Metaphors: Realistic vs. Form-based Phone Interfaces
Results
The first part of our hypothesis was that the task completion times for the RealPhone
interface would exceed task completion times for the form-based interface at a
statistically significant level. Otherwise stated: H1: m (tasktimereal)
> m (tasktimeform).
The alternative hypothesis is that task completion time for the form-based interface would exceed task completion
time for the RealPhone interface. Otherwise stated: H10: m (tasktimereal) <= m (tasktimeform).
The second part of our hypothesis was that the subjective satisfaction ratings for the
RealPhone interface would exceed the subjective satisfaction ratings for the form-based
interface. Otherwise stated: H2: m (subjective
ratingreal) > m
(subjective ratingform). The alternative hypothesis is
that the subjective satisfaction ratings for the form-based interface would exceed the
subjective satisfaction ratings for the RealPhone interface. Otherwise stated:
H20: m (subjective ratingreal)
<= m (subjective ratingform).
In evaluating the data gathered during our survey, we assumed H10 and H20 to be true and attempted to reject each at a significance level of 90%. For n=22, a 90% confidence level for our T-tests would require T > 1.323 to attain statistical significance.
T-Tests for Task Completion Time
Completion times were recorded for each of the twelve tasks. The raw data is presented in Appendix H and Appendix I. The tasks were compared using the Minitab Stat Package to perform T-tests at a 90% confidence level. Eleven of the twelve tasks failed to provide any statistically significant results. However, for task 6, the RealPhone times were significantly greater than the form-based times.
Real Mean |
Real S.D. |
Form Mean |
Form S.D. |
T |
p |
|
Task 1 |
12.3 |
6.85 |
14.8 |
15 |
-0.72 |
0.76 |
Task 2 |
10.37 |
7.79 |
10.18 |
8.32 |
0.08 |
0.47 |
Task 3 |
25.7 |
22.3 |
18.3 |
19.5 |
1.17 |
0.12 |
Task 4 |
8.38 |
4.9 |
9.2 |
7.74 |
-0.42 |
0.66 |
Task 5 |
14 |
13.5 |
10.64 |
4.99 |
1.08 |
0.14 |
Task 6 |
11.53 |
4.84 |
8.94 |
3.63 |
2.01 |
0.026 |
Task 7 |
8.64 |
3.6 |
9.95 |
6.16 |
-0.86 |
0.8 |
Task 8 |
10.1 |
3.38 |
11.6 |
11 |
0.6 |
0.72 |
Task 9 |
22.5 |
8.02 |
21.5 |
10.7 |
0.34 |
0.37 |
Task 10 |
15.5 |
13.2 |
15.8 |
13 |
-0.09 |
0.54 |
Task 11 |
14.75 |
4.02 |
17.7 |
10.2 |
-1.26 |
0.89 |
Task 12 |
9.53 |
6.97 |
8.31 |
4.21 |
0.7 |
0.24 |
T-test results for Mean Task Completion Times. n=22 |
 |
Graph of mean task completion times for the RealPhone and form-based interfaces. n=22 |
T-Tests for Subjective Satisfaction Questionnaire
Subjective satisfaction ratings regarding four different aspects of the interface
design were collected for both tests using our Subjective
Satisfaction Questionnaire. The raw data is presented in Appendix J and Appendix K. The ratings were
compared using T-tests at a 90% confidence level. Three of the four ratings failed
to provide any statistically significant results. However, for question 3, the RealPhone ratings were significantly more favorable
than the form-based ratings.
| Questions: 1. How would you rate the interface overall: Poor (1) -- Excellent (9) 2. The interface was: Confusing (1) -- Excellent (9) 3. The interface was: Unattractive (1) -- Attractive (9) 4. The interface was: Visually distracting (1) -- Visually non-distracting (9) |
Real Mean |
Real S.D. |
Form Mean |
Form S.D. |
T |
p |
|
Question 1 |
6.45 |
1.53 |
6.64 |
1.59 |
-0.39 |
0.65 |
Question 2 |
6.32 |
1.94 |
6.95 |
2.01 |
-1.07 |
0.85 |
Question 3 |
7.32 |
1.86 |
5.5 |
2.15 |
2.99 |
0.0023 |
Question 4 |
7.41 |
2.24 |
7.36 |
1.71 |
0.08 |
0.47 |
T-test results for Subjective Satisfaction Questionnaire. n=22 |
 |
Graph of Subjective Satisfaction Questionnaire Results. n=22 |
 |
Department of Computer Sciences Direct questions and comments to the student editorial team |