RESEARCH INTO INFORMATION QUALITY: A STUDY OF THE STATE-OF-THE ART IN IQ AND ITS CONSOLIDATION (Academic Paper) Luís Francisco Ramos Lima Universidade Federal do Rio Grande do Sul, Brazil
[email protected]
Antonio Carlos Gastaud Maçada Universidade Federal do Rio Grande do Sul, Brazil
[email protected]
Lilia Maria Vargas Universidade Federal do Rio Grande do Sul, Brazil
[email protected] Abstract: Information Quality (IQ), though not a recent study field, faces a problem in organizing its theoretical body. Furthermore, lack of a defined, robust field of knowledge for research in IQ is detected. The purpose of this article is to analyze the state-of-the-art in the scientific production within the area, focusing on thematic and methodological aspects. One-hundred and seventy one articles were catalogued within five major events. In order to organize the themes collected in IQ, the structure supplied by Conceptual Maps for knowledge organization was used; as a result of this approach, three central views for research into IQ are proposed: Organizational, Behavioral, and Operational.
Key Words: Information Quality; Conceptual Maps
1 INTRODUCTION The theme Information Quality (IQ) has been studied with great attention since the last decade and of particular note are the guiding concepts contained in the research developed by Wang [20] and Strong et al. [17, 18]. Since then, there have been several attempts to solve problems related to IQ in academic and organizational fields, as there is a pressing need for a methodology that measures to which level organizations can develop information products and services of quality for their users [12]. In recent years, the impacts of Quality of Information in organizations have gained notoriety, seen in the creation of events destined especially towards the sharing of knowledge regarding IQ, bringing together academic studies and ex cathedra work, like the International Conference on Information Quality (ICIQ), held at MIT (Massachusetts Institute of Technology), in its ninth edition in 2004, and the more recently the International Workshop on Information Quality in Information Systems (IQIS), coordinated by the ACM (Association of Computing Machinery), in its second edition in 2005. This situation, however, does not mean that academic production in IQ has achieved greater consolidation as a branch of knowledge; among the challenges recently pointed out for the development of IQ, Pierce [15] highlights that though research into IQ is not a novelty, “what is lacking is a unified body of knowledge that addresses data quality in its entirety”. The objective of this study is to carry out an analysis of the scientific production in the area of Information Quality in an effort towards the organization and consolidation of research on the theme. Keeping this in mind, the main events relative to the subject of Information Quality as well as those related to the area of Information Systems with related articles were visited. For this task, articles were
catalogued and analyzed according to the classifications proposed by Alavi and Carlson [2] and Hoppen and Meireles [10] for the bibliographical research of scientific articles. A presentation is made of the main subjects discussed in IQ, illustrated with the elaboration of a Conceptual Map, according to the model proposed by Novak [14], in an effort to establish the relationships among these topics. The article is structured as follows: section two presents a brief explanation of the research bibliography on Information Systems and its importance for IQ. The following section outlines the Research Methodology adopted in this study. Section four presents the Results of the Study, including the Conceptual Maps proposed for the organization of the State-of-the-Art in IQ. Lastly, the Final Remarks concerning the study are presented.
2 BIBLIOGRAPHICAL RESEARCH IN IS AND RELATIONSHIP WITH IQ Much has been studied regarding theoretical research and its main bases for the area of Information Systems [13]. Especially in the last decade, there have been several attempts at consolidation in the area, in an effort to give robustness and epistemological representativeness to the IS theory, in the methodological aspect as well as in the organization and dissemination of knowledge. In the last decade, a methodology for the evaluation of research articles was proposed by Hoppen et al. [9], which sought “more precise and available criteria, so as to encompass all the quality dimensions of a study and avoid that some criteria are forgotten or neglected”. This classification permitted IS area to attain greater facility and productivity in the evaluation of the research performed in the area. From a wider perspective, Bertero et al. [4] offer a critical analysis of the scientific production in Management Science, where IS is inserted as a theoretical field. In examining the applicability, the references used, the intrinsic quality of the material and the epistemological criteria, the authors reveal a need for “greater quality and significance in our field”. Benbasat and Zmud [3] diagnosed that in the 1980s, to ensure legitimacy, the use of concepts borrowed from other disciplines of reference was adopted, which generated a crisis in the identity of the IS area; as a solution, the authors proposed a set of properties (conceptual and phenomenological) that define the IS field. In response to this work, Agarwal and Lucas [1] reaffirmed that the crisis is real, though they criticized the restrictive nature of the approach adopted that is limited to a “micro-focus” of research in IS, and that “macro studies should convince our colleagues in other academic disciplines of our belief that IT is a key business driver in the 21st century”. Lunardi et al. [13] analyzed the Brazilian production in the IS area in detail, covering the main events and journals in this specific area and in Management. They found that the main characteristic in the IS area as a field of knowledge is the existing interaction between the IT and its application in organizations, mixing the ‘micro’ and ‘macro’ perspectives discussed above. All the above studies have the purpose of highlighting the importance of the scientific discussion in relation to the academic production, for the consolidation of the necessary characteristics for a field of knowledge. The same effort must be done in relation to the Information Quality, a field with an ever more robust production, and which may suffer the same ill identified by Benbasat and Zmud [3], which may lead to prevision of a ‘crisis of identity’ in the IQ field. In 2001, Chung et al. [5] attempted to define the frontiers of knowledge and furnish a structure for the field through the application of an exploratory survey, raising various aspects in regard to the characteristics (skills) of Data Quality. These characteristics were classified in accordance with the General Theory of Systems [5] in their three classes: mechanical, open and human systems. A review of the lines of research in IQ was carried out [11], in which the researchers pointed out that progress has taken place in IQ research in the last 20 years and the range of the research themes in the ICIQ (International Conference on Information Quality), major event in the area, has increased. They also presented a list with the main research interests of the participants, as presented in Table 1.
Aggregated Research Interests of Participants Assessment, Benchmarks, Metrics, Score cards, Optimization, Perceptions Methodologies, Frameworks, Tools, Paradigms
Types of systems: financial, cooperative, accounting, health care (HIPPA), governmental, data warehouses, multimedia, scientific
Identified at ICIQ Standards, Metrics, Tools, Benchmarking & Achievable goals IQ Paradigms, Jargon/terminology, standardization, Methodologies, frameworks & Visualization Functional Areas Strategic vs. Operational Governmental (voting), data warehouses, mailing lists
Application of data mining & Statistical Techniques to improve IQ Application of TDQM in different Cross industries Industries Application: Six sigma and lean applied to IQ problems Case Studies: Successes & Failures Best & Worse Practices Longitudinal vs. Snapshot Communication of DQ inside of Visualization Organization Cost of IQ Cost/Benefit Analysis IQ ROI Cost Justification Cross country, cross culture DQ Cross disciplinary Focus issues Cultural Issues & IQ ROI Ecommerce: Web quality, XML Web quality Exchange of IQ information Exchange of IQ information, Use of IQ information IQ considerations in systems and DB DQ of Software Design Design IQ education (self-contained courses, IQ education degree programs) IQ officer & specialists: role, Career Paths responsibilities, skills needed, certification Knowledge management, Business Knowledge management Intelligence Managing IQ IQ Governance Metadata Metadata National disasters National disasters/”Big Splash”, Legislation: Sarbanes Oakley Homeland security, legal & Public Policy Process Management Risk analysis Risk analysis Tags for IQ: their use & impact Table 1 – Research Topics
Year as a Conference Sessions 1998, 1999, 2000, 2001, 2002, 2003 1997, 1999, 2000, 2001, 2002, 2003
1997, 1998, 2000, 2001, 2002, 2003
1997, 2000, 2001
2002, 2003 2001
2003 1999, 2000, 2001, 2002
1997, 2001, 2002 2001, 2002, 2003
2000, 2001, 2003 2001, 2002
2001, 2002, 2003
3 RESEARCH METHODS Sample Definition In this study, 171 articles covering the period from 1995 to 2005 were reviewed. These articles were collected, depending on the availability obtained from online databases, such as The ACM Digital Library (Internet address: http://portal.acm.org), EBSCO Host (Internet address: http://search.epnet.com) and Proquest (ABI/INFORM) (Internet address: http://proquest.umi.com). Brazilian events were studied directly from the proceedings published on CD. To facilitate the search and organization the articles were divided into two groups: (1) articles originating from events considered of greater relevance in the area of Information Quality; (2) articles referring to the theme in other important Information Systems events. In the latter group: the following key words were used in the search: “quality of information”; “quality of data”; “information” or (relational indicator between the research terms) “quality”; “data” or “quality” and their analogs in Portuguese. Following the search, the texts were checked in order to confirm that the subject was related to the research theme. Articles from brazilian and international events were sought, when it was found that there was a scarcity of local production. All the events are well classified by the “Qualis” system from CAPES (Commission for the Improvement of People/Staff in Higher Education – a Brazilian government educational department), depending on the respective areas of evaluation. A proviso should be made with regard the ICIQ event, which permits the submission of texts from executive articles and presentations. A total of 22 executive presentations were catalogued from the 168 pieces of work viewed, and will not be included in the analysis. Finally, due to the variation in the quality of the papers, open analyses by event will not be made. Group
Initials ICIQ
Name International Conference on Information Quality
IQIS
International Workshop on Information Quality in Information Systems International Conference on Information Systems Congresso Anual de Tecnologia da Informação Encontro Nacional da ANPAD
1
ICIS 2
CATI ENANPAD
Editions Used 1st (1996)*, 4th (1999)*, 5th (2000)*, 6th (2001), 7th (2002)*, 8th (2003), 9th (2004)* 1st (2004)*, 2nd (2005)*
20
20th (1999), 21st (2000)
4
(2)
-
19th (1995)
Quantity 146
1 Total 171
(1) Events included in the research in which no article on the theme was found * Complete proceedings Table 2 – Events included in the research
Referential for the Analysis of Methodologies The greater amount of information was obtained from the proceedings of the events: titles, abstracts and key words. To ensure greater validity to the study, the full texts of the articles were also reviewed, in order to better identify the methodology and epistemological position adopted. The papers were classified as either practical or theoretical, according to the methodological strategy. The theoretical
studies were classified according to the proposition of Alavi and Carlson [2] for non-empirical research in their work on the development of the discipline of Information Systems; for the practical studies, the approach developed by Hoppen and Meireles [10] on the classification of scientific research was used. In the articles reviewing, it was decided to include an additional category to the practical studies Simulation – in order to fill a gap in the research, not taken up by the classifications adopted, notably influenced by the field of Computer Science, which denominates as ‘experiments’ or ‘simulations’ implementation studies in software for conceptual tests. Table 3 presents the definitions of each classification. Theoretical
Conceptual Illustrative
Practical
Applied Conceptual Qualitative
Survey Research Experimental Study Simulation
Describes frameworks, models or theories offering explanations and conceptual reasons. Directed towards practice; frequently contain recommendations for actions or steps to be followed in certain circumstances. Combine characteristics of both the Conceptual and Illustrative. Includes the techniques of Case Study, Participative Observation and Action Research, among other methods within the same interpretative position. Research work expressly indicated as survey or with the use of questionnaires with quantitative descriptions as a result. Considering the experiments performed with groups of people. ‘Experiments’ or ‘simulations’ as implementation of software for conceptual tests. Adapted from [2, 10]
Table 3 – Classification adopted
Conceptual Map Following the analysis of the results in accordance with the proposal above, a structure was elaborated to organize the knowledge produced on IQ. Thus, a Conceptual Map (CM) was chosen for the task. The Conceptual Map, a tool for the organization and representation of knowledge, was developed and proposed by Novak [14] and extended to diverse applications. Conceptual Maps can, for example, “be used to clarify or describe the ideas that are held by people on a certain subject” [8], and for this purpose, they are suitable for the present study towards the objective proposed by the authors of using the CM as a support tool for the bibliographical review. For the elaboration of the CM, concepts and ideas are necessary for the consequent association. To this end, the key words were collected from the articles reviewed according to the following criteria: • For each article, up to three key words were selected; • For articles with more than three key words, three were chosen according to the criteria of the evaluator; • For articles with fewer than three key words, there was no alteration • Finally, in those articles in which there were no key words, up to three key words were proposed in the evaluation according to the evaluator criteria. The process described above proffered the survey of relevant concepts for the IQ area, which were later represented graphically. Conceptual Maps are a subjective way to present and group concepts the relations between the concepts and ideas were proposed by the authors of the present study, except where the given article expressly contains such an association.
4 RESULTS OF THE STUDY Authors, Places and Countries A set of 281 different authors from 25 different countries and 158 institutions, among them academic and companies, had participated in the elaboration of the articles included in the study. The place with the greatest amount of production was the University of St. Gallen, in Switzerland, contributing with 8 articles in the events studied. An important fact: the 10 main academic research centers (which correspond to 6.3% of the institutions) are responsible for 39 articles, which represent 22.8% of the IQ production. Table 4 shows the top ten academic production centers. The company that most participated towards research production was Acxiom Co. (Internet address: www.acxiom.com), with 4 articles. Academic Center University of St. Gallen Hulboldt-Universitat zu Berlin Indiana University of Pennsylvania MIT (Massachussetts Institute of Technology) Politecnico di Milano Suffolk University Boston University Northeastern University University of Rome Worcester Polytechnic Institute
Country Switzerland Germany United States United States Italy United States United States United States Italy United States
Quantity 8 6 5 5 5 5 4 4 4 4
Table 4 – 10 Main Academic Research Centers
Analysis on the participation of countries in the studies shows the dominant presence of the United States of America in the production of IQ. Institutions from the USA participated in the production of 104 articles (60.8% of the total). Table 5 shows the 5 countries with the greatest volume of production in IQ. Only one Brazilian article was found and included in the study. Country United States Germany Switzerland United Kingdom Italy
Quantity 104 13 11 10 9
Table 5 – Countries with the Greatest Amount of Production in IQ
Research Strategies Within the sample, an equilibrated distribution was found between the research strategies adopted: the use of Theoretical and Practical approaches appeared in almost equal numbers. From the different kinds of research used, of particular note is the presentation of conceptual articles (23%), which is a further indicator of the efforts made at consolidation of the research in the area. The illustrative studies, which represents 19% of the articles, point to a parallel line of academic production that attempts to make the concepts so far developed available to professional practice, with ‘recipes’ for the use of the methodologies in companies and reports of the best practices in IQ. A third type that stands out is the Simulation, adopted for the establishment of concepts related to Computer Science; within this type, the articles are, for the most part, concerned with technological aspects such as databases, network and internet applications that use simulations to test theory. This type represents 20% of the articles analyzed. Table 6 presents the total numbers for each research type analyzed, and Figure 1 shows the relative distribution of the strategies adopted, and the chosen types for each strategy.
Strategy
Research Type Conceptual Theoretical Applied Conceptual Illustrative Qualitative Experimental Study Practical Survey Research Simulation Total
Quantity 40 20 33 29 0 15 34 171
% 23% 12% 19% 17% 0% 9% 20% 100%
Table 6 – Research Strategies
Illustrative 35%
Qualitative 37% Simulation 44% Conceptual 43%
Theoretical 54%
Practical 46%
Applied Conceptual 22%
Survey Research 19%
Figure 1 – Distribution of Strategies and Research Types
Methodology Utilization In the practical studies, the presence of a robust methodology was noted, due to ensure the validity of the research and the capacity for generalization. In the 78 studies identified as practical, the use of methodology varied in accordance with the type of study. The qualitative studies presented the greatest lack of methodology references; only 17% of the articles of this type expressly indicated the utilization of some methodology for the research development. The situation is inverted in relation to quantitative studies; 80% of the survey studies were based on methodological aspects, and 88% of the simulation studies. Table 7 presents the values found. Research Type
Qualitative Survey Simulation
Methodology? Quantity No 24 Yes 5 No 3 Yes 12 No 4 Yes 30 Total 78
% Use 17% 80% 88%
Table 7 – Use of Methodology
Conceptual Maps Chung et al. [5] recommend that research efforts must be conducted in the area of interpretative qualities (“ability to identify and define the organizational implications of IQ”). In order to elaborate the Conceptual Maps, it was sought to follow such skills, extended towards two challenges: (1) elaborate a main research structure in IQ, that would guide the associations performed, and (2) classify the terms studied within the proposed structure.
A secure path that may indicate the research tendencies in IQ is the observation of the development of work sessions at events over the years. The ICIQ event (in its ninth edition in 2004) well illustrates the evolution of the research topics in the area, as presented in Table 8. Year Session Practice Paper Session (I, II, III) 1996 Research Paper Session (I, II, III) State of the Art Practice in IQ Data Mining and Data Warehouses Data Quality and the Web Data Quality in Context 1999 Data Quality Management Emerging Research and Practice Models of Data Quality Assessing Information Quality Case Studies on Information Quality Data Warehousing and Data Mining Developing Information Product Maps Electronic Commerce and Networking 2000 Improving Information Quality IQ and Decision Support Information Quality Management Measurement of Information Quality Record Matching and Data Source Selection
Type Year Session P Along the Road to Data Quality P Building Systems for Quality P Data Quality in Business Functions P Data Quality Issues on the Web O Data Quality Measurement 2002 P Databases and Data Warehouse Cleansing O, P IQ Evaluation and Assessment IQ in Organizational Contexts P Politics of Data Quality P The Application and Extension of Integration PM O, P Conceptualizing Information Quality
Type Year Session P Business Applications P Cost Analysis O Cost based Cases P Data Mining P Data Quality Applications 2004 Data Quality Cases P P Healthcare Applications O IQ Research Frontier P IQ Strategy P Quality Metrics O, P Web/Internet Quality
P P O P O O, P P P
P P O, P P P O, P P O
IQ - Information Quality PM - Product Maps
Data Mining and Record Matching Data Quality Applied Impact and Effects of Data Quality 2003 Information Quality Evaluation and Assessment Information Quality Issues on the Web Managing Information Quality Methods, Techniques and Tools Organizational Issues in Data Quality
Type O O O P P P O O,B O P O, P
TYPES: O - Organizational B - Behavioral P - Processual
Table 8 – Research Sessions of ICIQ
This identification of the sessions allows suggesting a division of knowledge on IQ. Noting the themes of the sessions permitted the classification into three important categories, denominated IQ Views. In Table 8, the proposed classifications for each session are indicated, according to the following concepts: • Organizational View: refers to the impact of IQ in the organization, its management, influence on the structure and production, on strategic and tactical approaches; • Behavioral View: analyses the human aspect in IQ, in the performance of daily tasks, the insertion of IQ in professional skills and capacities and above all in the view of the internal and external clients of the organization. As seen in Table 8, only one topic contains articles that analyze directly this View; • Operational View: examines the technical and methodological aspects of IQ, such as measures, the application development, tools for data control and information systems; denotes an operational approach. In this way, the Main IQ Conceptual Map was elaborated, which will serve as a base for developing the respective maps for each view, incorporating concepts obtained from catalogued key words. Figure 2 shows the Main Map. From reading the map it can be deduced that IQ is determined by the set of three views, which are related to each other. From the articles reviewed, approximately half of them – 84 articles – contained key words in their contents. Eighty-seven (87) had no key words, and in these cases the key words were inferred according to the content. A total of 279 key words were identified, classified and standardized (e.g. the terms Data Warehouse and Data Warehousing were standardized according to the rate of occurrence), which resulted in a list of 269 key words for the development of the Conceptual Maps. The complete list of words is in Annex 1 of this paper. In the maps related to the IQ View, four types of concepts were incorporated. The father-concept identifies the corresponding view; the concepts in italics denote necessary inclusions made by the evaluator, due to a better comparison between the concepts. The concepts in bold identify the concepts suggested by more than two occurrences of the corresponding key word, and the non-highlighted concepts represent one occurrence of the key word. Figure 3 shows the key adopted in the following maps.
Figure 3 – Key to the Conceptual Map
Figure 2 – Main Conceptual Map
The Organizational View was composed by strategic and tactical elements that involve the direct determination of high management, or matters related to the organizational structure of the company. Some of the included concepts are: better practices in IQ, the relations of IQ through the organizational strategy, and benchmarks developed in the studied articles. Figure 4 shows the Organizational View related to IQ. The Behavioral View sought to incorporate aspects related to the agents involved in the Information Quality, such as consumers of internal or external information to the organization and the detailing of skills and resources of collaborators in relation to IQ. Aspects such as the personal needs of the subjects and professional training specifications are also inserted, as shown in Figure 5.
Figure 4 – Organizational View
In turn, in order to compose the Operational View, terms related to the implementation of quality processes, processes of IQ, methodology and measures, operational tools for data and information quality and problems related to IQ were selected. Figure 6 shows the Conceptual Map with reference to the Operational View of IQ.
Figure 5 – Behavioral View
Figure 6 – Operational View
5 CONCLUSIONS The analysis of the articles on IQ published in the studied events made it possible to present a panorama of the research carried out on the subject. It can be seen that the field of IQ has been undergoing a process of evolution in the production of knowledge on the subject, but with an imminent need for theoretical consolidation. The identification of the main subjects made it possible to identify the concepts studied within IQ, and facilitated the elaboration of a structure that organizes and presents the knowledge disseminated as a result of these studies. As observed, there is no apparent preference for the production of theoretical or practical research. However, the examination of the articles dedicated to the practical position indicates a greater care taken with the methodological rigor in the Case Studies. There is also a large volume of production with the use of Simulation, resulted from researches dedicated to the technological aspects of information, such as the structure of data banks and of IS. The existence of a large volume of theoretical production, using the ‘applied conceptual’ and ‘illustrative’ research types, denotes the concern with the application of research in the IQ field in the professional environment. This characteristic should be preserved, and serve as a reference for Brazilian research, as corroborated by Bertero et al [4].: “national production is therefore a phenomenon of academic environment, generated and consumed within itself”. Unfortunately, of the 171 articles studied, only one is produced in Brazil, which suggests there is little knowledge of or little interest in this subject, which is of proven importance and significant impact in organizations; to point an example, note Redman’s study [16], where it is estimated that the absence of a quality program results in costs, due to the low quality of information, of 20% of the profit of organizations. In order to organize the knowledge on IQ, the structure offered by Conceptual Maps were used. Thus, the division of this knowledge into three views was proposed: Organizational, Behavioral and Operational, in order to organize the concepts identified from the articles. Far from representing boundary limits, the maps should be developed, reviewed and enriched. A further area of possible research, based on the present material, is the inclusion of new concepts obtained from articles on IQ in national and international journals. According to Chung et al.[5], research into Information Quality is highly interdisciplinary. Instead of this representing an obstacle, it should be considered as a challenge to studies in the area of Management, because of its relevance; after all, in the last decade it was already noted that “poor Data Quality [and also Information Quality] can have a severe impact on the overall effectiveness of an organization” [19]. The evolution of IS, of the forms of work in organizations and even of the environment in which we live, have added several elements of Complexity: it is difficult to measure and manage the information, especially when there are problems in knowing the information we work with [6]. Recently, the same author suggested that knowledge and the decision support criteria used in these systems should be restructured, and added to high quality information in order to facilitate decision making [7], which indicates a continuing concern with the Information Management and its Quality.
6 REFERENCES [1] AGARWAL, Ritu; LUCAS Jr., Henry C. The Information Systems Identity Crisis: Focusing on High-visibility and High-impact Research. MIS Quarterly. September 2005, v. 29, n. 3, p. 381-398. [2]ALAVI, M.; CARLSON, P. A Review of MIS Research and Disciplinary Development. Journal of Management Information Systems. 1992, v. 8, n. 4, p..... [3] BENBASAT, Izak; ZMUD, Robert W. The Identity Crisis Within the IS Discipline: Defining and Communicating the Discipline’s Core Properties. MIS Quarterly. June 2003, v. 27, n. 2, p. 183-194. [4] BERTERO, C. O.; CALDAS, M. P.; WOOD Jr., T. Produção científica em Administração de Empresas: Provocações, Insinuações e Contribuições para um Debate Local. Revista de Administração Contemporânea. Jan/Abr 1999, v. 3, n. 1, p. 147-178.
[5] CHUNG, WooYoung; FISCHER, Craig; WANG, Richard. Redefining the Scope and Focus of Information Quality Work. In: Information Quality. New York: M.E.Sharpe, 2005, p. 230-248. [6] DAVENPORT, Thomas H. Ecologia da Informação. São Paulo: Futura, 1998. [7] DAVENPORT, Thomas H.; HARRIS, Jeanne G. Automated Decision Making Comes of Age. MIT Sloan Management Review. Summer 2005, v. 46, n. 4, p. 83-89. [8] GAVA, T.B.S.; MENEZES, C.S.; CURY, D. Aplicações de Mapas Conceituais na Educação como Ferramenta MetaCognitiva. Available: www.edu.ufrgs.br/trilha/mapas_conceituais/paginas/bibliografia.htm. Viewed in 27/09/2005. [9] HOPPEN, Norberto; LAPOINTE, Liette; MOREAU, Eliane. Um Guia para Avaliação de Artigos de Pesquisa em Sistemas de Informação. Revista Eletrônica de Administração. UFRGS, 1996, ed. 3, v. 2, n. 2. [10] HOPPEN, Norberto; MEIRELLES, Fernando. Sistemas de Informação: um Panorama da Pesquisa Científica Entre 1990 e 2003. Revista de Administração de Empresas. 2005, v. 45, n. 1, p. 24-35. [11] KAHN, B.K.; PIERCE, E.M. --[12] KAHN, B.K.; STRONG, D.M.; WANG, R.Y. Information Quality benchmarks: Product and Service performance. Communications of the ACM. Apr 2002, v. 45, n. 4, p. 184-192. [13] LUNARDI, G.; RIOS, L.R.; MAÇADA, A.C.G. Pesquisa em Sistemas de Informação: uma análise a partir dos artigos publicados no EnANPAD e nas principais revistas nacionais de Administração. XXIX EnANPAD Proceedings. 2005. [14] NOVAK, J.D. A Theory of Education. New York: Cornell University Press, 1977. [15] PIERCE, Elizabeth M. Introduction. In: Information Quality. New York: M.E.Sharpe, 2005, p. 3-20. [16] REDMAN, Thomas. On The Cost of Poor Data Quality. Available: http://www.dataqualitysolutions.com. 2003. Viewed in 06/12/2005. [17] STRONG, Diane M.; LEE, Yang W.; WANG, Richard Y. Data Quality in Context. Communications of the ACM. May 1997, v. 40, n. 5, p. 103-110. (a) [18] STRONG, Diane M.; LEE, Yang W.; WANG, Richard Y. 10 Potholes in the Road to Information Quality. Computer. Los Alamitos: IEEE Computer Society Press, August 1997, v. 30, n. 8, p. 38-46. (b) [19] WAND, Yair; WANG, Richard Y. Anchoring Data Quality Dimensions in ontological foundations. Communications of the ACM. November 1996, v. 39, n. 11, p. 86-95. [20] WANG, Richard Y. KON, Henry B. MADNICK, Stuart E. Data Quality requirements analysis and modeling. In: 9th International Conference on Data Engineering. April, 1993.
ANNEX 1 Key Words List Key Words Data Quality IQ Dimensions TDQM Information Quality Data Mining Data Cleansing Data Integration IQ Improvement Data Matching Data Warehousing Information Product Knowledge Management IP-MAP IQ Measurement DQ Assessment DQ Framework DQ Strategy DQ Tools Biological Data Sources Clustering Cooperative IS Databases Decision Making DQ Dimensions DQ Improvement DQ Management E-commerce Error Detection Extraction Firm Performance HealthCare Inconsistent Data Information Searching IQ Assessment IQ Framework IQ Survey IS Quality IT Investment Markov Chain Maturity Model Quality Levels Quality Metrics Record Matching Regular Expression Relational Algebra Web Data Web Sites Website Design Website Quality
# 29 13 13 12 11 7 7 7 5 5 5 5 4 4 3 3 3 3 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2
Key Words (one occurency) Account Plan Accounting IQ Accounting IS Active Data Warehouse Adaptive Query Processing Address Management Administrative Health Data AIMQ Methodology Algebra Analytic Hierarchy Process Approximate record matching Assessment Authenticity Authoritative Data Source Banking Data Management Business Intelligence Business Rules Certification Classification Cluster Analysis Communication Quality Communities of Practice Comparative Study Completeness Complex Sales Composition of Quality Dimensions Conflicting Data Conjoint Analysis Consumer Behaviors Content Management Corporate Governance Corporate Householding Cost Benefit Analysis Cost Measurement Critical Success Factores CRM Customer Data Integration Customer Data Management Customer Investigation Customer Loyalty Customer Segmentation Cyberspace Data and Information Quality Data Errors Data Evolution Life Cycles Data Extraction Data Integrity Data Management Data Mapping
Data Metrics Data Migration Data Product Map Data Production Data Provisioning Data Security Data Sources Data Standardization Database Evaluation Database procedures Decision Quality Deduplication Dependence between errors Dirty Data Distributed Systems DQ Analyst DQ Assurance DQ Benefits DQ Checklist DQ Control DQ Costs DQ Effects DQ Engineering DQ Feedback DQ Inference DQ Patterns DQ Problems DQ Program DQ Scorecard DSS Dual-Process Theory Duplicate Detection Enterprise Information Systems ETL Evaluation of IMP Event-driven Process Chains Evolving DSS Factors of DQ Feed Management Feed-back control systems Financial IS Financial Reform Fishbone Diagrams Freshness Fuzzy Search Fuzzy Theory General Systems Theory GQM Heterogeneous Networks
Information Accuracy Information Attributes Information Consumer Information Continuity Information Defensibility Information Engineering Information Errors Information Flow Model Information Governance Information Infrastructure Information Integration Information Integrity Attributes Information Management Information Management Capabilities Information Orientation Information Overload Information Sources Information Supply Chain Information Systems Information Validity Assessment Instrument Development Intelligent Agents Internet Applications Internet Regulation Interorganizational Networks IP Process IP Quality IQ Attributes IQ Criteria IQ Education IQ Management IQ Measurement Methodology IQ Metadata IQ Methodologies IQ Methodology IQ Metrics IQ Research IQ Specifications IQ Technologies IQ Tools IS Curriculum IS Evaluation IS Management Issues in DW Knowledge Engineering Large DB Logical Interdependencies Machine Learning Management of Data Resource
Maslow Hierarchy Measurement Tools Measures Measures of Interestingness Metadata IQ Missing Data Modeling Validation Molecular Biology Multi Attribute Utility Theory Multi-channel architectures Multidimensional Modeling Multi-source IS Name and Address Quality Name quality assessment Negotiation Strategy Network Intrusion Detection Object-oriented DB Ontologies for Knowledge Representation Organizational DB Organizational Process Partition Similarity Perturbation Poor information Preservation Private Linkage Probabilistic Decision Model Probability Calculus Procedures for Data Quality Management Process Map Project Evaluation QML Quality Assessment Quality Certification Quality evaluation Quality Function Deployment Quality in Modeling Query Real Options Rental Classifieds Results Root-cause analysis Safety Telephone Systems Sales IQ Sarbanes-Oxley Sensivity Analysis Service Network Similarity Simulation Small Business
Sociology of Knowledge Source Quality Structural Equation Modeling Supply Chain Survey System Quality Taxonomy Temporal information Theory-Specific DQ Time Series Time Series Modelling TQM UML Unstructured Text Use Case Model User Perception User Requirements User satisfaction Value of Information Virtual DB Virtual Knowledge Communities Website Quality Assessment WWW Impact
LEGEND AIMQ - AIM Quality CRM - Customer Relationship Management DB - Data Base DQ - Data Quality DSS - Decision Support Systems ETL - Extraction, Transfering and Loading GQM - Goal-question Metric IMP - Information Map Product IP - Information Product IS - Information Systems IT - Information Technology QML - Quality of Service Modeling Language TDQM - Total DQ Management TQM - Total Quality Management UML - Unified Modeling Language
