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

How Low Can You Go? Is the System Usability Scale Range Restricted?

Philip Kortum and Claudia Ziegler Acemyan

Journal of Usability Studies, Volume 9, Issue 1, November 2013, pp. 14 - 24

Article Contents


Methods

The following sections discuss the participants, materials, and procedure used in this study.

Participants

The population consisted of 32 undergraduate students from Rice University with an age range of 18 to 22. Gender was unspecified. For their participation, participants were given credit in partial fulfillment of a class requirement.

Materials

The System Usability Scale was used to capture participants’ subjective usability assessments of voting ballots. The modified System Usability Scale described by Bangor, Kortum, and Miller (2008) was used, and the word “ballot” was substituted for the word “system.” This change in wording is an accepted practice in System Usability Scale administration and has been shown not to have any detrimental effects on the reliability or validity of the measures (Sauro, 2011). The fully modified SUS used in this study is shown in Figure 1.

A set of 14 different paper ballots were administered to participants in the study and are shown in Figures 2 and 3. The ballots reflect a wide range of possible design choices, and were roughly modeled on examples from the real world, or were composites of different ballot designs. Each ballot has between four and six possible choices, using a question on which a typical participant could easily form an opinion. This design has two benefits. First, it allows the participant to focus on the voting process and the ballot—as opposed to any extraneous, evocative properties of that ballot. For example, the ballots do not have the typical name/party affiliation format and use non-controversial ballot choices, because using partisan races, real political figures, and controversial ballot selections has the potential to evoke emotional responses that might impact a participant’s assessment of the underlying usability (Ladd & Lenz, 2008). To verify that the ballots had no important evocative features, we conducted a post test that correlated the valence score for each ballot with its usability (SUS) score. This correlation (r2=.010) was not significant, F(1,12)=0.12, MSE=503.87, p=.738. Second, by using issues that a voter can actually form an opinion about (as opposed to slate voting or nonsense ballot choices), these ballot designs provide participants with the opportunity to cognitively engage in making a ballot selection.

Procedure

After signing an IRB-approved consent form, participants were given a packet that contained the 14 ballots, which were presented in random order. After completing each ballot, the participant was asked to complete a System Usability Scale. This procedure continued until the participant had voted and rated all 14 ballots. Upon completion of the formal rating process, participants were asked to identify the least and most usable ballots.

 

Previous | Next