Technical Translation: Usability Strategies for Translating Technical Documentation

Criteria 
From the relatively long list of criteria used to assess usability, it emerged 
that many were unsuitable for use in this study, either because relevant 
events did not occur with enough frequency to justify their inclusion or 
because they were impossible to quantify given the procedures and equip-
phrased
which
completing
of other 
phenomena which were observed during the test. The pilot study also made
it clear that additional work was needed to establish exactly how the criteria
should be applied in the main study.
When analysing the error criteria data, it was apparent that the method 
for handling data from the Post-Test Survey was not appropriate (see page 
234
the mouse while others immediately began re-reading the user guide. It 
day, the administrator may give more or less information to the par
Scripting the instructions given by the administrator, coupled with 
rule out the possibility of the administrator making 
ad hoc 
com
which might vary from session to session and consequently biasing or 
phrased 
slightly to make it easier to quantify them. An example of a re
criterion is “Errors at point where subject thinks task completed” 
was rephrased as “Number of times subject stops work withou t
a task”. It was also necessary to add additional criteria as a result 
the amount variability which may inadvertently come as a result of unscrip-
ted instructions from the test administrator. For example, o
n any particular
ment in use. As a result, some criteria were deleted while others were re
strictly enforced procedure for dealing with questions (see page 226) 

---

Experiment to Test the Impact of Iconic Linkage
the other error criteria. Apart from faulty logic, subtracting the number of 
correct answers from the total errors meant that any attempts to represent 
the data in the form of a graph produced seriously skewed results.
In addition, it was felt that the application of error criteria was subjective 
and, at times, inconsistent. This can be attributed to the fact that a single 
person was responsible for determining whether an incident actually corre-
sponded to one of the criteria. It was decided, therefore, to clearly define 
each criteria and what each one involved. The following list presents the 
be observed in order to be recorded. 
stops working to read the user guide. 
each participant in the post-task survey.
not associated with the task currently being performed. 
not associated with the task currently being performed OR where a user 
scrolls through several menus without choosing an option. 
pant asks a question relating to the way a task should be performed or 
whether a task has been completed. 
presses frustration verbally or where a participant’s body language (e.g. sigh-
ing) or facial expressions (e.g. frowning) indicate frustration. 
presses confusion verbally or where a participants body language (e.g. head-
scratching) or facial expressions indicate frustration. 
presses confusion verbally or where a participant’s body language or facial 
expressions indicate frustration (e.g. smiling). 
235
Criterion 1
Tasks Not Completed
: The failure of a user to complete a ta sk
Criterion 2
Times User Guide Used
: Each occasion where the participants 
Criterion 3
PTS Score
: The number of questions answered incorrectly by 
Criterion 4
Incorrect Icon Choices
: Where a user clicks an icon which is 
Criterion 5
Incorrect Menu Choices
: Where a user chooses a menu option 
Criterion 6
Verbal Interactions/Questions
: Each occasion where a partici-
Criterion 7
Observations of Frustration
: Incidents where a participant ex-
Criterion 8
Observations of Confusion
: Incidents where a participant ex-
Criterion 9
Observations of Satisfaction
: Incidents where a participant ex-
modified error criteria and the definition of each criterion which must
despite the administrator identifying the precise page in the user guide; 
ultimately resulting in the administrator giving explicit verbal instructions 
on how to complete the task. 
231). Instead, the number of iincorrect answers was added to the totals for 

---

Assessing Usability 
uses an incorrect shortcut key or types incorrect commands into a field. 
Criterion 11
Stopped Work without Completing Task
: Each instance 
where a participant mistakenly believes the task to be complete or where a