Transcript phl_541_student_t

INTERPRETATION OF DATA
In order to compare "two" or “more than
two” means and to indicate whether the
difference between them is "real" and not
due to "chance" factors, or, in other
words, the difference between them is
"statistically significant", different tests of
significance are performed.
The commonly used tests include:
• The "student's t test" (for comparison
of 2 means).
• The “analysis of variance” or “ANOVA”
(for comparison
means).
of
more
than
2
Significant and Non-significant Differences
Significant:
Between > Within
Non-significant:
Within > Between
120
Control
Treated
Increase in MAP (mm Hg)
100
80
60
*
40
*
*
20
0
0
1
2
3
4
5
Phenylephrine (g/kg)
Comparison of 2 means
6
120
Control
Treatment 1
Treatment 2
Increase in MAP (mm Hg)
100
80
*
60
*
40
20
*
*
*
3
4
*
*
0
0
1
2
5
Phenylephrine (g/kg)
Comparison of 3 means
6
Student’s “t” test
For unpaired data:
"Unpaired" data indicates that the two
means are obtained from 2 different
groups.
Mean
Standard error
e.g. Two groups of rats were used in a
study of the antihypertensive effect of
Losartan. The following results were
obtained:
Does
Losartan
have
a
antihypertensive action ???
significant
Then, by referring to the tables of "t"
distribution at different probability levels
and at the corresponding degree of
freedom, the computed "t" is compared to
the critical "t" (i.e. the "t" value obtained
from probability tables).
If the
computed
t > ifcritical
t at
0.05
On
other hand,
computed
t the
critical
t,
probability
level,
this means
that both
the
this
means that
the difference
between
lowering
in blood
pressure
is due to the
groups
is not
significant
at P<0.05.
drug in 95% of the cases, i.e. this
difference is expected to occur by chance
in less than 5% of the cases. In other
words, the difference between both groups
is significant at P<0.05.
In the provided example, the critical t
corresponding to 14 degrees of freedom
and 0.05 probability is equal to 2.145.
Therefore, since the computed t (3.6) >
critical t (2.145) this means that the
lowering in BP was due to the drug
under study in 95% of the cases and
that the difference between both groups
is significant at P<0.05.
What is your decision
The effect of Oral Hypoglycemic Drug on blood glucose
level was determined in 6 diabetic rats. The level of blood
glucose was as the following:
Before treatment: 210, 190, 220, 200, 240, 215
After treatment: 120, 110, 105, 100, 125, 115
Does this drug have significant hypoglycemic effect,
at
p< 0.05, knowing that tabulated t at degree of freedom 10 =
2.23.
What is your decision
The effect of Oral Hypoglycemic Drug on blood glucose
level was determined in two different groups of 6 diabetic rats each.
The level of blood glucose was as the following:
Control Diabetic rats : 210, 190, 220, 200, 240, 215
Drug-treated rats: 120, 110, 105, 100, 125, 115
Does this drug have significant hypoglycemic effect,
at
p< 0.05, knowing that tabulated t at degree of freedom 10 =
2.23.
What is Your
Decision
You give 9 students a slice of white
donuts and a slice of brown donuts.
Since you are such a skilled researcher
you present the slices of donuts in a
counterbalanced order across your
students.
You measure the number of grams of
white and brown donuts that each
person eats.
Is white donuts is preferred
over brown donuts?
Name
White
of
donuts
persons (grams
eaten)
Ahmed 40
Osman 30
Osama 100
Barney 80
Bretta
0
Betty
10
Barbara 20
Calvani 10
Ben
40
Brown
donuts
(grams
eaten)
50
40
60
90
60
0
50
30
40