Compur. Enciron. Urban S.~cms Vol. IO. No. 3i4, pp. 157-163. 1986 0198-9715186 $3.00 + 0.00 Printed m Great Bmam. All nghts reserved Copyright % 1986 Per&wnon Press Lrd ON AUTOMATED CORRECTIONAL DATA SYSTEMS JAMES M. TIEK Engineering Center. Rensselaer Polytechnic Institute, Troy, NY 1218G3590, U.S.A Abstract-The status of automated correctional data systems (ACDSs) in the U.S. is reviewed in this paper. IX can be stated that current ACDSs are indeed data-rather than information-systems: that is. their capabilities are limited. An alternate distributed correctional information system is proposed. AH but a handful of the 50 states in the C’S have implemented or are in the process of implementing some version of a computer-based or automated correctional data system (ACDS) for the storing and retrieval of offender-related data. In fact, many such systems have been upgraded or changed several times since their inception, which, in some cases, dates back to a dozen years or more. The creation of the Law Enforcement Assistance Administration (LEAA) in the Omnibus Crime Control and Safe Street Act of 1968 has significantly accelerated the development and proliferation of ACDSs. An obvious question is whether ACDSs have been effective or, more precisely, cost-effective? A recent study funded by the National Institute of Justice ]I] attempted to answer this question. Although the paucity of available information about ACDSs prevented the study’s authors from providing an explicit answer, they were able to undertake a critical review or assessment of existing ACDSs. This national assessment was based on available ACDS-related reports, structured telephone interviews, and brief visits to 26 ACDS sites. Finally, while this article primarily summarizes the study by Tien et al. [l], it also draws from a recent effort by Tien and McClure [2] in which they consider technical approaches for making computers more effective in public organizations. ACDS STATUS As of 1980, the 50 state correctional agencies had a total of 311 yr of ACDS experience, with an average of 4.2 yr and a median of 5.0 yr. These figures are significant in that the field of automation is one in which experience counts. Frequently the first system installed by a correctional agency was subject to special problems because the agency staff had not yet learned what the computer could do for them. Subsequent systems were often more successful because of the added sophistication of both users and data processing staff. In total, 40 states, Washington, DC, the Federal Bureau of Prisons and numerous regions and localities have ACDSs that are operational. (By operational, it is meant that at least one offender-based application is operating and officially in use.) Although current ACDSs run on a wide variety of different central processing units or mainframes, the IBM 370 and its look-alikes (such as Amdahl or Itel ~uipment) dominate the field, with 28 installations. Six states have, in addition to their large mainframes, minicomputers located in their institutions; in most cases, these are used for peripheral applications such as inmate fund accounting and psychological test scoring. The majority of the ACDSs are written in COBOL; some have parts written in assembler language as well. An important characteristic of any automated data system is the extent to which on-line processing is available. On-line processing refers to the ability to interact with the computer system through a terminaf device such as a teletype or a cathode ray tube (CRT). The functions of data entry, data editing, data retrieval, and data or file updating may be performed either on-line (i.e. via the terminal) or through batch processing. Thirty-five of the currently operating state ACDSs have some degree of on-line capability; that is, at least one of the above four stated functions can be performed via the terminal. Further, 18 of the 29 systems now under development will have some degree of on-line capability. The function of data or file updating deserves further