TO: Media Services Director
FROM: Student Inte
DATE: October 12, 2018
SUBJECT: Media Services Website Redesign Instructions Usability Data Analysis Plan
The type of study that was conducted was of randomized block design. It is of random design
because all participants in the study were selected randomly. The faculty, staff, and students were
sel
either only the written instructions (Set 1) or the written instructions with visuals (Set 2)), the
amount of time it takes for the faculty, staff, and students to complete the task because it is,
thought that students, faculty, and staff may respond differently to different sets of instruction,
The explanatory variables are the type of person (student, staff, or faculty) and which set of
instructions is used; both of the explanatory variables are categorial nominal. The response
variable is the time taken to complete the instructions which is quantitative continuous.
ted randomly. The design is then blocked because we want to sce how on average (using
I would summarize the data by creating the first 6 different boxplots and 6 different histograms
of the time taken, Each boxplot would be assigned a number (1 (faculty), 2 (staff), or 3 (student))
and a letter (A (Set 1) or B (Set 2).
Twould then combine all the boxplots labelled with a number (1, 2, or 3) and the letter B to
compare the averages of the time spent by faculty, staff, and students using set | of instructions
(w/o visuals) in a comparative box plots graph, using a bar graph of the median times of each
boxplot to make conclusions of about the relationship between the explanatory and response
variables. I would repeat the same process for the A-lettered boxplots to compare the averages of
the time spent by faculty, staff, and students using set 2 of instructions (w/ visuals). We could
repeat the process with the histograms if we are more concerned about making conclusions about
the the average times.
These averages can then be compared to generate some evidence for estimates to begin
answering the following questions. Is there a significant difference in the time taken to complete
the instructions when the instructions contain visuals versus when the instructions don’t contain
visuals. Is there a significant difference in the time taken to complete the instructions when the
subject is faculty, staff, or a student and is this difference the same across different types of
instructions?A possible confounding variable is the type of device used to do that the task (getting to the
website homepage and completing the instructions). Smartphones often load the website at
different speeds compared to that of most computers. Another possible confounding variable is
the internet bandwidth that each subject uses. Because this will most definitely not be constant
across all samples, it may cause changes in the time taken to complete the instructions. A
possible solution to both problems may be just gathering all of the experiment participants in a
controlled environment and have them complete one’s assigned set of instructions one-by-one
where every subject of every sample is using the same exact computer to complete the set of
instructions.