تحقیقات از تلاش دقت تجارت کردن و تاثیر خودکارآمدی در رفتارهای جستجوی وب

One of the major barriers to the electronic commerce is that Web users may waste a lot of time and effort to search for information. Previous research suggests that both goal and experience are important factors influencing user's Web search behavior. In the aspect of goal motivation, the effort–accuracy trade-off model provides a good explanation for the choice of a strategy. Yet, this trade-off model does not consider the impact of the experience. The Social Cognitive Theory (SCT) may make up for this weakness by means of the construct called self-efficacy, which reflects the effect of experience. In this study, we explore if there indeed exists a trade-off effect between effort and accuracy in Web searching behaviors and how self-efficacy may impact the trade-off effect. The results show that while the emphasis of accuracy will increase the level of effort by low self-efficacy subjects, the trade-off between effort-saving and accuracy-seeking does not exist for individuals of higher self-efficacy.

The rapid growth of World Wide Web (WWW) has significantly increased the quantity of data available to WWW users but at the same time reduced the accessibility of information [14]. Information explosion causes significant cognitive overload. The nonlinearity of the hyperlinked documents may further cause disorientation of people who lose their sense of location and direction. WWW users may waste a lot of time and effort in struggling with inconsistent information that is distributed on Web. Therefore, how one may find information effectively has become an important issue to WWW users. Research must be directed at understanding users' Web search behavior to provide guidance for the design of WWW applications. Several previous studies have responded to this concern. Hoffman and Novak [10] suggest two kinds of browsing behaviors: goal-directed and experiential. In this view, people who adopt a goal-directed strategy look for specific information, while the experiential users mainly attempt to make sense through their acts and the resulting system's responses. Similarly, Murphy [17] suggests that searching and surfing are two major ways to browse the Web sites. Searching refers to specific information finding activity while surfing means that people simply browse Web pages for curiosity or fun. In another study, Vandenbosch and Higgins [23] find that scan- and focus-search are two ways to acquire information through Internet. In scan-search, people simply browse with no specific question to resolve, while focus-search occurs when people want specific information for specific problem. Collectively, these studies suggest that goal is very important. People who have an explicit goal will adopt a directed, focused strategy to search. To be specific, in the aspect of goal motivation, the effort–accuracy trade-off model [12] and [20] provides a good explanation for the choice of a strategy. According to this model, people, as rational and adaptive decision-makers, have a reservoir of strategies from which they can choose to realize their goal. Furthermore, they will attempt to minimize the cost and/or maximize the return when selecting search strategies. Thus, people will choose a more sophisticated, and therefore, more effort-demanding strategy when their goal is to seek for an accurate result. Conversely, a less effort-consuming strategy will be employed when they do not care much for a specific outcome. Several laboratory experiments have confirmed this prediction [20]. However, many other studies have shown that in naturalistic environment, people often stick to the same strategy regardless of situational differences [13]. For example, in researching computer usage, Olson and Nilsen [19] find that expert users do not change their access method. Similarly, Hammond et al. [9] show that people rely on a rapid, low analytical control strategy to solve their problem if they are experienced and if there are a large number of cues. Finally, Beach and Mitchell [6] and [16] propose the image theory that stresses the intuitive and automatic aspects of decision-making in real-world settings. In their view, people are creatures of the past experience, which produce a strong image that they can use to test the acceptability and compatibility of an alternative when making decisions. An option is rejected when the weighted violation of the criteria exceeds some critical threshold. Furthermore, making judgment about the compatibility of an option with one's image is a rapid, smooth process that can be characterized as intuition. An analytical process is evoked only if there are more than one acceptable alternative. These findings collectively suggest that people intuitively adopt one well-known alternative to solve problems when they become experienced with this particular problem domain. Thus, to effectively apply the effort–accuracy trade-off model to explain real-world search behaviors, an additional factor, experience, must be considered. We should note that goal motivation and experience are not mutually exclusive. Rather, according to the Social Cognitive Theory [3], they are reciprocally determined. In this theory, experienced people who are also confident in their capability may adopt a goal level that is higher than one adopted by novices. Conversely, people of insufficient experience and low self-conviction may wonder and give up easily when facing trivial challenges. More importantly, experts may rely on intuitive strategies over analytical ones and can still manage to maintain their level of performance without expanding more effort. The Social Cognitive Theory therefore provides an appropriate perspective to complement the effort–accuracy trade-off model. Thus, in this study, we set out to reexamine the effort–accuracy trade-off theory by considering the impact of experience. Two questions are asked in this study. First, “is there an effort–accuracy trade-off on WWW search decision?” and, second, “how does one's experience influence the effort–accuracy trade-off on WWW search decision?” Note that according to the Social Cognitive Theory, people's success or failure in their past experience can be indexed by a psychological construct called self-efficacy. Thus, the second question is rephrased as “how does one's perceived self-efficacy influence the effort–accuracy trade-off on WWW search decision?” A three-trial experiment was conducted to address these issues. This paper is organized as follows. Section 2 reviews the theoretical basis from which four hypotheses are developed. 3 and 4 describe the research design and data analysis, respectively. Finally, we discuss the effect of self-efficacy on the effort–accuracy trade-off model in Section 5 and provide a conclusion in Section 6.

In summary, this study investigates if there is a trade-off effect between effort-saving and accuracy-seeking as well as the impact of experience, represented by self-efficacy, on this trade-off for Web searching behaviors. The result shows that self-efficacy is an important factor to Web search accuracy. On average, people with low self-efficacy will vary their levels of effort to seek for different accuracy goals, while people with high self-efficacy do not significantly alter their level of effort (i.e., Web search time) but still can find accurate result in their search more quickly than those of low efficacy. The reason appears to be because high self-efficacy individuals may hold a strong mental model of usefulness in the Web sites that they use. Conversely, low self-efficacy people apparently may get lost easily because they lack a similar strategy and are still trying to identify one. This study has several limitations. First, our sample population was predominantly young males with little computer background, and therefore, our result cannot be generalized to all populations. Nevertheless, in theory, the effect of self-efficacy on the choice of strategy (e.g., high self-efficacy people tend to use intuitive strategies and low self-efficacy ones use analytical strategies) has been shown to exist in general [5]. We believe that this issue deserves more research for further clarification. Next, we have not followed the procedure of typical laboratory experiments that investigate the trade-off between effort-saving and accuracy. In these experiments, subjects are taught a specific set of strategies that differ from one another in the required level of effort and the expected level of accuracy. The shortcoming of such arrangement is that the experimental tasks will depart substantially from those of the real world. Because our interest is in finding out if the trade-off effect exists in real-world environments, we have decided to allow subjects to choose their preferred strategies in our experiment. Third, the 20 min in each trial may not be enough for some nonexperiential users to find the result although it is far more than the average time spent. Fourth, this research relies on the use of time as the measure of effort. While this use is reasonable in the tradition of effort–accuracy research, it may be inadequate for assessing the mental effort of high self-efficacy WWW users. Indeed, the issues concerning expert intuition and how mental effort of intuition can be measured are difficult and beyond the scope of the present research. Finally, we only announce the time and price of the top five subjects, so that the others should justify their performance subjectively by comparing their answers with the announced ones. This indirect feedback may make these users misjudge their performance and affect the self-efficacy in the next trial. The contributions of this study are several. To research, we examine how self-efficacy may affect the trade-off between effort-saving and accuracy-seeking in Web search behavior. The results suggest that this kind of trade-off may not exist for the high self-efficacy group. For researchers who are to base their work on the trade-off model proposed by Payne et al. [20], it is important to consider the effect of self-efficacy. To practice, we find that experienced users who have a high level of self-efficacy will not change their level of effort in different situations. This kind of users may have developed a strong image of helpfulness for particular sites, which is both powerful and dangerous. People relying in this intuition can be quick and accurate, and yet, when they are wrong, they often become the worst performers. The Web site designer should be aware of this tendency of Web users and provide ways for recovery. Many opportunities for future research exist. One is that studies can be conducted to evaluate different designs of screen presentation. According to Vessey's [24] Cognitive Fit Theory, a fit between the search strategy and the presentation may enhance people's performance. Researchers can also compare the different effects of self-efficacy between direct and indirect feedback. Some past studies have suggested that feedback is an important factor influencing the level of self-efficacy. Thus, it is interesting to study the impact of the format, content, and speed of feedback toward self-efficacy. In addition, while this study roughly classifies the strategies used in Web search, future work may consider a microlevel analysis of user's mental model of strategies used. This will help to build a cognitive model of Web search behavior and aids in our understanding concerning how people form their image toward particular sites. A third potential area of research is error detection and recovery in Web search behaviors. Finally, research effort can be directed at studying how socialization may affect the image of Web sites. All of these can lead to useful guidelines for designing friendly and helpful Web sites.