Professional Documents
Culture Documents
Probability
Curve
By:
Keerthi Samuel.K,
Lecturer
Vijay Marie College Of Nursing
INTRODUCTION
• All the observations in the selected sample
should be around the mean. This phenomena
is called normal probability or normal
distribution.
• Some variables deviate from mean which are
measured by using measures of dispersion
like standard deviation.
INTRODUCTION
• For individual and discreete series poison
distribution curve is used whereas for
continuous data normal distribution is used,
which is also called normal probability
distribution.
• Many statistical data in medicine are
displayed in the form of normal curve.
POISSION CURVE OR BIMODAL
CURVE
NORMAL PROBABILITY CURVE
• It was first discovered by De.Moivre as a
limitation of poison model in 1773.
• But the credit comes to Karl Gauss who first
made the reference in 1809.
PRINCIPLES
• When a large number of samples like height
and weight are collected, a frequency
distribution table is prepared by keeping
small class interval. Then the features are
seen as:
1. Some observations are above the mean and
some are below the mean.
2. If they are arranged in an order , maximum
of them will be around the mean and fewer
at the extremes
PRINCIPLES
decreasing smoothly on both the sides.
3. Normally half the observations are
symmetrically distributed on each side of the
mean.
Normal curve
Total area = 1
x
μ 3σ μ 2σ μσ μ μ+σ μ + 2σ μ + 3σ
1
Means and Standard
Deviations
Example:
2. Which curve has the greater mean?
3. Which curve has the greater standard deviation?
B
A
x
1 3 5 7 9 11 13
Curve B is more spread out than curve A, so curve B has the greater standard
deviation.
x
6 7 8 9 10
Height (in feet)
z
3 2 1 0 1 2 3
z = Value - Mean =
xμ- .
Larson &Standard deviation
Farber, Elementary Statistics: Picturing the World , 3 eσ 16
Construction of a normal curve
• We need mean and SD . There are two methods:
z= Value - Mean
Standard deviation
• Find the ordinate at each of these distances from z table
• Multiply each of these by 0.39 and take this values
frequencies corresponding to midpoints.
m f Z
10-20 15 2 2-31/3=10
20-30 25 1 1-31/3
30-40 35 3 3-31/3
40-50 45 2 2-31/3
N=8
Construction of a normal curve