upa - home page JUS - Journal of usability studies
An international peer-reviewed journal

A Comparison of the Usability of a Laptop, Communicator, and Handheld Computer

Piia Suomalainen, Leena Korpinen, and Rauno Pääkkönen

Journal of Usability Studies, Volume 5, Issue 3, May 2010, pp. 111 - 123

Article Contents


Results

The following sections discuss the results of the subjects’ background information, the subjects’ experiences, the observation of the subjects’ work posture, and the typing and calculating tasks.

Background information of the subjects

Before this study, two of the subjects used laptops and communicators daily. One subject used a handheld computer daily. In addition, 24 (96%) of the subjects had at some point tested the laptop, 12 (48%) subjects had tested the communicator, and eight (32%) had tested the handheld computer.

Subjects’ experiences

All subjects (25) answered five questionnaires. Some of the questionnaires had sections where subjects could answer open questions, but not all subjects answered those open questions.

The subjects answered the question “How ergonomic were the devices (laptop, communicator, and handheld computer) in your opinion?” The scale was from 1 to 5. For the laptop, they answered as follows: 4 (16%) very poorly, 3 (12%) pretty poorly, 10 (40%) fairly, 8 (32%) pretty well, and 0 very well. For the communicator, they answered as follows: 8 (32%) very poorly, 12 (48%) pretty poorly, 3 (12%) fairly, 0 pretty well, and 2 (8%) very well. For the handheld computer, they answered as follows: 1 (4%) very poorly, 4 (16%) pretty poorly, 13 (52%) fairly, 5 (20%) pretty well, and 2 (8%) very well.

Figure 5 shows the data about question two: “Did you feel any stress in any part of your body when working with the devices?” Again, the scale was from 1 to 5, with 1 = very little stress, 2 = pretty little, 3 = pretty much, 4 = fairly much, and 5 = very much stress. For the laptop, subjects felt the most stress on their necks (average 3.0), then their backs (average 2.9), and then their shoulders (average 2.8). For the communicator, subjects felt the most stress on their backs (average 3.5), then their fingers (average 3.3), and then their eyes (average 3.1). For the handheld computer, subjects felt the most stress in their eyes (average 2.7), then their wrists (average 2.6), and then their backs (average 2.6).

Figure 5

Figure 5. The distributions and averages for the question: “Did you feel any stress in any part of your body when working with the devices?”

One person did not answer the question (Figure 5) and two others did not experience any stress using any of the devices. For the rest of the subjects, their answers indicated that when working with the laptop they felt the highest amount stress in their neck (fairly much), the second highest was in their back, and the third highest was in their shoulders. They felt the lowest amount of stress in their feet and the second lowest in their fingers.

Two of the subjects did not experience any stress on any body part when working with the communicator. For subjects that felt stress with this device, the highest stress was felt in their back (pretty much), the second highest stress was felt in their fingers, and the third highest stress was felt in their eyes. Subjects felt the lowest amount of stress in their feet and the second lowest amount of stress in their arms.

Four subjects did not experience any stress on any part of their body while working with the handheld computer. For subjects that felt stress with this device, the highest stress was felt in their eyes (fairly much) the second highest was felt in their backs, and then their wrists.

Over half of the subjects (56%) thought that writing with the laptop was pretty easy, and 36% of the subjects thought that writing with the communicator and with the handheld computer was pretty difficult. Almost half of the subjects (48%) thought that it was pretty difficult to calculate using the laptop. They thought calculations using the communicator were fairly easy (36%) and calculations using the handheld computer were very easy (60%).

Observation of the subjects’ work posture

The following sections describe the observations of the subjects’ work posture while using the laptop, the communicator, and the handheld computer.

Laptop

The observations of the observer (researcher) indicated that the majority of the subjects (68%) worked in such a position that their neck was at a 15 to 45 degree angle. In that position there was some degree of stress. Eleven subjects also had some degree of rotation in their neck. Thirteen subjects (52%) had only a little stress in their back.

The angle of the arm was pretty good. Only three subjects had very much stress in their arms. Eight subjects had some degree of stressing positions. The researcher made similar observations while subjects worked on the calculation portion of the tests.

Communicator

Fourteen subjects (56%) had very much stress in their back and nine (36%) had some degree of stress when working with the communicator. The positions of their arms were only a little stressful in almost all subjects (92%). The position of their backs was somewhat stressful for nine subjects (36%) and very stressing for eight subjects (32%). The researcher made similar observations while subjects worked on the calculation portion of the tests.

Handheld computer

Seventeen subjects (68%) had very much stress in their neck and six subjects (24%) had some degree of stress when working with the handheld computer. Ten of the subjects (40%) had very much stress in their arm when they worked with the handheld computer. All other subjects had only a little stress in their arms. Eleven subjects (44%) had very much stress in their back and nine subjects (36%) had somewhat stress. Subjects did the calculations in the same position as they typed.

Typing and calculating tasks

The subjects (25) wrote altogether 19,734 letters and made 45 typing errors (2.3 errors/1,000 letters) with the laptop. Everybody completed the typing task. The subjects made altogether 164 calculations with the laptop. Ten subjects completed all eight calculations within the given time. One person completed only two. The highest number of calculation errors/subjects was five.

With the communicator, the subjects wrote altogether 9,153 letters and made 62 errors (7.3 errors/1,000 letters). The subjects correctly completed 188 calculations (7.5/subject) with the communicator.

With the handheld computer, the subjects wrote 2,668 letters and made 31 errors (14.1 errors/1,000 letters). The subjects completed 186 (7.4/subject) calculations with the handheld computer.

The results for different devices were analyzed using nonparametric tests, Friedman’s 2-way ANOVA by ranks (3 samples, Table 1). The null hypotheses were rejected based on the following factors: (a) the distributions of typing characters for the laptop, the communicator, and the handheld computer were the same (Sig =0.000); (b) the distributions of typing errors for the laptop, the communicator, and the handheld computer were the same (Sig =0.004); and (c) the distributions of correctly calculated exercises for the laptop, the communicator, and the handheld computer were the same (Sig =0.011). The null hypothesis for the distributions of calculation errors for the laptop, the communicator, and the handheld computer were the same (Sig =0.097) so it was not rejected. Table 1 shows the pairwise comparisons for typing characters, typing errors, and correctly calculated exercises, which we could analyze based on the testing of null hypotheses.

Table 1. The results of pairwise comparisons for typing characters, typing errors, and correctly calculated exercises of different devices using Friedman’s 2-way ANOVA (Adj.Sig = Asymptotic significances, Std.Test.Statistic = standard test statistics)

Table 1

* p<0.05

Table 1 shows that the differences between typing characters in all pairs were significant (p< 0.05). Asymptotic significances (2-sided tests) were displayed. The significance level was 0.05. Differences between typing errors were significant in pair communicator-handheld computer. For the calculation test, the differences of correctly calculated exercises were not significant.

Previous | Next