Title: Group Value And Intention To Use: A Study Of Multi-Agency Disaster Management Information Systems For Public Safety
APA Citation: Lee, J., Bharosa, N., Yang., J, Janssen, M., & Rao, H.R. (2011). Group value and intention to use- A study of multi-agency disaster management information systems for public safety. Decision Support Systems, 50(2), 404-414
According to the research conducted by Stephenson & Anderson (1997), the use and performance and communication and decision support systems in disaster management have attracted the attention of researches for over two decades. The use of information systems in the early 70s was limited and the range of applications was constrained by the type of equipment available and limitations on access. Mainframe computers were used in disaster management applications in the early 1970s and were accessed solely by defense-related civil emergency planners (Stephenson & Anderson, 1997). In the 1980s, the growth of computer network communications through the use of bulletin board systems provided the vehicle to send and receive e-mail and transfer files over slow modem links.
However, it is only recently that researchers have started to emphasize decision support and information sharing across different disaster management agencies as well as the quality of shared information (Lee, Bharosa, Yang, Janssen, & Rao, 2011). For example, in recent disasters such as the September 11, 2001 terrorist attacks, SARS Outbreak in Asia in 2003, and Hurricane Katrina in 2005, the emphasis to share and use quality information across disaster management information systems has been paramount.
The research conducted by Lee, et.al (2011) is geared toward the improvement of disaster management information systems design for public sector applications. The contribution of the research is twofold. First, the study is conducted based on the understanding of information systems success models. Second, the study comes from the examination of information quality and systems quality attributes expected to play important roles in the disaster management domain.
The research conducted by Lee et.al. (2011) builds on the DeLone and McLean Model of Information Systems Success. DeLone and McLean (1992) developed a model of information systems success to measure the success of an information system at both the individual and organizational level. The categories of information system success include system quality, information quality, use, user satisfaction, individual and organizational impact. These categories provide a framework for measuring information systems success. DeLone and McLean (year?) discovered that use and user satisfaction are closely related and as a result of this interaction between use and user satisfaction, certain “net benefits” will occur (DeLone & McLean, 2003). Net benefits will either be positive or negative, though if the net benefit of an information system is negative, the tendency for use will decrease as well as user satisfaction.
According to Lee et.al (2011), the expected group value and intention to use an information system has come up with several single-item constructs which operationalize the term information systems success. These constructs include: perceived task support, expected group value, information and systems quality, information accessibility, and information timeliness as it pertains to the relevance of information. Relevance of information is information that is applicable and helpful for the task at hand. In a prior study Lee et.al (2005) conducted, they argued emergency responders are overly concerned about information overload and being distracted by information that’s irrelevant to the task they’ve been assigned to. Several studies were conducted and found evidence that information overload occurred in a number of disaster management information systems (Dawes, Cresswell, & Cahan, 2004; Helsloot, 2005).
Research Design & Methodology
The research conducted by Lee et.al. (2011) was set during a simulation emergency flood event which took place between The Netherlands and Germany in which fifty-one people participated. The goal of this simulation was to improve multi agency communications amongst disaster management agencies. The simulation took place over three days where different parts of the exercise were conducted in sequential order starting with strategic, tactical, and operational. The data gathered from one part of the exercise will be provided to other disaster management information system users in the next part of the exercise.
The participants were given a paper based questionnaire which asked questions on the following single item constructs: perceived task support, expected group value, information and systems quality, information accessibility, and information timeliness as it pertains to the relevance of information. The scoring system used was based on a seven point interval scale. Once the surveys were gathered, the scores were analyzed using a statistical method called path analysis which examines casual relationships between constructs. Path analysis is a structural equation modeling (SEM) technique used to address the correlation and strength between variables in information systems research (Dow, Jackson, Wong, & Leitch, 2008). The data analyzed used path analysis because it assumes that all constructs are measured without error (Dow, et.al, 2008) and because of time constraints on the participants, they weren’t able to analyze the data using a more advanced SEM technique.
According to Bharosa, Lee, & Janssen (2010), a similar study was conducted in 2007 where a disaster management scenario took place over three days. The participants in the exercise were unaware of what was going to happen and the situations were neither controlled by the participants or the researchers. Through the use of questionnaires and observations, data was collected and analyzed using an analysis of variance (ANOVA) statistical technique and found that the participants were more inclined to access the information they need from the information system rather than to share it with other participants using the system.
Data Analysis & Results
Because of the time constraints and small number of participants involved in this exercise, only forty six participants submitted their questionnaires. The results of the questionnaire were analyzed using path analysis because of the time constraints and found that if an information system has an expected group value, users will intend to use this system because of the value expected. Unexpected results were found in this research. For example, perceived task support did not show a direct effect on a user’s intention to use the disaster management information system.
In parallel to the quantitative data collection, the research team also collected qualitative data before, during, and after the survey (Lee, et al., 2011). The collection instruments used were document analysis, semi-structured interview techniques with the participants and the disaster management information system architects, and participant observations. The responses from the participants using the disaster management information system found that other participants weren’t sharing the information needed to coordinate with the other participants on the system because they had other stuff to worry about. The “other stuff” here refers to agency and role specific unrelated to the exercise (Lee, et al., 2011).
Recent studies have found evidence of poor information sharing in disaster management information systems (Dawes, et al., 2004) such as the case was during the recovery process after the World Trade Center attacks which resulted in lack of coordination between the various disaster management agencies involved in the disaster management recovery process. As a result, information coming in from the field to the disaster management information system users was not coming in at a timely manner making the participants unaware as to what was happening in the field during the time of the exercise.
Discussions & Future Research
The results of the data analysis revealed that expected group value is a key determinant of intention to use in the public sector disaster management domain while perceived task support has an indirect effect on the intention to use a disaster management information system (Lee et al, 2011). According to Lee et al., the four constructs: perceived task support, expected group value, user satisfaction, and intention to use can be used as indicators of information systems success. However, the results show that the system is only successful on an individual level versus an organizational level which is shown in the correlation between perceived task support and expected group value.
There were limitations in this study mainly with time. The exercise was conducted over a three day period and the participants who were involved were real disaster management personnel who were tasked with other obligations outside of the simulation. Because of the time constraints, the data analyzed using path analysis was the only option that befitted the scenario. As a result of the data analysis technique used, this leaves the reliability and validity of the study open to interpretation and to further research. Had there been additional time, advanced data analysis techniques such as SEM could’ve been used as well as multiple item constructs.
Future research in disaster management information systems may extend this study into different directions. First, the definition of information systems success in multi-agency disaster management is difficult to define. The study conducted by Lee only shows information systems success as one preferred measure rather than one for all situations. Second, to define the major determinant of expected group values is another direction this research could go in and that the type of data analyzed could show that the type of disaster management task would moderate the relationships between the variables. Third, additional studies that would replicate this one would be able to use larger datasets and multiple-item constructs which would test and refine the results discovered in this study.