Tip:
Highlight text to annotate it
X
Welcome to Epi Info 7 training.
In this video you'll learn how to perform a complex sample analysis
in the visual dashboard.
I'm in the visual dashboard and I'm using a fictitious dataset of weekly
classroom absenteeism in elementary schools sampled from each of three
counties.
We want estimate the proportion of students absent in the previous week.
We have a stratified cluster design so we need to use a complex sample analysis
to estimate.
I'll open a line list for reference to show all the variables.
We have County, School, Size, Classroom, Selection Probability, Weight (which is the
inverse of the probability,)
Absent, and High Absent.
Now let's navigate to the Complex Sample Means in the Advanced Statistics
section.
And we have to choose our fields.
Means of is
Absent,
Weight
is Weight,
Stratify by
is County,
Primary Sampling Unit
is School,
and I'll click Run.
The output shows the weighted mean and other statistics
taking into consideration the study design which answers our first question
of average absenteeism in the three counties. Now to answer the second
question, we'll run a frequency to find those classrooms whose absent proportion
exceeded .1;
and not just any frequency but a Complex frequency.
So I'll navigate to Advanced Stats again
and I'll choose Complex Sample Frequencies.
Frequency of is High Absent,
Weight is Weight,
Stratify is County,
and Primary Sampling Unit is School.
And I'll click Run.
This output shows the weighted percents and other associated stats.
The High Absent column
has both a Yes and a No section and the column percent is what we're looking
for.
And it indicates that 16% of the classes in the three counties
exceeded .1 absent in the previous week.
You've just been shown how to perform a complex sample analysis in the visual
dashboard.
Thanks for watching and look for more user support videos coming soon.