Page 162 -

6.7 Analysis of diaries 149

crash? Or was it a hard drive crash? It is necessary to provide participants with a list
of terms and how they should be used, along with specific details of what should
be recorded.
It is also very important to define for participants when they should make a di-
ary recording. Just saying, “when you feel like it” is not sufficient as, many times,
this will not provide enough motivation or clarity. Often, diarists do not immedi-
ately sense the importance of their entries and, especially with diaries that are rela-
tively unstructured, one of the big challenges is convincing participants that what
they are doing is important. They may feel that there is nothing to report, that
what's going on is mundane. At the same time, the number of diary entries should
not be linked directly to payment for participation. For instance, if participants are
paid, $2 for each diary entry, there is a good chance that they will attempt to make
many diary entries. In the Time Diaries to Study User Frustration sidebar, if the
method had been modified so that users were paid $5 every time that they filled out
a frustration experience report, the chances are good that users would get frustrated
very often and fill out many reports, regardless of how they were feeling; this could
bias the data so that it is unrepresentative. Any payment should be for regular par-
ticipation but should not be linked directly to the number of entries. Participants
should get paid for taking part in the study, regardless of the number of entries. Each
diary entry should be triggered by an event, a time, or a sense of importance, not by
financial compensation.
Throughout the period of the study (and 2 weeks is often an appropriate length of
time), it might be necessary to encourage participants to keep making diary entries.
If diary reports are turned in during the study period (not only at the end), you may
be able to monitor the diary reports and give feedback to users who are not providing
useful data. For instance, in a diary study of the information-seeking needs of mobile
device users (Sohn et al., 2008), the participants were sent five text messages a day,
reminding them to send in text messages which served as basic diary entries (and
which were then followed up later in the day). It is always a good idea to give feed-
back to diary participants, not on their specific entries (which might bias the data) but
on the existence of their diary entries, on a regular basis. Another interesting study
reported on the use of smartwatches in the service of reminding participants when to
record a diary entry. For a diary relating to tracking food items eaten, whenever the
smartwatch detected hand gestures that typically represented hand-to-mouth eating,
the smartwatch sent a message to a smartphone app, reminding the participant to
record what they were eating (Ye et al., 2016).

6.7 ANALYSIS OF DIARIES
Once the diaries are collected, the next step is to analyze the diary entries or reports.
Depending on the media used to collect the diary entries (such as paper), it may be
necessary first to transfer the diary entries to an electronic format. Hopefully, if any
handwriting was done in paper diaries, the handwriting is legible and not open to
potential debate!