Full Text
L Carter Gobe Associates, Inc.. A Carter Gobie Lee CORTLAND COUNTY NEEDS ASSESSMENT Final Report January 11, 2007 ---PAGE BREAK--- CORTLAND COUNTY NEEDS ASSESSMENT . REPORT TABLE OF CONTENTS SEC1ION ONE TRENDS & PROJECTIONS I 1-1 Poulation. cr:; Stanstics . . . rcacera:iar Rate . . . . . Jail Average Daily Population...,.. . 1•-6 Admissions . 1-9 JailAverageLengthofStay .1-10 OriminaljusticeSystem . 1-12 SECTION TWO PROJECTIONS Projections . . 2-1 -1 Model 2ADM per 1000, Population 2-1 Model 3—Part 1 Arrests Per Population .2-2 .2-2 . 2-3 Model6—ARIMA 2-3 r.lndel7—.4!ES 2-3 Summar’ofResults .2-4 PeaRog and Bed Needs 2-5 Recommendation . .2-6 ---PAGE BREAK--- INTRODUCflON In order to determine the future capacity of a jail svstem we must examine ts hIstory Using trw countys population trends, crime patterns, arrest numbers, and historical jail data, we can predict the future bed needs of a jail system through careful analysis. This data is provided by various sources, as cited. Population Cortiand County has maintained a steady population between 1996 and 2005, experiencing an overali nec ease JO as sno r’ b Thble r 2003 tIe ocojiatiop g e b 09 o 442 rd duas Ths s a a ncr bange han me counts exoererced n an other ‘year from this span Ins daia s p a ded b tne US Census Bureau as reported in March 2006. Fgure I below s a graphic ihustration of the data of Table 1, reflecting the population trends of Cortland County between 1996 and 2005 it shows a reiattveiy large increase in population in 2003. and a relatively large dec ease n 2005 C eral te popu’aon has noi changed sgn’f canty s nce 1996 CORTLAND DOUR IT NEEDS ASSESSMENT FNAL REPORT TRENDS & PROJECTIONS Source US Census Bureau, March 2006 C:r’er GoNe Assocriates — ---PAGE BREAK--- The popiation growth as projected by Hudson Valley Regional Council and Cornell University s illustrated by Table 2 below. The table shows the projected population ot Cortland County in 5 year ntervals through 2025. This projected trend shows a decline in population of or 867 individuals, between 2005 and 2025. Tab!e 2 CORTLAND COUNTY NEEDS ASSESSMENT flNAL REPORT Figure 1 TRENDS & PROJECTIONS Historic Cortland County Population 46:000 45:500 $5000 iUi 155 1999 2U0J uu1 2 221 68 867 ---PAGE BREAK--- CRIMINAL JusTice Crime Reported cdme data collected from Viminia University Uniform Crime Index Report and DMsion of Criminal Justice Services is shown in Table 3 below University Uniform Crime index Report only provided data up to the year 2002, so the data for 2003-2005 was provided by the Division of Criminal Justice Services, The comes illustrated are ‘ndex crimes high level offenses compiled by the Uniform Cnme Reports (UCR) and published annually by the FBI in order to gauge fluctuations in volume and rate of reported come. index crimes are divided into two categories: violent crimes and property crimes. The violent crimes inclu.ded are “urde aoe robbery and agg avated assault wnile the oropert/ crimes inciuded are bug arv arcel theft, and motor vehicle theft. Crim.e rates are calculated as the number of crimes reported per 1000 persons. The data shows a 40% drop in the overall rate of index crimes in Cortland County; however, some types of crime are up. Forcible rape has increased by 63% since 1 996. The number of murders has increased by a deceivingly high 300% due to four murders being reported n 2005 versus one in 1996, but no murders were reported in any of the intermediary years. Arson and robbery reports have also increased, but only by two incidents each. Property crimes have taken a significant drop of 42% overall, due to a large decrease in larceny-theft and motor vehicle theft, while violent crimes as a whole have remained mostly stable, As of 2005, the crime rate is at 28 offenses per 1.000 persons — a 40% drop from the 46.3 rate in 1996. Table 3 Historic Reported Crime Idex Crirnealcnt — I I 2 Tpr I’ M’ Ii, Li , i/F .11 1’ tI i’ 5 44 i dCrneTcA — — id54j 187fr 1 7221 i63 i 1 - I 5_1114j. 33.1 1 355 341 C’ RaIr . 5 3R U 7 32.8 311 2r 251 21r CORTLANO COUNTY NEEDS ASSESSi1ENT FEAL REPORT TREiDS & ---PAGE BREAK--- CORTLAND COUNTY NEEDS ASSESSMENT flNAL REPORT TRENDS & PROJECTIONS Arrests Figure2 Sr.:es Virgne Uni.versii UnifDrm Chmo ndex Rpcd (0epcIin 199v20020 : ‘ While reported crime data is an indicator of the crime rate in Cortland County, it does not correlate directly with actual arrests. In order to predict the future needs of a facility, it is necessary to examine the actual arrest data for that community. Table 4 illustrates arrest data, also from Virginia Un;versty Uniform Cnme Index Report and Division of Criminal Justice Serices. Again, University Uniform Crime Index Report provided data up to the year 2002, and the Division of 6riminal Justice Seivices pro’Med the data for 2003- ‘The dza sc tra ‘Jor a rests are &a 1, frcm year h ear vh typiv ore ‘r a s s er 1 ea bj bat q oibe rape arrests aJe no eased s gn can’l, ears ‘ c.rime arrests overall ha.ve increased b.y 10% sini...e: 1996. Conversely. pn..•pehy arrests are down 27% since 1996. This is a result of larceny-theft arrests, the n.ost arrest-heavy index crime in the county, droppino by 43 since 1996 desnite all other property crime arrests being un The significant decrease In larceny-theft arrests has a.lso led to a decrease of 20% in overall arrests. In 2000, total arrests were up after a brie.f decline n 1999 Arrests dccl ned agan in 2001 bu over the roflowing hwo years they increased until scam. in 2umjon 2003 had a t,cctat of :381 indsix crime arrests, a 6.4% increase over the 233 arrests in Historic Reported cri iTS 1999 2000 2001 2002 2003 2004 2005 ---PAGE BREAK--- Pr-ert, otArs[ i.u’ io2 281j i 1 Sour.es VgaUie Unom Ppcit Roai.ng •S-1cC2). DMinot Se;s, Figure 3 below graphically illustrates data from Table 4, showing the number of arrests per year separated by property crime arrests, violent crime arrests, and total arrests. It shows a small spike in overall arrests in 2000, followed by a larger spike in 2003 after a twoyear increase in arrests. Figure 3 CORTLAND COUNTY NEEDS ASSESSMENT FNAL REPORT TRENDS & ROJECTONS - 74 75 1 2000 2001 2002 2003 200.4 1.002- 4’ , , , , ---PAGE BREAK--- CORTLAND COUNTY NEEDS ASSESSMENT NAL REPORT TRENDS &PROJECONS ncarceration Rate incarceration rate (IR) s calculated as the ratio of a jails average day population (ADP) to total county population, expressed as a rate per 1000 persons. Table 5 below shows that the incarceration rate for Cortiand County has not significantly changed since 1998, dropping from 1.3 inmates per 1000 persons to 1,2 inmates per 1000 persons. This stability can be explained by the countys largely unchanged county population and ADP. Table 5 Historical Incarceration Rate 3448 9 48 947 48 473 4a2i 48 92 48 922 03. Pop cerationRate’ 10 Saurceu U.S. Scares Bureau, March 2006 Cu4asd Co.r.nfy Jai{ SO. hshc& Reps:t JAIL STATISTICS A jails population is examined by three basic measures: The admissions or intake (ADM) The average daily population (ADP) The average length of stay of inmates (ALOS) Admissions is the number of inmates processed into the facility Average daily popuiaton measures th.e approximate number of inmates in the facility on a given day. Average length of stay is the average amount of days spent by inmates in the facility. The relationship between toese factors is expressed by the foliovong formula: ADP = (ADM x ALOS) . 365 Average Daily Population Cnanaes in cnme and arrest rates affect the admissions and population of a jail. The average daily JLJ dUOs :P a’ a aB s a rneasu e’oe cf ne a eagM bd court &s s tJ3ed ce a c a early basis, ADP is a census based number. rather than being formula denved. 5 FJP indicates, the toical bed soace needed by a au system, which makes it 3 yeryj5eu measure cf ce oc t s rCD a 5 c a ated uapa s cuerei a oce d a e an e c c se’ ---PAGE BREAK--- CORTLAt4D COUNIY NEEDS ASSESSMENT FINAL REPORT TRENDS & PROJECTIONS During some short periods of time, a facility may experience a brief spike in its population. For this reason, future space needs can not be predicted based on ADP alone, In order to account for these spikes. a peaking factor is calculated for each historical year The three months with the highest ADP values are averaged together as the ‘three month high. The peaking factor is then calculated as the percentage difference between the three month high and the years overall ADP. Table 6 below illustrates the ADP from 1998-2005, with the three highest months of each year highlighted. Table 6 June December 63 Annual 64 64 Sooroc: C1!end Cooty taIis7caI Report 48 47 43 49 51 57 51 5R Table 7 below illustrates the yearly ADP from 19982005. as well as the three month high and peaking factor for each year. The ADP of 2005 was 58 inmates, a 9% decrease from 1996, and about equal to the 8year average of 59 inmates. The relatively low ADP in 2001 may be a reflection of that years decreasing arrests, just as the abnormally high ADP in 20032004 may reflect the 2003 spike in arrests. The period of increasing ADP from 2001 through 2004 also follows a trend of increasing ALOS for those years. The highest peaking factor recorded was 18% in 2002, but the 8-year average is 12i%. The peaking factor for 2005 was 11%. close to the 8-year average. Table 7 Historical AOP and Peaking Factor 411 ii ‘II II II it 3 Month Peaklnr 5% rSO/ 149% 77% 1 121 0% 121% S rcer ooJ r, 5 r Sc& iaure 4 s a arnh.c 7ustratIOfl of the .p nrIrs rnUro.ic ai o?e frcm 1 ‘Jrouh 2O5. it LisrOes DOtfl and annual ADP. Historical 4 J February l March April May Daily Population 48 58 48 5’ 51 00 S 4 57 40 6 July 53 66 cugust l — 83 5: — September 68 Dctober 50 71 November - 55 63 5’ 71 51 74 56 51 — 61 58 51 ---PAGE BREAK--- Av&ajc Daily PcpWation Tw ‘ ‘ r : rpil’c ust tinn f h& merage breaKdwn 01 nmates fc’ri iniar’ 1i9ó t k;ar•M c: I flfl’iE3 hro e d’ h seneiioed. re-trtaI dnd ther v;ucn Cudcs st ‘ccJes fl. L’J ::os ln of S flfl:e tJ—tra 5 ah)Ut e 1 or. •Th . t o :‘m.o nDrciri 4’ r’d ohr prate ‘oi;rsi. Figure 5 Historic vcrag Population Breadown January 1998 March 2006 CORTLAND COUNTY NEEDS ASSESSMENT RNAL REPORT TRENDS & PRQJECTONS Figure 4 ---PAGE BREAK--- COuNTY NEOS ASSESSIIENT - T ble i-h oricA era Daii Popu tron Hi H ii H Sen e ced Pretnat ° _j__4ZZZ 291r oar a ‘nt ry b k o nf h I s OP bro n down -v re-t n, ced and ot r nmas AD° ot ore-trial is n an ncreasin0 trend, wthle ADP at sentand nnats ha d eased a t:t rn 1 o 01 U. o u Ii n r ly d rn t d b rn t s o ty tn lU” Jl ftn atudte tn dor on ype o th oth r o a f ii Sn e und t b gr g 0 rweer the pre-U.a! nouiaon has amost outnumbered the sentenced pcpuaton n all but a ev months Fig re 6 Historic Average Daily Population a’ - - c_ -N j’ N’ — mi o dm-ssnn (ADr it ar another census based number that refers to all parsons admtiJ to a ja regrdless h n diffrn bt n ndiiu I r th U. h r a zr_d t’rLner peroci nf jmp t a;t a dau )iit abef,-ta ISS fl1 J21n (J - I P- r ar: i’r r - am - : r 4 r r 1r - c — - . r - ---PAGE BREAK--- Overar Ie %T aie Carter (krbte Table 9 H on alA ion Figure 7 below s a graphic iflustration of the admissions data from Table 9, shovPng male, female, and total it shows that female admissions are relatively stable, and that total are driven by admissions of n‘tale inmates, Figure 7 Average Length of Stay Source Cortland Ccainty Jail statisroal Report 2005 Average Length of Stay (ALOS) is the average number of days that an inmate stays in the jail. It is calculated by multiplying the ADP by 365 and dividing that number by the annual admissions. The average length of stay in Cortland County is shown in Table 10 below, The county’s ALOS has seen a significant increase in the past five years, particularly in the period from 2000 to 2004, where it increased from 17 days to 2.9 days, an average increase of 3 days per year. in 2005, ALOS fell to 23 days but was a % ‘e cc 1a a erage The a erage ength of sa ver W i e °e en ea eriod a 22 days. wrile the average of the last 5 years Is 24 days A.LOS has risen 30% since 1999. Male A.[Q5 has increased 28% since 1999 Female ALOS is up as well., having increased by 62% since 1999. Male and female ALOS was unable to be calculated for 1998 due to mal.e and female ADP being unavailable for that year. A sharp increase of ALOS can indicate delays in efhcientiy moving cases through the system for the pre trial cop,ilatton or longer sentences for Lne locally senmiced peculation CURT LAND COUNTY NEEDS ASSESSMENT FINAL REPORT TRENDS & PROJECTIONS Historic Admissions Li: 1992; 2000 2001 2002 2003 2024 ---PAGE BREAK--- Table 10 Historical Average Length of Stay Cources: County J ni Stotinycat Reonut Carter CubE Aueuciateu ioure 8 below is a oraphic representation of the countys ALOS from 1999 to 2005 which shows the steady increase of ALOS from 2000 to 2004. Figure 8 Figure 9 below is a graphic representation of male and female ALOS from 1999 to 2005 It shows that the ALOS of both male and female inmates has seen an increasing trend over the last 7 years. Figure 9 CORTLAND COUNTY NEEDS ASSESSMENT F:NAL REPORT TRENDS & PROJECTtONS Historical Average Length of Stay ---PAGE BREAK--- RTLMOCCUNTYNEEDSASSESSENT FAi, REPCRT TRE[S & CORTLAND COUNTY CRIMINAL JUSTICE SYSTEM The Cordand County Jatis average daUy population is controlled by the criminal justice system within the county Historical data of crime and arrests, general population, court processing and jail population provrde critical data regarding factors that impact on the jails average daily population. However, local criminal justice policy and practice provide insight into the data, Judicial sentencing practices prosecutorial decison makinm probation violation practices and law enforcement inibatives impact how best to interpret historical data, In that regard interviews were conducted with key local criminal justice participants in order to pan insoht into local criminal justice practice in Cortland County. Law Enforcement Initiatives Between 2002 and 2004 State, County and municipal area law enforcement agencies nstituted a drug enforcement task force that resulted in a high number of arrests within the Cortland County region. These initiatives were reflective of the high number of arrests in 2003. The task force strength was reduced due to lower staffing initiatives of some of the agencies involved in the task force efforts. As a result the drug task force efforts were diminished for a period of time. Staffing issues have rec.ently been resolved and the drug task force efforts are being revitalized. The results of this revitalization should be realized in the near future. At agencies have commtted to the efforts of the drug task force for the foreseeable future. These efforts are anticipated to increase the arrest rates previously realized in 2c03 As these arrests increase, it is anticipated that admissions to the ial will have a corresponding increase. Criminal Courts — The criminal courts are meeting time standards for disposing of criminal cases. New York. State and American Bar Association standards for case completion are met or exceeding in the past year. Efforts to complete case disposition in a more efficient manner have resulted in a fast track Pre Sentence Investigation (PSI) process. This process reduces the amount of time for completion of the PSI. As a result, the length of stay within the county jail has not exceeded acceptable levels. While the length of stay has risen by five days since 1999 (17 days to 22 days), it has been on decline since 2004 when only one judge was sitting (second judge was out for extended period due to illness) The reduction fri the length of stay in the jail is an indication of good case processing of pretriaI cases. Additionally, the number of trials has remained within an acceptable level Indications are that the number of felony cases has increased during the past two year period. Cortland County had the reputation of having the highest number of sex abuse cases per capita in the State. Local jail sentencing practices has remained somewhat of a stagnant protocol. The most common practice is to sentence to the jail in one of two forms: 1) 6 months jail, as part of a flveyear term of probation and 2) :ntermittent sentences (vieekends) This practice has been a rnajcr reason the jail population has ap.•proxims.tely 50% of its average daily population (ACE) as. locally sentenced, Intermittent sentence.s are 1 one j ce1 %r off’ndere p nfj eic 1 eJ ad tne jr tee r a anc or The courts. ‘have’ be.en usln.q a variety of afternative t.o ra cars on proorams as a result of overcrowd[ne. at t.he jail. Electron. monitorin.g is beln.g u.sed a..s an afl.ernative for pretrial defendants, The program is new and has had a roes sure of success• thus far. The Probation Departrr.ent aperates the Aiterna.tives to onu” ora ocaro- s- e ee ehv approximatehj six mo.rdhs with .about 12. cases u.nder the progra.m thus fd.r. A n•ew Drug Court. ha.s• also :.rr ---PAGE BREAK--- CORTLAND COUNTY NEEDS ASSESSMENT FN REPORT TRENDS & PROJECT!ONS been implemented that will hopefully have a positive impact on future jail pouuations but that uses intermittent sentencino as part of the protocol. Dstrict Attorney — Tre current D nas beer in offie br approxima e 1 srs al hougn flC as cee wtO me D. office for more than twenty years. fle D.a. ndcated tna: whe ne nas not implemented any new policies regarding plea bargaining practice, he has taken a more focused view on case reviews within the office that may result in more trials than in the past. He also has been instrumental in resurrecting the Police Drug Task Force. He has a belief that if the jail were not overcrowded there are people in the community today that would be in the jail. He was not in total agreement with the new electronic monitoring program that has been developed for the pretrial population. He did acknowledge that the program has been somewhat successful in the short period since its inception. Probation Office The probation office is charged with operating the Alternative to Incarceration (ATl) Program in Cortland County. The County has been operating a pretrial release program since 1985 as part of the ATI Program. In 2004 the ATI Program was incorporated into the Probation Department. The Pretrial Release program reviews pretrial defendants backgrounds and provides a recommendation to the court as to release pending trial, As a result of this activity, they are very familiar with the pretrial population in the jail. They use an objective assessment tool in determining release recommendations. As mentioned previously, they have begun to use electronic monitoring as an enhancement toward release conditions. The Electronic Monitoring (EM) program has been in existence for about a year and is used sparingly by the courts as a result of jail overcrowding, EM is used as a condition of pretrial release as well as a condition of probation. At the time of the interview there were six people on Electronic Monitoring, 4 on probation and two on pretrial release. There have been a total of twelve people on EM since its inception. The office also supervises those who have been found guilty of an offense and placed on probation by the courts.At the time of the interview there were approximately 520 criminal cases on actve probation ‘k ‘f ri” “r r,rtc , Th dr 0 a uU .a 2 office also feels that there are persons that have been released that would be incarcerated if not for overcrowding, especIally females. ---PAGE BREAK--- TLAN COUNTY N 0 A SSMEN — I ORT JECkN PR0JEOn0NS Seven projection models were developed from the vanous data sources collected Models 1-3 are general population based, Models $ and S were developed as mathematical extrapolations of the jail average daIly population and are used for demonstrative purposes, Models 6 and 7 are statistical based projections with different focus and emphasis, relying on R-Square values for reliability. Two sets of projections were initially developed, one using 10 year historical data and another using five year historical data for average daily population projection models, It was determined that the ten year historical data presented an overall trend that was not representafive of current trends and cractices, After nteraews with key criminal lustice practitioners it was agreed that the f:ve ‘year historical trend was more realistic toward future growth natterrs. Model I Incarceration Rate A projection model based on incarceration rate was created by finding the calculated average of the incarceration rates of the past five years, representing an upswing in ADP, and applying that average to projected future populations. The average incarceration rate was found to be 1.18 inmates per 1.000 persons. The ADP was then projected using the following formula: Ths model predicts an ADP uf 56 in the year 2020 9b ( n I ‘ pu I R 1, I ,ie, Model 2 ADM per IMOO Population The rt ProJectIOn rncdeI pwdicts DM baced on tue ratIo of htoral 4DM tu rstrial popu:atn For I ton al yn r Dl 1 wa dlv d d by poculan uver I oO rng ru rati DM t 100 r on ny ra then yieldi rag of 1 dm1 si 1 000 p in th mg II to rn i ict Because ths proiechbn mves a nredicted ADM vaJue another sten must be taken to nredct the ADP The averse ALOS of the last five ‘years was caiculated to be 24 days 2 DP was then projected usmq the fimi,r.u frrni ---PAGE BREAK--- CORTLAND NEEDS ASSESSMENT PROJECHONS FNAL REPORT M 13 Arets opulton proecti n mod& p d ADM u ng histo rrests hi oncal DM nd his r I popul ti A rano of tmal arrests to popuiat:on was calculated and a ratio of DM to totai arrests was calculated ‘hen mltipIed these two rat:os give a ratio ot ADM to populaton The rao ot Di. to nopJiation is then niulbohe b the proiected population to oat the orediuted ADM for that ‘ear Since we want a projection of ADP, we again use the following formula. where 24 is the calculated average of the historical AbCS: DM 24: This mode! predicts an ADP of 69 in 2025. T- Projected ADP WI 5-lear AESS of 23r fio - - lI’• - C JflJ — - U - e em h Tn mode! orcb future P b using the historical annual pe entage ha ges in D The percent ge Ddng vas a:rjiaed beten a h tastorical ear from 2001 tn 20C5 snce thes ears reprser: an JC5:iPO n ADP and those wuiues r thn a’eraged Th arage annual en entage cnang 4 7 Rd e -ent 0 • — ng a ,ei;a icr fnLon rrnja The annual percentage change can not simply be multiplied by 5, because the percentage change has to be ancUed to each di Rual ea Using fo mula ne 5 year c cemage crae was ‘d to be Tzeeaace aseapc c : a ro eePIasUJrutor2J5 ---PAGE BREAK--- CORTLAND COUNTY NEEDS ASSESSMENT FAL REPORT PROJECflONS L - gFi Model 5 - ADP by Numerical Change A fifth mod& predicts future ADP by using the histoncal annuaf numercai change in ADP. The numerical chanqe was calculated between each historical year from 2001 to 2005. since these years represent an upswing in ADP. and those values were then averaged. This average numerical change. 225 annually, was multiplied by 5 to obtain an expected 5-year numerical change of 11, This value was then added to 2005s historical ADP of 58 to obtain n expected ADP for 2010, and so on, until reaching an expected ADP for 2025. This model predicts an ADP of 103 in 2025. This projection was done using a computer:zcd ARlf (0.1 1 1 nodel (also called Box-Jenkins). This model has an R-square value of 504 and predicts an AD cf 114 2025 Model 7 — Multiplicative Winters Exponential Smoothing (Linear Trend, Multhcativa Seasonabty) 1. ru .r at:. H!t _is. n—i :hu :tn c. le I i nunr i;0 ‘j j h ti1, are aJJoJ toor subtra :ud frur :ne urJ•_ ‘jan non model a 1 Ju f I 12 anl ;‘redicts an a; dO n 2’ Model 6 ARIMA ---PAGE BREAK--- Summary of Results The various models presented here project a range of ADP in 2025 between 42 and 69. in Table 1 below. the credicted ADP of each model from 2010 to 2025 is illustrated. Table 11 Summary Projected ADP I II I Incarceration Rate 57 57 57 56 ADM per 1000 Population 66 66 66 65 Part I Arrests per Population 70 70 70 69 ADP by Percentage Change 73 92 116 146 ADP by Numerical Change 69 81 92 103 ARIMA 74 87 101 114 Multiplicative Winters Exponential Smoothing 65 66 68 69 - These results are illustrated graphically by Figure 1 below. Figure 10 CORTLAND cou NEEDS ASSESSMENT FNAL REPORT PROJECOONS Projected ADP 40 20 -8 ncarceraton Arresis per Pouiaton ADP by Number Change Erponental Smooth Wig Unear, Mu’Api-catve ADM per I ) Poputat.on mbP by Percentage Change .—ARMA ---PAGE BREAK--- CORTLAND COUNTY NEEDS ASSESSMENT ONAL REPORT . PROJECTIONS Peaking and Bed Needs Because the popuiahon of a jaii often exceeds its average daily porulation. t is necessary mat the eaong factor be represented. The average peaking factor in Cortiand Counry over the past 10 yeats has been 2 —di “a , a classi a cc acc s eccr”o”ded Tk 5 ea’c a ecc’ edsd a e a be ab’e to accommodate 22% more than the projected ADP for 2025. Tab:e 3 below shows the bed needs projected by each model for 2012 through 2025. Bed needs are the resuc of appying the peaking and ciassification factors, in this case 22% to tne projected ADP, The highest projected bed need is 179, by the ADP by Percentage Change model, The lowest projected bed need is 69, by the incarceration Rate model. Tabte 12 Summary Projected Bed Needs Incarceration Rate 70 70 69 69 ADM per 1,000 Population 81 80 80 79 Part I Arrests per Population 86 85 85 84 ADP by Percentage Change 89 113 142 179 ADP by Numerical Change 85 98 112 125 ARIMA 90 106 123 139 iplicative Winters Exponential Smoothing 79 84 Recommendation The jail is currently overcrowded, havtng to board out nmates (especially females) and create de nopulatton programs by the courts n order to accommodate current needs, although general p.oouiation growth and cr!me have either been stable or declining. Future growth of the general populahon n Cortland s anticipated to be stable with very little growth or no growth anticipated. The driving force behind the current jail crowding has more to do with criminal justice practice and policy than historical and future growth patterns. A reemphasis on drugs and possible out: of county drug trafficking enforcement appears to be a driving force on future jail bed çrowth needs. it should also be noted that violent crime and arrests have been on the rise since 2001 These factors as well as the courts sentencing practices are having major impacts on the jails population. The first thre . models are general population based, The general population growth rate is anticipated to be stable or declining, Historical data also does not support a strong relationship between general pop•ulrtion and jail population. ---PAGE BREAK--- J J.T4 ¶ swan Models four and five were developed more as demonstrative models than from strong statistical foundation and reflect straight line mathematical extrapolations. The last two models were developed based on statistical projection models. The higher the R value the stronger the predictability rating. The ARIMA model has a .60 R value while the Multiplicative model has an R value of .51. Assuming that criminal justice practice in Cortland County remaki as currently practiced and based on the historical data and system interviews, it is recommended that Cortland County should plan for approximately 140 jail beds by the year 2025. p. claGTXhocaTa a 11122007 24