life which was lost in my case. I feel a deep sense of empathy for his family and what they must continue to endure in terms of pain. But this tragedy was never supposed to happen. I dont absolve myself of all guilt. I, out of naiveness, out of influence, out of the ignorance of knowing the consequences, agreed to do a crime: a robbery.60
As shown in Figure 3, from 1962 until 1981, an average of two youth offenders in the United States entered prison each year with life without parole sentences. Beginning in 1982, the number began to rise markedly, peaking at 152 youth in 1996. Although the number has declined since 1996, it has never returned to the much lower figures from the 1960s to mid-1980s.
While the absolute number of youth sentenced to life without parole has decreasedsince 1996, the nationwide proportion of youth sentenced to life without parole for murder has increased relative to the total number of youth arrested for or reliably implicated in murders nationwide (known murder offenders).61
Source: Data provided from thirty-eight state correctional departments and additional other sources for the states of Alabama and Virginia.
Human Rights Watch and Amnesty International have compared the number of known murder offenders (including all degrees of murder and felony murder) who were below the age of eighteen for each year from 1980 through 2000, with the number of child offenders who entered prison during those same years with a life without parole sentence. As shown in Table 4, the proportion of youth murder offenders who entered prison with life without parole sentences constituted an ever growing proportion of the number of known youth murder offenders. For example, the percentage going to prison with life without parole in 2000 was three times greater than the percentage in 1990. The data thus suggest an increasing punitiveness toward youth murder offenders.
Table 4: Youth Murder Offenders and Youth Offenders Sentenced to Life without Parole
Source: H. Snyder, T. Finnegan, Y. Wan, and W. Kang, "Easy Access to the FBI's Supplementary Homicide Reports: 1980 2000," 2002, available online at: http://ojjdp.ncjrs.org/ojstatbb/ezashr/ , accessed on September 14, 2005 (using data from the Federal Bureau of Investigation, Supplementary Homicide Reports 1980-2000 [machine-readable data files]). Data on dates of entry to prison contained in data provided by thirty-eight state correctional departments and additional other sources for the states of Alabama and Virginia.
Comparing the imposition of life without parole sentences on children and adults convicted of murder casts additional light on the increasing punitiveness toward child offenders. As shown in Figure 4, in eleven out of the seventeen years between 1985 and 2001, youth convicted of murder were more likely to enter prison with a life without parole sentence than adult murder offenders.62 Even when death sentences are included, as shown in Figure 5, in one quarter of the same seventeen years, child murder offenders were more likely to receive either the death penalty or life without parole than adults. In the remaining years, adults were only slightly more likely to enter prison with either life without parole or death sentences (between 1.3 and 0.1 percentage points)a remarkable finding given that during most of the years studied, large numbers of states had abolished the juvenile death penalty. On its face, this data suggests that states have often been more punitive towards children who commit murder than adults. At the very least, it suggests age has not been much of a mitigating factor in the sentencing of youth convicted of murder.
Source: The data are from the National Corrections Reporting Program, which is sponsored by the Bureau of Justice Stastics. NCRP data downloads are available online at: http://www.icpsr.umich.edu/NACJD/NCRP/, accessed on September 6, 2005.
Source: see Figure 4, above.
There is wide variation among the states in the number of youth offenders serving life without parole sentences, as shown in Table 5, below.
New Jersey, Utah, and Vermont all have laws allowing life without parole for child offenders, but as of the end of 2003, none of them had any youth offenders serving the sentence. For this report, the federal bureau of prisons reported that they had zero youth offenders serving life without parole. However, we have since located at least one inmate, Jose A., who was fifteen at the time of his crime and is serving life without parole in the federal system.63
Table 5: Total Youth Serving Life without Parole by State
Source: Data provided by thirty-eight state correctional departments and additional other sources for the states of Alabama and Virginia.
In Figure 6, below, we present the rates at which states impose the sentence relative to state youth populations. The range in the rates is extraordinary: Louisana has the highest rate, 109.6 per 100,000 youth age fourteen to seventeen, a rate that is 730 times larger than Ohios lowest rate of 0.15. The national rate is 14.20.
Source: Data provided by thirty-eight state correctional departments and additional other sources for the states of Alabama and Virginia. Population data were obtained from Population Division, U.S. Census Bureau, Table 2: Annual Estimates of the Population: April 1, 2000 to July 1, 2003, released in September 2004, available online at: http://www.census.gov/popest/states/asrh/SC-EST2003-02.html, accessed on September 6, 2005.
The differences in the state rates of life without parole for youth do not correlate directly to differences in rates of violent crime by youth. As shown in Figure 7, there are states with high rates of youth violence but low rates of youth with life without parole sentencing such as Delaware, Illinois, and Maryland; and there are states with the reverse: high rates of youth with the sentence and lower rates of youth crime, such as Pennsylvania, and Michigan. Michigan’s rate of youth crime is equal to New Jersey’s; yet, Michigan has sentenced 306 youth to LWOP whereas New Jersey currently has zero youth serving the sentence. Missouri, on the other hand, has both relatively high rates of youth crime and high rates of sentencing youth to life without parole.
Source: sentencing rate data provided by thirty-eight state correctional departments and additional other sources for the states of Alabama and Virginia. Arrest rate data from Federal Bureau of Investigation (FBI), Crime in the United States, 1997, Uniform Crime Reports (UCR), available online at: http://www.fbi.gov/ucr/97cius.htm, accessed on September 14, 2005. Population data from the Bureau of Census, Estimates of the population of states: 1997. Although many states had crime index data available for subsequent years, the data from 1997 provided rates for the largest number of states, with reporting coverage ranging from 33 to 100 percent of counties. No 1997 data were available for Florida, New Hampshire, and Vermont.
State criminal justice policies and practices clearly play a large role in the different rates of life without parole sentences for youth. One of those policy choices is that of making life without parole a mandatory sentence for certain crimes, regardless of whether it is committed by an adult or a youth. As shown in Table 6, the eight states with the highest rates of sentencing youth to life without parole all make the sentence mandatory upon conviction for certain crimes. The five states with the lowest rates of sentencing youth to life without parole (other than those that do not impose the sentence on youth at all) make the sentence discretionary. In these states, in which the judge retains the ability to weigh individual characteristics of defendants, the much lower rates of sentencing suggests judges assessments that life without parole is not an appropriate sentence for youth offenders.
Table 6: Mandatory or Discretionary Life without Parole by State
*Rate per every 10,000 youth aged 14-17 in state population as of the 2000 Census.
Source: Sentencing statutes of forty-one states. For state totals, data provided by thirty-eight state correctional departments and additional other sources for the states of Alabama and Virginia.
No examination of criminal justice in the United States is complete without a discussion of race. Therefore, we collected data on the total number of youth offenders in each racial group serving life without parole. Our data reveal that blacks constitute 60 percent of the youth offenders serving life without parole nationwide and whites constitute 29 percent.64 In addition, the data show that black youth nationwide are serving life without parole sentences at a rate that is ten times higher than white youth (the rate for black youth is 6.6 as compared with .6 for white youth). Neither the data we compiled nor other available sources answer the key question: are children from racial minorities sentenced to life without parole more frequently than white children convicted of similar crimes and with similar criminal histories?65
As with the national totals given above, the percentage of minority youth serving life without parole are often very different from the percentage of white youth serving the sentence in a particular state. Again, while the differences are dramatic, we do not know the crime rates, criminal histories, or other race-neutral factors that would allow us to draw conclusions about racial disparities in the sentencing policies of states. However, research studies have found that minority youths receive harsher treatment than similarly situated white youths at every stage of the criminal justice system, from the point of arrest to sentencing.66 For example, Amnesty Internationals research indicates that one reason for the over-representation of black and other minority children in the criminal justice system is racial discrimination by law enforcement and justice authorities.67 In addition, in a study of youth in Floridas juvenile justice system researchers found:
[W]hen juvenile offenders were alike in terms of age, gender, seriousness of the offense which promoted the current referral, and seriousness of their prior records, the probability of receiving the harshest disposition available at each of several processing stages was higher for minority youth than for white youth.68
Table 7 presents, by state, the racially disaggregated rates of youth sentenced to life without parole per 10,000 youth aged fourteen through seventeen.69
Table 7: Rate of Youth Offenders Serving Life without Parole by Race and State
*Rate per every 10,000 youth aged fourteen to seventeen in state population as of the 2000 Census
Source: Data provided by thirty-eight state correctional departments and additional other sources for the states of Alabama and Virginia. Population data were obtained from the U.S. Census Bureau, State Population Data Sets, available online at: http://www.census.gov/popest/states/asrh/files/SC-EST2003-race6-AL_MO.csv and http://www.census.gov/popest/states/asrh/files/SC-EST2003-race6-MT_WY.csv, accessed on March 4, 2005. Calculations are based on Census 2000 data.
In every single state, the rate for black youth sentenced to life without parole exceeds that of white youth. The highest black rate in an individual state40.5 (in Iowa)is just under nine times greater than the highest white rate of 4.2 (in Louisiana). The highest Hispanic rate of 16.6 (in North Dakota) is 3.6 times greater than Louisianas rate for whites.
Table 7 also indicates that there is a sizeable spread between the highest and lowest rates within racial groups, but the differential for black youth is by far the greatest. For black youth, the life without parole rates range from 40.6 in Iowa to .3 in Tennessee. For Hispanics, the highest rate is 16.6 in North Dakota, and the lowest is .6 in Florida. Finally, for white youth, the range is much smaller, from 4.2 in Louisiana to .04 in Minnesota.
In Figure 8, we have ordered the states according to the size of the ratio of their black rates of youth offenders sentenced to life without parole compared to white rates.
Source: See Table 7, above.
The highest black/white ratio is that of California: black youth in that state are 22.5 times more likely to be serving a sentence of life without parole than white youth. Mississippis black to white ratio is the lowest in the nationblacks are only 1.7 times more likely to be serving the sentence than whites. Our inability to draw conclusions about racial disparities in sentencing from this data highlights the need for states, the federal government, and independent experts to compile disaggregated data on this issue.
Emily F. was age
fourteen in this photo and age fifteen when she committed her crime.
Emily F. was reportedly violently raped when she was twelve years old. She was placed by her mother in a group home in Iowa, where she was treated for depression. Emily, who committed her crime in 1994 when she was fifteen years old, believes her crime is linked to her use of Prozac, a medication that is used to treat depression by increasing the amount of serotonin, a natural substance in the brain that helps maintain mental balance. 70
According to Emily, she told her doctor three times prior to her crime that she was having violent hallucinations that she feared were triggered by the Prozac. Emilys mother also told the doctor that she herself had to be taken off of Prozac because of negative side effects. 71 Five days after that meeting, Emily was arrested for murdering her aunt, a woman that she had no reason to harm and had spent every summer and most weekends with since [she] was three years old. 72
At Emilys trial, contradictory evidence was introduced about whether her crime was premeditated or an impulsive reaction to her medication. One resident of the group home alleged that Emily told her of a plan to kill her aunt and
to take her money, and mentioned this plan several times in the days and hours leading up to the crime. This same resident noticed defendant appeared more and more volatile and upset near to the day of the crime. 73
However, Emilys psychiatrist testified that she did not know the difference between right and wrong and was incapable of understanding the nature of her acts on October 25 because she was in a psychotic state. 74 He stated that just prior to the murder, Emily experienced an affective storm which totally overwhelmed all other operations of her brain, leaving her in a dangerous limbic psychotic state before murdering [her aunt]. 75 However, the states psychiatrist, Dr. Taylor, disputed any claim of insanity and noted the various medications defendant was then receiving, including Prozac, would not have had any adverse consequences for her. 76 Dr. Taylor also noted that the defendant was receiving Thorazine at the time of the murder, which he said had the effect of making one less inclined to be aggressive.
 These states are: Alabama, Arizona, Arkansas, Colorado, Connecticut, Delaware, Florida, Georgia, Idaho, Illinois, Iowa, Louisiana, Maryland, Massachusetts, Michigan, Minnesota, Missouri, Nebraska, New Hampshire, North Carolina, Pennsylvania, Rhode Island, South Carolina, South Dakota, Virginia, Washington, and Wisconsin.
 All data discussed in this chapter are on file with Human Rights Watch. Please see Appendix B for a detailed description of research and statistical methods.
 Each state department of corrections has its own method for coding the type of crime committed by its prisoners. Some states break homicide crimes down into categories (e.g., first degree murder, second degree murder, etc.), and some simply group all homicide crimes into the general category of murder. Moreover, in many states the category of first degree murder or murder includes both intentional homicide and the felony crime described in the text as felony murder. Therefore, a variety of codes or offense terms are used by states to describe the crime for which each incarcerated individual was convicted. This makes it difficult to determine which types or sub-categories of youth crimes tended to result in a life without parole sentence.
 This sample was derived from a survey of 281 letters received by Human Rights Watch from youth offenders across the United States, 172 of whom gave detailed explanations of their role in the crime. These prisoners serving life without parole who wrote to Human Rights Watch were a self-selected group, not a random sample. We did not expressly ask for information about the level of culpability of the offender (i.e., whether the individual was or was not the triggerperson in a felony murder crime), and therefore prisoners were not aware that this was a question we were interested in, so it is possible that we received more correspondence from those individuals who felt their sentence was disproportionate. In particular, it is possible that individuals serving life without parole for felony murder convictions were more likely to write to Human Rights Watch out of a belief that their sentence was unfair.
 Thrown Away: Children Sentenced to Life without Parole in Colorado, A Human Rights Watch Report, February 2005, p.18-19.
 Second Chances: Juveniles Serving Life without Parole in Michigan Prisons (ACLU of Michigan, 2004), p.4, available online at: www.aclumich.org/pubs/juvenilelifers.pdf, accessed on September 13, 2005.
 This percentage reflects the gender of child offenders serving life without parole in thirty-nine out of the forty-two states that impose the sentence. We had no data from Idaho and no data on the gender of youth offenders from Virginia.
 For example, male youths are about thirteen times more likely to be arrested for murder than females. U.S. Department of Justice, Office of Juvenile Justice and Delinquency Prevention, Juvenile Offenders and Victims, National Report Series, December 2001, available online at: http://www.ncjrs.org/html/ojjdp/nrs_bulletin/nrs_2001_12_1/contents.html, accessed on September 13, 2005.
 Eileen Poe-Yamagata and Jeffrey A. Butts, Female Offenders in the Juvenile Justice System: Statistics Summary (Office of Juvenile Justice and Delinquency Prevention, June 1996), p. 13, available online at: http://nicic.org/Library/013515, accessed on September 13, 2005.
 The gender and crime of its child offenders serving life without parole in Virginia are not included because the state did not provide us with that data.
 This sample was derived from a survey of 281 letters received by Human Rights Watch from youth offenders across the United States, 96 of whom gave detailed explanations of their previous criminal offense history. These prisoners serving life without parole who wrote to Human Rights Watch were a self-selected group, not a random sample. We did not expressly ask for information about offenders previous criminal histories, and therefore prisoners were not aware that this was a question we were interested in, so it is possible that we received more correspondence from those individuals who felt their sentence was unfair. In particular, it is possible that individuals serving life without parole for their first criminal offense were more likely to write to Human Rights Watch because of this sense of receiving an unfair sentence.
 We are aware of only one offender in our sample who was previously convicted as an adult of first degree murder before the first degree murder crime that led to his sentence of life without parole.
 Human Rights Watch telephone interview with Mitchell Strutin, October 28, 2004. See also Pennsylvania v. Pelzer, 531 Pa. 235 (May 29, 1992).
 Pennsylvania v. Daniels, 531 Pa. 210, 217 (Penn. S. Ct., May 29, 1992).
 Pennsylvania v. Pelzer, 531 Pa. 235 (May 29, 1992).
 Letter to Human Rights Watch from Stacey T., Chester, Pennsylvania, May 20, 2004 (on file with Human Rights Watch).
 The number of known murder offenders is the best proxy for the number of youth convicted of murder, because specific data for murder convictions of youth offenders does not exist. Known murder offenders is a term used by criminal justice professionals and includes not only all individuals arrested for murders (including all degrees of murder and felony murder) but also individuals identified by witnesses or reliably identified as perpetrators but not arrested (usually because the offender was killed). For more information on the methods by which known offenders are classified, see Easy Access to the FBI's Supplementary Homicide Reports: 1980-2000, U.S. Department of Justice, Office of Juvenile Justice and Delinquency Prevention, available online at: http://ojjdp.ncjrs.org/ojstatbb/ezashr/asp/methods.asp, accessed on July 22, 2005.
 There are no published data that identify the number of adults receiving life without parole sentences in the United States. We used as a proxy the number of adult offenders entering prison with life without parole sentences for murder convictions. The proxy should be extremely close to the actual number, since only death or some other extraordinary development would prevent someone sentenced to life without parole from actually entering prison.
 Letter to Human Rights Watch from Jose A., United States Penitentiary Allenwood, White Deer, Pennsylvania, March 9, 2004 (on file with Human Rights Watch).
 Native Americans constitute .8 percent, and Asian Americans are .9 percent of the total number of youth offenders serving life without parole sentences. Note that the 2000 Census used separate categories for race (White, Black, American Indian, Hawaiian, and Asian) and ethnicity (Hispanic or Latino and non-Hispanic or Latino). Therefore, all people who identified themselves as Hispanic, regardless of their race identification are counted in the Hispanic populations. Those people who identified themselves as White-Hispanic are also counted in the White population; those people who identified themselves as Black-Hispanic are also counted in the Black population; and so on.
 After combing existing datasets on crime rates, Human Rights Watch spoke with Tom Zelenock, the project leader of the National Archive of Criminal Justice Data project at the University of Michigans Inter-University Consortium for Political and Social Research. He stated that Uniform Crime Reports (UCR) Data provide state break-downs of overall race of offenders and state break-downs of age of offenders, but UCR data does not provide a state break-down of the race of offenders of (or below) a particular age. This is a widely acknowledged shortcoming in the data. Zelenock said he is sure people have tried to extract juvenile race data and have done it incorrectly, because it is just not possible from data currently collected in the United States. Human Rights Watch telephone interview with Tom Zelenock, on March 24, 2005.
 See., e.g., Eileen Poe-Yamagata and Michael A. Jones, And Justice for Some (Building Blocks for Youth Initiative for the National Council on Crime and Delinquency, 2000), available online at: http://www.buildingblocksforyouth.org/justiceforsome/jfs.html, accessed on September 14, 2005 (finding that youth of color are overrepresented and receive disparate treatment at every stage of the juvenile justice system); Mike Males and Dan Macallair, The Color of Justice: An Analysis of Juvenile Adult Court Transfers in California (Justice Policy Institute, Building Blocks for Youth Initiative, Feb. 2000), available online at: http://www.buildingblocksforyouth.org/colorofjustice/coj.html, accessed on September 14, 2004 (The Color of Justice) (showing that youth of color are 8.3 times more likely than white youth to be sentenced by an adult court to imprisonment in a California Youth Authority facility); Jolanta Juszkiewicz, Youth Crime/Adult Time: Is Justice Served? (Pretrial Services Resource Center, Building Blocks for Youth Initiative, Oct. 2000) available online at: http://www.buildingblocksforyouth.org/ycat/ycat.html, accessed on September 14, 2005 (showing over-representation and disparate treatment of youth of color in the adult system and questioning the fairness of prosecuting youth as adults).
 Betraying the Young: Children in the U.S. Justice System (Amnesty International, November 20, 1998), available online at: http://web.amnesty.org/library/Index/engAMR510601998, accessed on September 14, 2005.
 Donna Bishop and Charles Frazier, A study of race and juvenile processing in Florida, a report submitted to the Florida Supreme Court Racial and Ethnic Bias Study Commission, 1990 (cited in he Color of Justice).
 In both Table 7 and Figure 7, we included only those states that had at least one child offender of each race in the rate comparisons.
 U.S. National Library of Medicine and U.S. Institutes of Health, Fluoxetine, brand names Prozac, Sarafem, MedlinePlus Drug Information, available online at: http://www.nlm.nih.gov/medlineplus/druginfo/medmaster/a689006.html, accessed on April 15, 2005.
 In October 2004, the U.S. Food and Drug Administration (FDA) asked antidepressant manufacturers to include warnings about possible side effects, specifically suicidal tendencies in children taking these medications. In a press statement, the FDA stated: Pediatric patients being treated with antidepressants for any indication should be closely observed for clinical worsening, as well as agitation, irritability, suicidality, and unusual changes in behavior, especially during the initial few months of a course of drug therapy, or at times of dose changes, either increases or decreases. See FDA Public Health Advisory, Suicidality in Children and Adolescents Being Treated With Antidepressant Medications, October 15, 2004, available online at: http://www.fda.gov/cder/drug/antidepressants/SSRIPHA200410.htm, accessed on September 15, 2005.
 Letter to Human Rights Watch from Emily F., Iowa Correctional Institute for Women, Mitchellville, Iowa, March 13, 2004 (pseudonym) (on file with Human Rights Watch). Human Rights Watch was also contacted by a young man serving a life without parole sentence in Alabama and his mother, both of whom link his criminal offense to the fact that he was on an antidepressant medication at the time.
 Iowa v. Fetters, 562 N.W.2d 770 (Iowa Ct. App. Feb. 26 1997).