Data Analysis and Results

To examine the effects of table of contents (TOC) and the use of frame on information searching and browsing, three different types of data were collected: scores on users' performance and speed in searching and browsing for information, along with the satisfaction questionnaire results.

The score on users' performance was calculated by counting the correct answers each user retrieved from an experimental system. The experimenter awarded one point for an unambiguously correct answer, a half-point for a partly answered response and no point for a wrong answer or abandoned question. Exceptionally, user's speed was calculated by counting unabandoned questions.

Results

The results are generalized according to the dependent measures: users' performance, users' speed and users' satisfaction.

Users' Performance

The users' performance in information searching and browsing with traditional TOC was statistically significant more accurate than with expandable TOC (F (1,32) = 5.42, p < 0.05). This finding did not support Hypothesis 1 that predicted users' performance with expandable TOC would be more accurate than with traditional TOC.

The mean score of the single-frame users was 11.208 and that of the multiple-frame users was 10. 875. However, the difference was not significant. This finding did not support Hypothesis 4 that predicted users' performance with multiple-frame would be more accurate than with single-frame.

However, there was a significance of the effects of interaction between table of contents and frames (F (1,32) = 4.32, p < 0.05). Users under expandable table of contents performed better under the single-frame treatment than the multiple-frame treatment. In contrast, users under traditional table of contents performed better under the multiple-frame treatment than the single-frame treatment.

Users' Speed

The users' speed in information searching and browsing with traditional TOC was statistically significantly higher than with expandable TOC (F (1,32) = 7.28, p < 0.05). This finding did not support Hypothesis 2 that predicted t users under expandable TOC treatment would have higher speed than those under traditional TOC treatment.

The mean speed of users under single-frame treatment was 8.00, which was exactly the same as that of users under multiple-frame treatment. This finding failed to support Hypothesis 5 that predicted that users' speed under multiple-frame treatment would have higher speed than those under single-frame treatment.

The effect of interaction between table of contents and frame was statistically significant (F (1, 32) = 7.28, p < 0.05). Users under traditional table of contents treatment possessed higher speed with multiple-frame treatment than with single-frame treatment. In contrast, users under expandable table of contents treatment possessed less speed with multiple-frame treatment than with single-frame treatment.

Users' Satisfaction

To test users' satisfaction, two dependent variables were employed: perceived degree of disorientation and perceived ease of use.

Perceived degree of disorientation. The mean scores on users' perceived degree of disorientation under the interaction between expandable table of contents and single-frame, traditional table of contents and single-frame, expandable table of contents and multiple-frame and traditional table of contents and multiple-frame treatments were between 2.28 and 3.37. Since seven category Likert scales were used in the satisfaction questionnaire, it must be reasonable to conclude that users did not feel disoriented while navigating through the experimental systems.

Further analysis was made to determine the significant difference in users' perceived degree of disorientation. The results of this analysis are summarized in Table 1.

Table 1: Source Information for 2¬2 ANOVA of Mean Scores on Users' Perceived Degree of Description

Source	DF	Sum of Squares	Mean Square	F	P
TOC	1	8.3457	8.3457	7.56	0.0097
Frame	1	0.1372	0.1372	0.12	0.7268
TOC ¬Frame	1	2.5364	2.5364	2.30	0.1395
Residual	32	35.3388	1.1043

The results of the ANOVA in Table 1 indicate a significant difference in perceived degree of disorientation between the two groups of users in expandable TOC and traditional TOC (F (1, 32) = 7.56, p < 0.01). The mean score of expandable TOC users was 3.574 and that of traditional TOC users was 2.611. Thus, the users under traditional TOC treatment reported feeling significantly less disoriented than those under expandable TOC treatment. This finding did not support Hypothesis 3 that predicted navigation would be more satisfied using expandable TOC as opposed to traditional TOC. Users must be more satisfied if they feel less disoriented.

Table 1 indicates that the main effect of frame failed to reach significance (F (1, 32) = 0.12, p = 0.7268). This finding did not support Hypothesis 6 that predicted the navigation would be more satisfied with multiple-frame as opposed to single-frame. However, users under multiple-frame treatment reported feeling less disoriented than those users under single-frame treatment, because the means score of perceived degree of disorientation of users under multiple-frame treatment was 3.031, whereas that of users under single-frame treatment was 3.154.

The finding in Table 1 indicated as well that the interaction between table of contents and frame failed to reach significance (F (1, 32) = 2.30, p = 0.1395).

Perceived ease of use. The mean scores for users' perceived ease of use under the interaction between expandable table of contents and single-frame, traditional table of contents and single-frame, expandable table of contents and multiple-frame, and traditional table of contents and multiple-frame are 4.21, 5.03, 4.46, and 5.56 respectively. The score higher than 4.00 suggests that for users the hypermedia system was perceived as easy to use.

Further 2¬2 ANOVA was employed to test the significant differences in users' perceived ease of use. The between-subjects were once again table of contents and frame (see Table 2).

Table 2: Source Information for 2¬2 ANOVA of Mean Scores on Users' Perceived Ease of Use

Source	DF	Sum of Squares	Mean Square	F	P
TOC	1	8.2656	8.2656	7.04	0.0123
Frame	1	1.3611	1.3611	1.16	0.2898
TOC¬Frame	1	0.1736	0.1736	0.15	0.7032
Residual	32	37.5903	1.1747

The results in Table 2 indicate a significant difference between perceived ease of use means for the two levels of table of contents (F (1, 32) = 7.04, p < 0.05). The mean score of users' perceived ease of use under traditional table of contents treatment is 5.292, whereas the under expandable table of contents treatment is 4.333. In other words, users perceived the traditional table of contents treatment easier to use than expandable table of contents treatment. This finding did not support Hypothesis 3 that predicted that navigation would be more satisfied using expandable TOC as opposed to traditional TOC. The users who perceived more ease using navigation tools should be more satisfied with the navigation tools than those who perceived less ease using them.

The results of ANOVA in Table 2 indicate no significant difference was found between the two types of frame (F (1, 32) = 1.16, p = 0.2898). This finding did not support Hypothesis 6 that predicted the navigation would be more satisfying using multiple-frame as opposed to single-frame. However, users perceived multiple-frame (mean = 5.007) as easier to use than single-frame (mean = 4.618), which failed to differ significantly.

Similar to the previous finding, Table 2 revealed that no interactions between table of contents and frame were detected (F (1, 32) = 0.15, p = 0.7032). A comparison of means scores of all groups shows that for users traditional table of contents and multiple-frame was perceived as the easiest navigation tools to use. On the other hand, expandable table of contents and single-frame was perceived as the most difficult navigation tools to use.

The results of data analysis indicated several statistically significant effects. However, no hypothesis was supported by the experimental data.