Conclusions

---

The results of this research confirm that providing more powerful computing platforms to individuals who are engaged in traditional office-oriented tasks typical of those encountered by lower level organization workers can result in improved ratings of the system. Our results also confirm that for the right task/application combination providing a more powerful computing platform allows users to become more productive. However, this increase in productivity only occurred when the more powerful platform was associated with a significant decrease in navigation time.

Navigation is critical as users make modifications such as those studied here. While navigation is not the primary objective, it is an unavoidable secondary task. Interestingly, only 20% to 26% of the task completion time was actually spent making the required modifications, while 59% to 78% was spent navigating the desired locations. Navigation was also responsible for the majority of the difference in task completion times observed when comparing the low- and high-demand tasks. While the navigation required to complete these tasks was identical, 56% to 75% of the increase in task completion times were due to increases in the time spent navigating.

These results provide some insights that could prove useful when tasks can be custom designed or users can be trained, but more importantly they highlight directions for additional research. Interestingly, both the design-oriented insights and new directions for research focus on navigation. Many navigation strategies are supported by the applications we utilized and our participants were free to employ any strategy they felt was appropriate. Given the influence navigation has on total task completion times, users should be encouraged to employ the most efficient techniques possible. This could affect courses where users learn to use new applications, with additional time being spent teaching the most efficient navigation techniques. Perhaps more importantly, users should be given guidance as to which techniques should be used and when they should be used. When tasks can be designed, they should be presented in such a way as to help users select and implement effective navigation techniques. For instance, some efficient methods for navigation require knowledge of surrounding text or nearby landmarks (e.g., headers). Providing this guidance appears to be more important as modifications become bigger.

Each modification can be viewed as a navigation event followed by a modification. While the navigation required for the low- and high-demand tasks is the same, navigation took significantly longer for the high-demand tasks. In other words, the size of the required modification has a significant effect on the time required to navigate to the desired location. Helping users separate navigation from the actual modification may prove useful in reducing these effects. One alternative would be to provide users with navigation-oriented information, such as appropriate landmarks or nearby text, in addition to modification-oriented information. Specific techniques need to be explored and conditions under which they prove beneficial must be identified.

Only those users completing the Excel-based high-demand experienced an increase in productivity when using more a powerful computing platform. Perhaps more importantly, this increase in productivity corresponds to the only condition where navigation became more efficient when using a more powerful platform. This supports our view that efficient navigation is a critical factor in determining whether or not a more powerful computing platform will result in improved productivity. Additional research that investigates the differences between navigation within Excel, Word, and PowerPoint may provide insights into which strategies become more efficient when used on more powerful platforms.

While deleting or inserting characters was the primary goal, navigation emerged as the dominant influence on task completion times. Increasing productivity requires that users understand the navigational techniques that are available and make appropriate choices. It may be possible to assist in this decision making process by presenting tasks appropriately, but additional research is necessary before specific recommendations can be made. Research that investigates the efficiency of alternative navigation techniques may allow applications to be redesigned to more effectively to exploit the additional capabilities of more powerful computing platforms. Finally, it is important to recognize that this study focused on clerical activities representative of those that lower level organization workers may encounter. Additional research that explores different tasks may prove useful in understanding the relationship between hardware performance and productivity for managers working in decision-making environments.

---