Conclusions and Discussion

Current and Desired Services and Format: Where are the Gaps?

In keeping with Rossi and others' (1999) recommendations for a user-needs assessment, we explored the issue of target audience. According to our key informants, users and potential users of Predictive Services, fire managers at the local, district, regional, State, and national levels should be the primary target audience for products and services. Although the public was listed by about one-fourth of respondents as an audience, serving this target audience presents challenges. There is ample evidence that the publics vary in their degree of knowledge about risk and, in particular, fire (see, for example, Winter and Cvetkovich 2003). Whereas experts might want technically relevant and appropriate information, the publics may want culturally relevant and value-based information (Fischoff and others 1984, Plough and Krimsky 1987). Serving the layperson through the products and services would have different implications than serving the fire management community. For example, executive summaries or annotations attached to data presentations may be of greater interest to lay people. Some of our key informants from the fire management community expressed an interest in this type of summary information.

The overlap between data offered through Predictive Services and other resources (such as National Weather Service) may need to be re-examined by Predictive Services. Overlap was sometimes viewed as a redundancy and, perhaps, a misallocation of resources. The daily fire weather forecasts, red flag warnings, Incident Management Situation reports, drought information, and interagency reports all seemed to be on target as products currently offered that are of high utility to our respondents. To augment these services, the websites might be adjusted to streamline information searches and to have more real-time updates (including removal of information that is no longer accurate or timely). Products that seemed of little interest and might be deleted included regional monsoon updates, upper air soundings, Predictive Services forms, and state-of-the-fuels reports. However, before these are dropped entirely, it might be helpful for the NPSG to explore with some key contacts what the intended purpose and barriers to use are. It may be that the information is of interest and would be useful in a different form. Some products are offered regionally rather than nationally, so consideration of availability is essential to interpreting low levels of use among these nationally distributed respondents.

A number of respondents seemed to have little awareness of the products and services offered and expressed a desire for more information on Predictive Services. Communication aimed at the fire management community to gain increased awareness of Predictive Services seems in order. In addition, respondents expressed a desire for a Web-based orientation or training, and a glossary to assist the user. These comments suggest that respondents who are currently aware of Predictive Services see benefit in imbedding more user support into the products to facilitate utilization of the products as well as to ease comprehension of the products.

Lessons from Acceptability of Risk

Our respondents expressed a clear desire for accurate and timely information. When uncertainty was characteristic of the data, some indicated a desire to have a clear disclosure of the information’s limitations and constraints. When respondents were asked to choose between erring on the side of caution when uncertainty might be involved in reporting, the majority leaned towards a cautious approach. However, we did not present detailed narratives or scenarios with tangible situations in our questions (see, for example, Kneeshaw and others 2004), which may have yielded a more complex picture of risk-reporting preferences. Specifically, if one had asked the respondent to contrast uncertainty in risk estimates for fires involving human life versus those burning in uninhabited areas, the results might have been different.

The Role of Trust and Confidence in Reliance

Most respondents expressed trust and confidence in the information provided by Predictive Services. Few respondents selected trust-related barriers as impediments to utilization. A few comments suggested that a personal working relationship with Predictive Services personnel would be essential to building and maintaining trust. In spite of this expressed trust, respondents tended towards not relying on Predictive Services in decisionmaking. A majority indicated they rely more on other sources than Predictive Services and were somewhat likely to take action based on Predictive Services. The regressions predicting reliance and taking action suggest that trust and confidence is the significant predictor in both cases (with gender, years in position, and level of education contributing to the overall equation, but not as individual significant contributors). This finding affirms the importance of trust and confidence in the delivery of Predictive Services products and services. The inverse relationship between years in current position and reliance and taking action support Reyna’s (2004) finding that more experienced decisionmakers tend to capture the gist of factors leading up to a decision, rather than relying heavily on technical details. Expertise leads to a different way of information processing, most likely reflected here. This also falls in line with the work of Siegrist and others (2000) that shows an inverse relationship between the level of knowledge about a topic and the importance of trust. According to this line of research, those who know more about an issue tend to rely less on trust in making determinations about issues than do those with less knowledge.

Lessons for Risk Communications and Information in Threat Management

Our findings offer some lessons for risk communications and information in threat management. The importance of marketing the products and services to potential users was reflected in a desire among many respondents to know more about what Predictive Services offers. Additionally, risk information could be of greater value if Predictive Services provided tools to facilitate comprehension (e.g., appendices, glossaries, executive summaries). Stating constraints or assumptions used to gather, analyze, and report data could facilitate utilization further. This information would allow the user to better understand and make informed decisions about using the information presented. Although Reyna (2004) found a tendency to use the gist of information in arriving at decisions, she also found that decisionmakers prefer to distill information and arrive at this gist or fuzzy information on their own. Therefore, we do not recommend that a distilled approach be the only method of information presentation. The perception that there is an undesirable overlap between resources available from Predictive Services and other sources was an interesting revelation. Whereas overlap is intentional in this case, the driving force for it was user convenience. It may be that establishing Predictive Services as a unique niche for information would be preferred. Then links to other reliable sources for distinctly different information could be presented. This would eliminate overlap, but point the user to where they could find other information of interest.

Opportunities to Enhance Risk Communication

Respondents indicated that building relationships, dialogue with those providing the data, and assistance with interpretation and underlying assumptions might facilitate their use of Predictive Services. Because trust and confidence influence both users’ reliance on products and services and their likelihood of taking action, more of a direct connection to the fire management community might be desired. This could be facilitated through the Geographic Coordination Centers and might be addressed through a hosted chat link or a hotline that users could call for assistance. Face-to-face briefings might also facilitate familiarity and relationships.

Tools to Move Us Forward

This user-needs assessment highlighted products used by fire personnel and factors that might be facilitators and barriers to Predictive Services usage. It helped clarify informant views on reliance on the products and services and the role of trust in that reliance. Ideas for refinement were offered that might help improve the existing products. These include considering the target audience when developing communication strategies, addressing overlap in available products and services, providing accurate and timely information that discloses assumptions and limitations, and developing Web-based tools that facilitate use. Study results also demonstrated the importance of trust in respondents’ decisions to use Predictive Services information. Fostering relationships with users by involving them in the development and maintenance of products and services might increase user trust and usage of Predictive Services.