Imagine, for a second, that you
would like to test a hypothesis involving participants who selected specific options on
several different questions. Yes, it is possible, and it is very easy,
if you know what questions you are looking for.
To teach you how to do this, I will give you a few
examples from the
Liver Flush Survey. We are going to test a few hypothesis in
order to learn how to filter accumulated survey data.
Hypothesis: People who do not
complete a thorough bowel cleanse prior to flushing are more likely to
experience problems during or after a liver flush.
To test this hypothesis, the first thing
I am going to do is to select all people who have completed a thorough bowel
cleanse. I want to find out what is the level of problems associated
with those who completed a thorough bowel cleanse, and then I am going to
compare those data with those who haven't done any bowel cleansing.
Let's search for all people who completed a
thorough bowel cleanse. It is easy cause a thorough bowel cleanse is one
of the main questions. Let's go to that question page:
Here it is:
http://curezone.com/faq/p/r.asp?a=28,571,2826&s=21&u=50
Results of that question:
1 |
No. Never. |
15 of 55 |
27% |
|
2 |
Yes, before flushing |
32 of 55 |
58% |
|
3 |
Maybe. I completed a bowel cleanse, but I
don't really know if it was thorough enough. |
7 of 55 |
13% |
|
So, we have 32 people (58%) who completed a
thorough bowel cleanse before flushing. To continue working with only those
32, you now click on
58% which is actually a clickable link.
That link brings you to a page that lets
you select a second question.
We are going to select this one as our second question: Major Risks? Major
symptoms and problems encountered during a liver flush:
http://curezone.com/faq/p/r2.asp?a=28,571,2826&s=21&u=50&o=10352&u2=4
Results of that question:
1 |
Yes. I will select below: |
4 of 18 |
22% |
|
2 |
No. |
14 of 18 |
78% |
|
As we can see,
22% of those who completed a thorough bowel cleanse encountered some
major problems.
Now, we are going to repeat the same process,
but this time we will select those who did not complete a thorough bowel
cleanse.
1 |
No. Never. |
15 of 55 |
27% |
|
2 |
Yes, before flushing |
32 of 55 |
58% |
|
3 |
Maybe. I completed a bowel cleanse, but I
don't really know if it was thorough enough. |
7 of 55 |
13% |
|
To continue working with only those 15
who selected NO, you now click on
27% .
That link brings you to a page that lets
you select the second question: Major Risks? Major symptoms and
problems encountered during a liver flush:
http://curezone.com/faq/p/r2.asp?a=28,571,2826&s=21&u=50&o=10351&u2=4
Results of that question:
1 |
Yes. I will select below: |
3 of 11 |
27% |
|
2 |
No. |
8 of 11 |
73% |
|
As we can see, 27% of those who completed a
thorough bowel cleanse encountered some major problems.
So, the question is: Is there a
significant difference between 22% compared to 27%?
Keep your feet on the ground, and remember that
at time of writing this, very few people actually answered all survey
questions, so we have only 3 people who encountered major risks and
8 who didn't.
Those numbers, 3 and 8 are far
too small to draw any valid conclusion.
The most important thing here is that you have now (I hope) learned how to
filter survey data.
Next hypothesis that you should test:
Hypothesis:
People who do not complete a thorough bowel cleanse AND people who do not
complete a thorough parasites cleanse are more likely to experience problems
during or after a liver flush.
Testing this hypothesis asks for filtering data
even deeper.
First we select all people who have
completed a thorough bowel cleanse:
Here they are:
http://curezone.com/faq/p/r.asp?a=28,571,2826&s=21&u=50
1 |
No. Never. |
15 of 55 |
27% |
|
2 |
Yes, before flushing |
32 of 55 |
58% |
|
3 |
Maybe. I completed a bowel cleanse, but I
don't really know if it was thorough enough. |
7 of 55 |
13% |
|
Then we click on
58% select the second question: Parasites Cleanse:
http://curezone.com/faq/p/r2.asp?a=28,571,2826&s=21&u=50&o=10352&u2=53
1 |
Yes, before flushing |
14 of 18 |
78% |
|
2 |
Yes, on the same day of liver flush |
6 of 18 |
33% |
|
So, there are 14 people (78%
) who completed both parasites cleanse and bowel cleanse.
Now, we click on
78% and select the third question: Major Risks:
http://curezone.com/faq/p/r3.asp?a=28,571,2826&s=21&u=50&o=10352&u2=53&o2=10766&u3=4
1 |
Yes. I will select below: |
3 of 14 |
21% |
|
2 |
No. |
11 of 14 |
79% |
|
This is the result: 79% did not suffer any
major symptoms or problems.
We repeat the same process, but this time we
filter and find those that have not completed neither of cleanses:
http://curezone.com/faq/p/r3.asp?a=28,571,2826&s=21&u=50&o=10351&u2=53&o2=10765&u3=4
This is the result. 100% of those who never
competed a Bowel Cleanse and never completed a Parasites Cleanse
NEVER
suffered major problems or symptoms.
Of course, at time of writing this, number of
people who never completed cleanses is 10. That number can not be considered as
a representative number for all the people who are flushing. Let's wait until
we have 100 people before we try to draw any conclusion.
If you have any question or a suggestion related
to data
filtering, please ask that question inside
Survey Discussion Forum.
What you can not do?
You can not select any of the Standardized Questions as your first
question. Standardized Questions must be selected as your last question,
when filtering data. These are the limitations of the software.
You can select maximum 3 questions. So far, software allows max of 3
questions.
To exercise, try to test next hypothesis:
1: People who suffered constipation before flushing are more likely to
encounter problems during a liver flush.
2: People who suffered constipation before flushing don't drink enough
water. (Water intake is one of the standardized questions and must be selected
as a last question.)
3: Women are more likely to encounter problems during a liver flush then
men. (Gender is one of the standardized questions and must be selected
as a last question.)
|