R e s u l t s
|
R e s u l t s |
|
Raw Data
Following are the mean times and standard deviations to complete
each treatment.
| a | Multi-Column Times |
Drop-Down Box Times |
||||
10 items |
30 items |
50 items |
10 items |
30 items |
50 items |
|
Mean: |
8.5 |
9.9 |
11.9 |
10.9 |
14.0 |
18.0 |
Standard Deviation: |
2.9 |
2.4 |
4.0 |
4.3 |
4.7 |
5.2 |
The solid bars represent the means while the line bars
represent the standard deviations.
2x3 ANOVA
A 2x3 ANOVA was performed with a =.05 . The main effects for list-style
presentation the number of items both indicated significant differences. List style
presentation by number of items interaction has also showed significant differences.
Tests of Between-Subjects Effects
Dependent Variable: TIME
Source |
Type III Sum of Squares |
df |
Mean Square |
F |
Sig. |
Noncent. Parameter |
Observed Power(a) |
Corrected Model |
572.878(b) |
5 |
114.576 |
13.410 |
.000 |
67.049 |
1.000 |
Intercept |
3304.107 |
1 |
3304.107 |
386.709 |
.000 |
386.709 |
1.000 |
STYLE |
270.682 |
1 |
270.682 |
31.680 |
.000 |
31.680 |
1.000 |
NUMITEMS |
203.006 |
2 |
101.503 |
11.880 |
.001 |
23.760 |
.985 |
STYLE * NUMITEMS |
99.191 |
2 |
49.595 |
5.805 |
.011 |
11.609 |
.806 |
Error |
153.795 |
18 |
8.544 |
||||
Total |
4030.780 |
24 |
|||||
Corrected Total |
726.673 |
23 |
|||||
| a Computed using alpha = .05 | |||||||
| b R Squared = .788 (Adjusted R Squared = .730) | |||||||
T-Tests
Our t-test results indicate that for each of the three list sizes (10,
30, 50), the differences between the two styles are statistically significant. That is,
p<.05 for each comparison.
Paired Samples Test (T-Test): Multi-column list vs. drop-down list for three list sizes.
| a | a | a | a | 95% Confidence Interval of the Difference |
a | a | a | |
List Size |
Mean |
Std. Deviation |
Std. Error Mean |
Lower |
Upper |
t |
df |
Sig (2-tailed) |
10 |
2.438 |
3.244 |
.662 |
1.068 |
3.807 |
3.681 |
23 |
.001 |
30 |
4.075 |
3.717 |
.759 |
2.505 |
5.645 |
5.371 |
23 |
.000 |
50 |
6.142 |
3.608 |
.736 |
4.618 |
7.665 |
8.340 |
23 |
.000 |
Error Rates
The chart below shows the average number of errors incurred by the
subjects for both the multi-column and drop-down lists. Note that the mean number of
errors is less than one for each treatment, and that the drop-down list boxes yielded a
greater number of errors for both the 30-item and 50-item lists.
Although the accuracy for the multi-column lists appears considerably better, t-tests indicated that the results were not statistically significant for any of the three list sizes; as a result, our error rate results are inconclusive.
T-tests (Paired, 2-tailed)
List Size |
Sig |
10 |
p < 0.13 |
30 |
p < 0.17 |
50 |
p < 0.21 |
 |
updated 5.9.98 |
 |