You are on page 1of 19

Tests of Significance

of Small Samples
Small Samples
• When size of sample is less than 30

• Normality is not achieved

• The sample values are not sufficiently close to the


population value

• Contribution of Sir William Gosset

• Gosset gave a test popularly known as ‘t-test’

• These tests are based on ‘t’ distribution


Properties of t-distribution
1. The variable t-distribution ranges from minus infinity to plus
infinity

2. Thus f(t) is depending on degrees of freedom (𝜈)

3. Like the standard normal distribution, the t-distribution is


symmetrical and has a mean zero.

4. The variance of the t-distribution is greater than zero.

5. Thus, the variance of the t-distribution approaches the


variance of the standard normal distribution as the sample
size increases.
Test of
Hypothesis
about the
Population
Mean
Student’s distribution
• The t-distribution is commonly called
Student’s t-distribution or simply Student’s
distribution.
• The t-distribution is used when sample size
is less than 30 and the population standard
deviation is unknown.

𝑋ത − 𝜇
𝑡= 𝑋 𝑛
𝑆


∑ X− X 2
• Where S =
n−1
Ex. 1

• The systolic blood pressure of 10 persons in the age group of 45-50 is


given below:

148, 128, 147, 127, 150, 145, 124, 140, 142, 149

• In the light of the data, discuss the suggestion that the average
systolic blood pressure of the population is 150. (for ʋ = 9, t0.05 = 2.26)
Ex. 2

• An operator claims that he produces 40 articles in an hour. A sample


of 10 random hours shows the turns out as 43, 45, 38, 37, 41, 42, 44,
39, 43, and 38. Is the claim of the operator reasonable at 5%
significance level? Assume the distribution of hourly turnout of the
operator to be normal and critical region at 5% level for 9 df to be
2.262.
Ex. 3

• A random sample of size 16 has 53 as mean. The sum of squares of


the deviations taken from the mean is 135. can this sample be
regarded as taken from the population having 56 as mean?

• (for ʋ = 15, t0.05 = 2.13)


Testing Difference Between
Means of Two Samples
(Independent Samples)
𝑋ത 1 − 𝑋ത 2 𝑛1 𝑛2
𝑡= 𝑋
𝑆 𝑛1 + 𝑛2

Where , 𝑋ത 1 = mean of the first sample


• 𝑋ത 2 = mean of the second sample
• 𝑛1 = number of observations in the first sample
• 𝑛2 = number of observations in the second sample

• S = Combined Standard Deviation


Σ 𝑋1−𝑋1 2 +Σ 𝑋2−𝑋2 2
• 𝑆=
𝑛1+𝑛2 −2
Ex. 4
• Two types of diet were used on 7 and 5 chickens reared in a research station.
• The weight of the chickens which were fed with high protein diet in ounces is
given below.
• 13, 16, 12, 17, 15, 17, 15
• The weight of the chickens which were fed with low protein diet in ounces is
given below.
• 9, 11, 15, 11, 14
• Test whether there is significant evidence that additional diet has increase the
weight of chickens. (For 𝑣 = 10, t0.05=2.23)
Illustration 2
• Two types of drugs were used on 5 and 7 patients for reducing their
weight. Drug A was imported and Drug B indigenous. The decrease in
the weight after using the drugs for six months was as follows:

Drug A 10 12 13 11 14
Drug B 8 9 12 14 15 10 9

• Is there a significant difference in the efficacy of the two drugs? If not,


which drug should you buy? (For 𝑣 = 10, t0.05=2.223)
Illustration 3
• For a random sample of 10 persons, fed on diet A, the increased
weight in pounds in a certain period were:
10 6 16 17 13 12 8 14 15 9
• For another random sample of 12 persons, fed on diet B, the increase
in the same period were:
7 13 22 15 12 14 18 8 21 23 10 17
• Test whether the diets A and B differ significantly as regards their
effect on increase in weight. Given the following:
d.f 19 20 21 22 23
Value of t at 5% level 2.09 2.09 2.08 2.07 2.07
Test of Hypothesis about the
difference between Two Means with
Dependent Samples
Means of dependent samples
• Paired t-test
• If two samples are dependent, they must have the same number of
units
• E.g., find out the effect of training, difference in the efficacy of two
drugs
𝑑ത 𝑛 Σ(𝑑−𝑑)2
•𝑡= where 𝑠 = where as
𝑠 𝑛−1
• 𝑑ҧ is the mean of the differences of the paired values
• S is the standard deviation of the differences
Ex. 1
An IQ test was administered to 5 persons before and
after they were trained. The results are given below.

Candidates 1 2 3 4 5
IQ before training 110 120 123 132 125
IQ After training 120 118 125 136 121

Test whether there is any change in IQ after the training


programme. (t0.05 = 4.6 for 4 df)
Ex. 2

• The following data relates to weights of 10 persons before joining a


health club and after 6 months from joining it. Test whether there is a
significant change in the weights after 6 months from joining the club.
Use 5% significant level t 0.05 = 2.26 for 9 df.

Weight before 120 125 115 130 123 119 122 127 128 118
Weight after 111 114 107 120 115 112 112 120 119 112
Ex. 3
• The weight of 10 guinea pigs (in grams) when brought in the
laboratory and after one month were recorded in the following table.
Calculate and conclude whether the gain in weight is statistically
significant or not. (t 0.05 = 2.26 for 9 df)

In the
49 41 37 41 42 37 39 38 41 35
beginning
After one
52 43 46 52 46 38 42 41 42 38
month

You might also like