4. Discussion

4.1 Interpretation of Results

It seems clear that as the browsing window is made smaller, users waste more time scrolling and searching for the link they are looking for. Although a few users claimed that they were not as comfortable with a large window as with a medium (or even small) window, nearly all had a best time with the large window and worst time with the small window.

4.2 Statistics Explained

To get the time per link, we divided the total time to get from first to last link in each link set and divided by the appropriate number of links. We then used Microsoft Excel to perform average, variance, standard deviation, and ANOVA calculations on the data. We did not perform repeated T-tests because doing so provides inaccurate results because of compounding alphas, as explained here.

4.3 Anomalies

In our second link set (refer to the experiment section of this report), there was a link that required the subject to click on a text link 'Home'. However, one subject associated this printed word with the Home button on the Netscape toolbar. If were were to run this experiment again, we would avoid this ambiguity by choosing a link that was less confusing.

4.4 Subjective Comments

Each subject filled out a subjective satisfaction survey (as refered to in the experiment section of this report). The survey focused primarily on which screen size the user felt most comfortable with and what they felt the ideal screen size for browsing is.

The results for what size the subjects felt most comfortable with were as follow:

Every subject's answer to the window with which they were most comfortable corresponded directly to what their choice of the best browing size window would be with only one exception. That one subject preferred a narrower browser that would take the entire screen vertically but only approximately 2/3 of the screen horizontally. This experiment was not designed to measure the preference of horizontal or vertical scrolling, therefore this type of window size was not used, but it is worth noting.

4.5 Comments

Our hypothesis was originally that the larger the screen size the faster the browsing time would be. Our data and statistics show this to be the case for our subjects. Most subjects preferred larger window sizes as well, noting that it allowed them to avoid scrolling to have to find the information they were looking for. Even those subjects who preferred smaller window sizes still performed better on the larger windows despite their preferences. It is still our opinion that the larger the window size, the faster the web browsing will take place.

It is interesting to note that Yahoo! already tries to limit vertical scrolling by presenting as many links as are readable on each page. Other search engines and web sites also use this approach indicating that our experiment helped to confirm what web designers already seem to know.


Continue
Return to the Title Page for "Screen Size and Web Browsing"