Fundamentals, Standard Error, Estimation, Interval Estimation, Hypothesis, Characteristics of Hypothesis, Testing The Hypothesis, Type I & Type II error, One tailed & Two tailed test, Tabulated Values, Chi-square (2) Test, Analysis of variance (ANOVA)Introduction, The Sign Test, The rank sum test or The Mann-Whitney U test, Determination of Sample Size
2. Some Fundamentals
Total No of Units Population (N) &
Sample (n)
Parameter (θ) : The Statistical Measures of Population
are called Parameters.
Population Mean (µ) = ∑xα/N
Population Proportion (P) = X/N
Where X is the no of units possessing
some attribute
Population Variance (σp
2
) = ∑(Xα- µ)2
/N
3. Some Fundamentals
Statistic (T) : The Statistical Measures of Sample are
called Statistics.
Sample Mean (X) = ∑x/n
Sample Proportion (p) = x/n
Where x is the no of units possessing
some attribute
Sample Variance(σs
2
) = ∑(X - X)2
/n-1
* SD of population & sample can be calculated by taking the
square root of variance.
4. Standard Error
A measure of precision achieved by sampling (for
mean & proportion) is called sampling error.
S.E. (X) = σ/√n
S.E. (p) = √pq/n
5. Estimation
Inferring about population parameter on
the basis of statistic is called estimation.
Point Estimation
Interval Estimation
8. Hypothesis
Hypothesis means a tentative result about any
situation or a assumption to be proved or
disapproved. For a researcher it is a question
that he intends to solve
Example: The performance of trained employees
is better than the performance of untrained
employees
9. Characteristics of Hypothesis
It should be clear & precise
It should be capable of being tested
It should state relationship between variables.
It should be stated in a simple terms
It must be able to tasted in a reasonable time
10. Testing The Hypothesis
1. State the null hypothesis (H0) & alternate
hypothesis (Ha or H1).
Example : In a bulb production company, manager
assumed out of 100 bulbs on an average 5
bulbs having some defect.
Null hypothesis H0 : µ = 5
Alternative hypothesis Ha : µ ≠ 5
11. Testing The Hypothesis
2. Establish a level of significance
5% or 1%
3. Choosing a suitable test (calculate the value)
4. Conclude the result
If calculated value < tabulated value then H0
accepted
If calculated value >tabulated value then H0 rejected
12. Type I & Type II error
H0 is True
H0 is False
H0 Accepted H0 Rejected
Type I error
Type II error
Correct
Decision
Correct
Decision
13. One tailed & Two tailed test
If Ha is the type of the greater than or the type of lesser
than, we use one tailed test.
Null hypothesis H0 : µ < 5
Alternative hypothesis Ha : µ > 5
If Ha is of the type “whether greater or smaller” then
we use two tailed test
Null hypothesis H0 : µ = 5
Alternative hypothesis Ha : µ ≠ 5
14. One sample test
When a single sample is taken from the
population and researcher estimate the values
of parameters on the basis of single sample
statistics then one sample test is used to test the
significance of characteristics of sample i.e.
statistics
z-test if n > 30
t-test if n < 30
15. Tabulated Values
Type of z – test 1% 5%
One tailed 2.33 1.645
Two tailed 2.58 1.96
For t – test tabulated valued are referred from statistical table at
desired degree of freedom
16. Two sample test
When two samples are taken from the
population then two sample test is used to test
the significance of difference of sample means
z-test if n > 30
t-test if n < 30
17. Chi-square (χ2) Test
Chi-square test is used either to determine the
association between the two variables or to test
the difference b/w the attributes of two sample
Example:
Whether quinine is affective in controlling fever?
Whether the consumption pattern of two ice
creams is different from each other?
18. Process
Formulate the Hypothesis
Calculate the value of χ2
Compare the calculated value with
tabulated value at the desired degree
of freedom
Conclude the result
If calculated value < tabulated value then
H0 accepted
If calculated value >tabulated value then
H0 rejected
19. Analysis of variance (ANOVA)
Introduction
When we have two sample then to determine
the significant difference between sample
means we apply z-test & t-test, but in the case of
more than two samples we cant apply these
tests.
ANOVA is used when multiple sample cases are
involved, using this technique one can draw
interferences about whether the samples have
been drawn from populations having same
mean
20. The Sign Test
The sign test is the simplest of the non-parametric
tests. Its name comes from the fact that it is
based on the directions (i.e. signs of pluses or
minuses) of a pair of observations and not on
their numerical magnitude.
In any problem where sign test is used, we count:
No. of + signs
No. of – Signs
No. of 0’s
21. Process
Formulate the Hypothesis
Population proportion is always taken 0.5
Calculate the sample proportion which is the
ratio b/w +ive signs & total no. of signs.
Apply z-test formula for proportion & conclude
the results
22. The rank sum test or
The Mann-Whitney U test
This is a very popular test to determine whether
the two independent samples have been drawn
from the same populations or it is used to
determine whether there is a significant
difference between two sets of population
23. Process
Formulate the Hypothesis
Combine two samples and rank them from
maximum to minimum
Calculate the value of U, μ & σ by using
appropriate formula
Apply z-test formula [ z = (U – μ)/σ ]& conclude
the results
24. The one sample run test
The one sample run test is used to judge the
randomness of a sample on the basis of the order
in which the observations are taken.
A run is a succession of identical letters (or other
kinds of symbols) which is followed and preceded
by different letters or no letters at all.
25. Process
Formulate the Hypothesis
H0 :There is a randomness in series
Ha :There is no randomness in series
Count the no. of symbols or letters
n1 = number of occurrences of type 1
n2 = number of occurrences of type 2
r = number of runs ( no. of sets)
Calculate Mean (µr) & S.D. (σr) by using formulas
Calculate value of z and conclude the results.