Scribd
Upload
16 views

Test Statistic_Note

The document explains test statistics used in hypothesis testing, distinguishing between parametric and non-parametric tests. Parametric tests assume normal distribution and require specific conditions, while non-parametric tests do not assume a specific distribution. It details various tests such as the Z-test, t-test, and Pearson's chi-squared test, including their hypotheses, test statistics, and decision criteria.

Uploaded by

JUST RELAX
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
16 views

Test Statistic_Note

The document explains test statistics used in hypothesis testing, distinguishing between parametric and non-parametric tests. Parametric tests assume normal distribution and require specific conditions, while non-parametric tests do not assume a specific distribution. It details various tests such as the Z-test, t-test, and Pearson's chi-squared test, including their hypotheses, test statistics, and decision criteria.

Uploaded by

JUST RELAX
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Test Statistic

Test Statistic The decision to accept or to reject a null hypothesis is made on the basis of a
statistic computed from the sample. Such a statistic is called the test statistic.
There are 2 different types of test statistic.
They are,
a) Parametric Tests
b) Non-Parametric Tests

Parametric test
Parametric tests are statistical tests that make certain assumptions about the parameters
of the population distribution from which the sample is drawn. These tests are typically
used when the data follows normal distribution.
Assumptions:
− Normal Distribution: The data should be approximately normally distributed,
− Homogeneity of Variance: The variances within each group being compared should be
equal.
− Interval or Ratio Data: The data should be measured on an interval or ratio scale.

Non parametric test


Non-parametric tests do not assume a specific distribution for the population and are often
referred to as "distribution-free" tests. These tests are used when the data does not meet
the assumptions required for parametric tests.
Characteristics:
− No Assumption of Normality: These tests do not require the data to follow a normal
distribution.
− Ordinal, Interval, or Ratio Data: Can be used with ordinal data (ranked data) as
well as interval and ratio data.

Parametric tests have specific assumptions that need to be met


1.Normality
2.Homogeneity of variance
3.Independence
Large sample Small sample
Large sample
The sample size n is greater than 30 (n≥30) it is known as large sample. For large
samples the sampling distributions of statistic are normal(Z test).

Small sample
If the sample size n ils less than 30 (n<30), it is known as small sample. For small
samples the sampling distributions are t, F and χ2 distribution.

Degree of freedom
− It is defined as the no. of independent observation which is obtained by subtracting the
no. of constrains from the total no. of observations.
− df = no: of items in the series – 1 = n-1
Where, n is the number of variables in the series in a row or column
Z - test

• Z - test is used to determine whether two population means are different when the
variances(SD known)are known. (in general)
• Sample size is greater than 30. (n ≥ 30) Otherwise, use a t test.
• Quantitative data
• Data should be normally distributed.
• Data should be randomly selected from a population
• When the samples are taken from population are independent

t - test
• The Student t-test is probably the most widely used parametric test.
• The t-statistic was introduced in 1908 by William Sealy Gosset.
• t - test compares the difference between two means of different groups
• All the observations must be independent
• Variable normally distributed
• Sample size should be small (n< 30)
• Quantitative data
• Random samples

Test of Hypothesis of Difference in the case of independent samples


Student’s t-test is used to compare the mean of two completely separate groups. It is also
known as independent sample t-test.
» Hypothesis:
H0: there is no significant difference between 2 means (𝝁𝟏 = 𝝁𝟐 )
H0:𝜇1 = 𝜇2 H1:𝜇1 ≠ 𝜇2

» Test statistics:

̅ 1 −X
X ̅2
t=
S.E
1 1
𝑆. 𝐸(𝑆𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝑒𝑟𝑟𝑜𝑟) = 𝑠√(𝑛 ) + (𝑛 )
1 2

̅ 1 )2 + Σ ( X2 –X
Σ ( X1 –X ̅ 2 )2
s (SD) = √
n1 + n2 – 2

» For t- test the degree of freedom is n1 + n2 – 2


» Then get the table value (critical value, using df. & level of significance

» Decision:
If the calculated test statistic is greater than the appropriate critical table value of the test,
then reject the null hypothesis.
Test of Hypothesis of Difference in the case of dependent samples
Paired t-test(dependent Sample t test) can be used to compare the means of two repeated
measures or compares study subjects at 2 different times (paired observations of the same
subject) [(before - after) (pre – post) ]

“Sometimes multiple measurements are made on the same person, RMANOVA to be used in
such situation”

» Hypothesis:

H0: there is no significant difference between 2 means


H0:𝜇1 = 𝜇2 H1:𝜇1 ≠ 𝜇2

» Test statistics:

̅
X
𝑡= 𝒔
√𝒏
x = x1 – x2
𝜮 (𝑿 − 𝑿) ̅ 𝟐
S. D = 𝑠 = √ 𝒏 −𝟏

» For Z test the degree of freedom is n-1


» Then get the table value (critical value, using df. & level of significance

» Decision:
If the calculated test statistic is greater than the appropriate critical table value of the test,
then reject the null hypothesis.

Pearson's chi-squared test

• Non parametric test


• The sample must be randomly drawn from the population
• Both variables are categorical (ordinal or nominal)
• All observations are independent
• Expected values of cells should be 5
• The chi-squared test compares the significant difference between a set of the observed
frequencies in a contingency table and a set of the corresponding expected (theoretical)
frequencies of a sample.
• Or The Chi Square statistic is commonly used for testing relationships (association)
between categorical variables
» Hypothesis:

H0: There is no significant difference between the observed frequencies and the expected
frequencies.
Or
There is no association between two categorical variables.

» Test statistics:

(𝑶𝒊 - 𝑬𝒊)𝟐
𝝌𝟐 = 𝜮 𝑬𝒊
Where,
Oi = observed value(s)
Ei = expected value(s)

» For paired t- test the degree of freedom is (r – 1) (c – 1),


[(𝑁𝑢𝑚𝑒𝑟 𝑜𝑓 𝑟𝑜𝑤𝑠 − 1)(𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑐𝑜𝑙𝑢𝑚𝑛𝑠 − 1)]
» Then get the table value (critical value, using df. & level of significance

» Decision:
If the calculated test statistic is greater than the appropriate critical table value of the test,
then reject the null hypothesis.

You might also like