<p>The purpose of this Master’s thesis is to study how well the online catalogue of The Music Library of Sweden works when ordinary users are searching for sheet music. The issues studied are: Why do the users fail to find the sheet music they are looking for? Do the users take advantage of the possibilities that the classification codes of the SAB system offer? If not, why?</p><p>The source material of the study is 144 comments from a user questionnaire about the online catalogue and, in addition, the search log for one day (513 queries). The comments were sorted into seven groups depending on the type of information need, for example searches for certain instrumentation or genres and problems with spelling. Each analysis ends with suggestions for improvements of the catalogue and its instructions.</p><p>The thesis show, according to the Zipf’s principle of least effort, that sheet music searching in the online catalogue is demanding too much effort and knowledge of the users. Searching with SAB classification codes is too complicated which may be the explanation why only 2 percent of the queries in the log are code searches. In 51 of the 76 cases where users failed to find sheet music the queries would have succeeded if the patrons had used classification codes. The SAB system is thus not used as often as it deserves.</p><p>The variety and amount of failed queries in the material show that sheet music is difficult to search, for several reasons, for example the complex construction of uniform titles of musical works and that the searchable vocabulary of the online catalogue often differ from the users’ own.</p>

This thesis has been motivated by past research, problems and realizations that online library catalog users frequently perform subject searches – using keywords, subject headings and descriptors – and these searches have yielded unsatisfactory results. Web-based catalogs or WebPACs (Web-based Online Public Access Catalogs), belonging to the so-called third generation of online catalogs and providing a wide variety of search options, remain largely underutilized despite the continuous advancement of information retrieval systems. Users still encounter a number of problems, such as those related to translating their concepts to the language of the catalog’s system and cross-references prepared to this purpose. Subject access in online library catalogs can be provided through different access points. To that purpose natural and controlled indexing and retrieval languages are used, and each among them has its advantages and downsides. Natural language indexing is performed by the computer, in which process words from defined fields are automatically extracted. Controlled indexing languages are those in which selection of terms to be assigned to documents is manually performed. These are, for example, classification systems, subject heading languages and thesauri. During the 1970s, a consensus was reached that the best retrieval results are gained when using both types of indexing languages simultaneously. Apart from indexing languages, it is necessary to take into account user search behavior; and while designing user interface one has to allow for the users’ skills and knowledge - ensuring instruction, help and feedback information at every step of the retrieval process. The aim of the research was to determine the variety and quality of subject access to information in WebPACs of British university libraries, including searching by words or classification marks, natural and controlled languages, browsing options, and forming simple and complex queries in order to conclude about existing advancements, offered models and employed methods and compare them to WebPACs of Croatian university libraries.

The purpose of this Master’s thesis is to study how well the online catalogue of The Music Library of Sweden works when ordinary users are searching for sheet music. The issues studied are: Why do the users fail to find the sheet music they are looking for? Do the users take advantage of the possibilities that the classification codes of the SAB system offer? If not, why? The source material of the study is 144 comments from a user questionnaire about the online catalogue and, in addition, the search log for one day (513 queries). The comments were sorted into seven groups depending on the type of information need, for example searches for certain instrumentation or genres and problems with spelling. Each analysis ends with suggestions for improvements of the catalogue and its instructions. The thesis show, according to the Zipf’s principle of least effort, that sheet music searching in the online catalogue is demanding too much effort and knowledge of the users. Searching with SAB classification codes is too complicated which may be the explanation why only 2 percent of the queries in the log are code searches. In 51 of the 76 cases where users failed to find sheet music the queries would have succeeded if the patrons had used classification codes. The SAB system is thus not used as often as it deserves. The variety and amount of failed queries in the material show that sheet music is difficult to search, for several reasons, for example the complex construction of uniform titles of musical works and that the searchable vocabulary of the online catalogue often differ from the users’ own.

The transformation of a manual paper-based moderation process into an electronic moderation (eModeration) process poses unique challenges. These challenges concern academic processes, people and the user experience of interactive systems. eModeration can improve the user experience of assessment processes while lowering the risk of delaying the process or losing scripts. Despite the benefits associated with optimising assessment procedures, particularly examination procedures, the use of eModeration in South Africa is limited. There are several possible reasons for a lack of eModeration adoption ranging from infrastructure and technical issues through to organisational and human factors. The focus of this study is on the human factors involved in eModeration. Since no User Experience Evaluation Framework for eModeration existed at the time of this research, an in-depth study was conducted based on the experiences of eModeration users in the context of private higher education institutions. The study focused on identifying the most important user experience constructs for the evaluation of an eModerate system within the context of private higher education institutions in South Africa towards proposing a framework. The study was based in the fields of Information Systems and Human-Computer Interaction with eModeration being the application domain. The research used a Design Science Research methodology, which involved the development and testing of a User Experience Evaluation Framework for eModeration. The data generation methods included interviews with deans, eModerators and management, as well as a survey that included responses from both moderators and deans. The research was conducted at Midrand Graduate Institute and evaluated at Monash University. The study makes a validated contribution towards identifying the most important user experience constructs. The identified constructs were utilised in the design and development of the User Experience Evaluation Framework for eModeration, which can be used along with the evaluation criteria tool to evaluate eModerate systems. / Information Science / D. Litt. et Phil. (Information Systems)

The objective of this study was to provide information on the characteristics of a specific group of international college students from a developing country in order to assist system managers in the selection of OPACs.