Page 323 -

312 CHAPTER 11 Analyzing qualitative data

Table 11.2 Some Examples of Statements to Look for While Coding
Statement Examples
Objectives Use computers for educational purposes
Actions Enter a password, chat online
Outcomes Success or failure, whether the objective is achieved
Consequences Files unintentionally deleted, a specific application abandoned
Causes Limited memory, dated equipment
Contexts User is computer savvy, user works with classified information
Strategies Avoid specific tasks, multimodal interaction

described the outcome of the action. Was the action successfully completed? Did the
action completely fail? Was the action partially completed? Whenever an action is not
completely successful, you may want to pursue the consequences or costs of the unsuc-
cessful action: Is the consequence highly detrimental? Does it cause the user to lose
several days of work? Does it prevent the user from completing some tasks on time?
Is it a minor nuisance or is it so frustrating that the user decides to abandon the action?
Causes are also associated with failed actions. Whenever an action completely
or partially fails, it is worth pursuing the causes of the failure. Does the failure trace
back to the user or the application? If it is caused by the user, what kinds of capabil-
ity are involved? Is it due to cognitive overload? Is it due to lack of attention? Is it
due to physical or perceptual limitations? Or is it due to the interaction between two
or more of those factors. Statements about the context of the interaction or usage are
also important. Different types of user may report different satisfaction levels for
the same application with similar performance measures because the comparison
context is drastically different (Sears et al., 2001). Finally, descriptions of interac-
tion styles and strategies are also valuable information that is hard to examine during
empirical lab-based studies.

11.4.2.2 Ask questions about the data
A good way to help detect interesting patterns and connections in data is to con-
stantly ask questions about the data. In Section 11.4.2.1, we listed a series of ques-
tions that you can ask once you identify an interesting action in the data. Those
questions can be related to the specific action, its outcome, and its consequence, as
well as the causes of failed actions. Most of those questions are practical questions
that may help you identify interaction challenges and design flaws.
Corbin and Strauss (Corbin and Strauss, 2014) discussed the art of asking ques-
tions in a larger context with the primary objective of theory development. They pro-
posed four types of questions and two of them are particularly important during the
analysis phase: sensitizing questions and theoretical questions. Sensitizing questions
help coders better understand the meaning of the data: What is happening here? What
did the user click? How did the user reach the specific web page? Theoretical ques-
tions help the researchers make connections between concepts and categories: What
is the relationship between two factors? How does the interaction change over time?