Skip navigation.
Home | About Us | Contact Us
Rick Scott, Governor
Florida Department of Corrections, Secretary Michael D. Crews

Florida Department of Corrections
Michael D. Crews, Secretary

Detailed Response to
The Miami Herald Article

March 21, 2001
Department of Corrections Detailed Response

"KIDS IN PRISON", page 2 of 10, first bulleted item:

Herald Statement:
"...homicides comprise just two of every 100 juvenile adult court convictions. Less-serious burglaries, drug offenses and thefts account for the majority."

DOC Response:
These figures may not be completely accurate or tell the entire story concerning the seriousness of the offenses for which juvenile offenders are sentenced as adults. The Herald's data source is the sentencing guidelines database which is not complete (discussed later). We examined juvenile admissions to the department (prison and supervision) which is very complete and accurate. Of the 21,380 offenders sentenced to DOC from 1995 to 1999 who were juveniles at the time of their crimes, 36% were sentenced for less serious burglaries, drug offenses, or thefts. However, if you consider sale, manufacturing, and trafficking in drugs as serious crimes, the proportion drops to 25%. Therefore, The Herald may be overstating the proportion of juveniles treated as adults who are less serious offenders.

Also, when you look at prison admissions from 1995 to 1999, shows that the juvenile offenders are much more likely than adult offenders to have committed more serious and violent offenses. We found that 8.1% of juvenile prison admissions were convicted of murder or manslaughter, while only 4.6% of adult admissions were convicted of this type of crime. In calendar year 2000, the discrepancy is even more pronounced: 7.5% of juvenile-at-offense inmates vs. 3.8% of the remaining inmate population are in for murder or manslaughter offenses. The comparison below further supports the fact that juveniles entering prison are much more likely to have committed more serious crimes than adults entering prison during the period from 1995 to 1999.

The top five offense categories for each group follow:

Juvenile (17 or younger) at Offense
Robbery with Weapon (17.9%)
Burglary of Dwelling (12.4%)
Drugs-Manufacture/Sale/Purchase (7.7%)
Armed Burglary (6.5%)
Robbery without Weapon (6.0%)


Adult at Offense
Drugs-Manufacture/Sale/Purchase (15.1%)
Burglary of Dwelling (8.9%)
Drug Possession (7.4%)
Burglary of Structure (5.6%)
Stolen Property (3.9%)

"KIDS IN PRISON", Page 3 of 10 - First and fourth Paragraph:

Herald Statement:
"Young convicts leaving adult prison are more likely to continue with crime than juveniles accused of similar crimes leaving juvenile programs, studies by the The Herald and two sets of researchers found. The conclusion: Children sent to juvenile programs are more likely to stay out of trouble than those routed to prison."

DOC Response:
The study conducted by The Herald is not scientifically rigorous enough to make such an unequivocal conclusion. To ensure that you have comparable groups (except for the fact that some offenders went through the DJJ system and others went through the DOC system) it is critical that all factors known to be highly associated with recidivism are accounted for and accurately measured. This is not the case in this instance for two reasons.

First, only the variables of "age, prior records, race and severity of crime" were measured by The Herald. The study failed to account for variables known to be associated with recidivism such as gender, family background, program and work involvement in custody, educational level achieved, and institutional adjustment. Excluding these recidivism predictors calls into question the results.

Second, the validity of how well The Herald measured "severity of crime," and "prior records" is questionable. The Sentencing Guidelines data used indicates the type of crime, such as manslaughter, burglary of a dwelling, etc. This does not provide indicators of how many victims were involved, the level of injury to victims, if a gun was used, and a host of other facts that describe the true seriousness of criminal acts. Also, the guidelines data does not necessarily contain a complete picture of offenders' prior records, which is a critical variable in explaining recidivism rates.

We responded to The Herald in November 2000 when they indicated they were going to rely on the study entitled "The Transfer of Juveniles to Criminal Court: Does it Make a Difference?" by Donna Bishop, et. al. (1996) to address the issue of recidivism amongst DOC and DJJ offenders. We stated the following:

This is of little relevance to the question of whether juvenile offenders should be transferred to adult court today. First, while the results of the study were published in 1996, the study is based on youths sentenced in 1987. This was prior to the establishment of the Department of Juvenile Justice. Also, the adult prison system, as the authors point out, was plagued by early prison release, brief lengths of stay in prison, and inmates serving less than 30% of their sentences. This lenient punishment system resulted in an inability to adequately prepare offenders using academic improvements, vocational training, etc., prior to prison release. Prison cannot have a significant deterrent effect on offenders if many of them were incarcerated for less than six months.

Second, the authors admit they did not adequately measure the correct amount of time that released offenders were on the street and exposed to opportunities to re-offend. Also, their measurement of prior record is flawed because they match cases based on 0, 1, 2 or 3 or more prior referrals to the juvenile justice system. Their data shows that 57.2% of the cases had 3 or more prior referrals. Matching transferred and non-transferred cases in the 3+ priors group is flawed because one can-not consider these offenders as necessarily equivalent. It is possible that the prior record of those offenders transferred to adult court were, in fact, significantly longer than those who were not transferred.

"KIDS IN PRISON", page 5 of 10, first bulleted item:

Herald Statement:
"Miami's Joseph Tejera is one of Florida's teen inmates, and he fits the state profile perfectly. Tejera was prosecuted for burglary, the most common juvenile adult court conviction, nearly one in four cases."

DOC Response:
While it is true that nearly one in four of every juvenile offender admitted to supervision committed a burglary; there are several types of burglary. Only 8.8% sentenced during 1995-99 committed burglary of a dwelling or occupied conveyance, the type committed by Mr. Tejera. It should also be noted that in addition to burglary of a dwelling, Mr. Tejera was also convicted of grand theft and criminal mischief. Under the Criminal Punishment Code (the sentencing law in effect at the time of offense) the burglary of a dwelling, a level 7 offense, is sufficient in and of itself to warrant a prison sentence. Mr. Tejera was given a mitigated sentence to supervision which, upon his technical violation, resulted in the revocation of his supervision and the imposition of a mandated state prison sentence.

"YOUNG INMATES REPORT HIGHEST RATE OF ASSAULT", Page 2 of 5 - Second Paragraph:

Herald Statement:
"The DOC logged 362 assault complaints involving juvenile victims in the five-year period - one for every two juvenile inmates."

DOC Response:
The Herald incorrectly calculated the rate of assaults alleged by adult or juvenile inmates against juvenile inmate. They divided the number of alleged assaults in which the victim was a juvenile (362) by the average inmate population over the five year period (622). There are two problems with this calculation. First, there were a total of 3,631 different inmates in prison from 1995 to 1999 who were juveniles at some point during their incarceration. Therefore, the report should have stated "one for every 10 juvenile inmates" (362/3,631) instead of "one for every two inmates." The only way the number The Herald is quoting would be accurate is if the 622 juvenile inmates were the same inmates over the entire five-year period. This is obviously not correct. An analogy would be if you wanted to calculate the cost per meal in a restaurant and used the average number of customers per night (15) and the restaurant's gross income over a year was $100,000. Using The Herald's math, you would say that the average customer spent $6,667 per meal ($100,000/15). In reality, there were a total of 5,475 customers (15 * 365 days) and the average meal cost $18.26 ($100,000/5,475 meals). Second, The Herald assumed the 362 assault allegations are all different inmates. In fact, approximately 15% of these alleged assault cases would have been eliminated if the records involving the same inmates were deleted. If this had been considered, the ratio would have been closer to one in 12 instead of one in two.

"YOUNG INMATES REPORT HIGHEST RATE OF ASSAULT", Page 4 of 5 - Ninth Paragraph:

Herald Statement:
"However, the DOC could not provide the number of reported assaults sustained between 1995 and 1999, saying the data are not computerized."

DOC Response:
We stated the following to your question about whether there is information on whether reported assaults were sustained in an e-mail message to Geoff Dougherty from the department's Chief of Research, Bill Bales on March 16, 2001. "The data in a fixed-field format on whether a case is sustained or not sustained has only been available since September 2000, which is beyond the period of time you studied. The outcome of cases has historically only been recorded in a text format."

To state that "the data are not computerized" is inaccurate, as was reported to The Herald in writing several days before the article was printed. The information is contained in the Inspector General's database that is used for management purposes by the department. However, until the department's new administration made it available in a fixed-field format (i.e., numerical) amenable for research purposes, it was available in text form. Also, with some additional effort, The Herald researchers could have searched the text field in the database to determine if the assault allegations were substantiated.

"KIDS IN PRISON: A PROFILE", page 1 of 1, 2nd chart from the bottom on the right:

Herald Statement:
"40 percent of children convicted in adult court get probation. Yet many end up in prison. Children tend to violate rules of probation and community control, for everything from failing to be home by nightfall to committing new crimes."

DOC Response:
Our analysis of the guidelines data indicates that 61.8% of juvenile of-fenders receive a supervision sentence (46.4% probation, 15.4% community control), not the 40 percent reported by The Herald. Also, the chart in the profile indicates that 33% of the youths versus 25% of the adults placed on supervision violate their supervision term for technical reasons or due to a new crime. If fact, these figures, do not include violations for new crimes which could affect the comparison in failure rates for juveniles and adults.

Also, with the statement that "many end up in prison", The Herald is providing their readers with the wrong impression. In fact, our data shows that only 8.3% of the juvenile offenders who were placed on supervision and were technically violated (i.e., did not commit a new crime) were subsequently sentenced to prison. In comparison, 13.3% of the adults who technically violated were sentenced to prison. Therefore, adult offenders who violate the conditions of their supervision are more likely to be sentenced to prison than juvenile offenders.

"HOW THIS STORY WAS DONE", Page 1 of 2 - Second Paragraph:

Herald Statement:
"The department could not produce reliable data on how many juveniles in prison are sexually assaulted..."

DOC Response:
This is not true. The department cannot legally provide any information that could lead to the identification of the victims of alleged sexual assaults. There-fore, the department could not provide The Herald with personal information that would lead to the identification of these offenders, such as date of birth. However, the department did provide a document to The Herald which displayed the number of victims, by each age, for each year from 1993 to October 2000 for investigations of sexual offenses. Of the 506 investigation over this entire period, only 7 (1.4%) involved a juvenile victim. The Herald chose not to present these facts to their readers.

"HOW THIS STORY WAS DONE", Page 1 of 3 - First Bulleted Item, First Paragraph:

Herald Statement:
"The paper assessed the prior criminal records of juveniles in adult court by using a DOC sentencing guidelines database, which contains criminal histories on defendants."

DOC Response:
While the sentencing guidelines scoresheet allows for the capture of prior record information, it is known from experience that this information is often lacking or incomplete. The predominant reason is that the information is not always readily available to the preparer. Therefore, to rely on this data for a complete and accurate representation of the prior criminal record of offenders is problematic.

"HOW THIS STORY WAS DONE", Page 1 of 3 - First Bulleted Item, Second Paragraph:

Herald Statement:
"That database (sentencing guidelines) covers only the most recent three years of a defendant's juvenile history."

DOC Response:
FS 921 states that "Juvenile dispositions of offenses committed by the offender within three years prior to the primary offense are included in the offender's prior record where the offense would have been a crime had the offender been an adult. Juvenile dispositions of sexual offenses committed by the offender three years or more before the primary offense are to be included in the offender's prior record if the offender has not maintained a conviction-free record, either as an adult or a juvenile."

"HOW THIS STORY WAS DONE", Page 1 of 3 - First Bulleted Item, Third Paragraph:

Herald Statement:
"The sentencing database is missing data on about 65 percent of the cases handled in Florida..."

DOC Response:
This 65 missing data rate stated by The Herald in inaccurate. The DOC is statutorily required to calculate a compliance rate for the sentencing guidelines scoresheets each year. This rate, in essence, tells how many scoresheets (to prison or community supervision) were received compared to how many should have been received. For FY1999-00, the rate of compliance was 69.6%. This means that the Department received 69.6% of the scoresheets that were used to sentence offenders to the Department during FY1999-00. This compliance rate was available to The Herald on the department's website.

"HOW THIS STORY WAS DONE", Page 2 of 3 - Last Paragraph:

Herald Statement:
In discussing their methods of studying assault rates between DOC and DJJ, The Herald states, "The Herald attempted to limit its study to DJJ cases in residential facilities. Data problems made that impossible, so the analysis includes offenders in nonresidential programs, too."

DOC Response:
This approach likely resulted in erroneous comparison figures. It seems logical to assume that juveniles NOT in residential programs would not be assaulted by other juveniles or adults if they are not confined with other offenders. This is not the case in DOC, since only inmates were examined. The Herald should have deter-mined through DJJ if assaults are reported by non-residential offenders, even if at a lower rate than residential offenders. If this is not the case, then they should have taken them out of the rate calculations. It also seems odd that the DJJ database does not include the basic information on the types of facilities in which their offenders are housed.