Abstract:
Integrated Library Management Systems (ILMS) are essential components in libraries that support the processing, storage, management, and retrieval of data. Before the advent of ICT in libraries, Egerton University's main library data was manually created and managed in card format; arranged in a way that facilitated access. Librarians made many errors in the development and management of the metadata especially when filing records whose words/terms were wrongly written or abbreviated. They were also overwhelmed by the mass of conventional paper records in trays, some with data illegibility. The need to automate library operations and subscribe to an ILMS such as AMLIB software that could reduce human labour and errors was critical. However, the AMLIB software has not been able to eliminate human errors entirely, but rather the errors changed their formats from paper to machine-generated. The errors continue to elicit conflicts between library staff and the library users, occasioning the necessity for this study, which aimed at exploring the types of human errors that occur while using the AMLIB software, explaining the sources of errors, and examining the staff skills and training needs in working with AMLIB software at Egerton University library. The study adopted the Unsafe Acts Model by Desai, which is an information system theory that deals with errors in software, to illustrate the interaction between the independent and dependent variables. The case study undertaken describes the current state of AMLIB software in Egerton University library, by use of quantitative and qualitative data. Purposive and simple random sampling was used to sample the target population that comprised staff and students of Egerton University. Questionnaires and interview schedules were used to collect data from the participants. The Statistical Package for Social Science (SPSS) program version 21 was used to analyse data that is presented in descriptive statistics, tables and graphs. The study encountered major human errors in omissions, errors of commission, and errors of record redundancies. The errors occur mainly because of insufficient staff training, complex system processes and procedures, staff fatigue, inadequate supervision, and defective security checks at the library exit points, among others. The study recommends regular staff training, adequate system control, close staff supervision, and sufficient surveillance at all library exit points as ways to minimize human errors in the use of the AMLIB software.
