18 extra multiple-choice questions for Tests Based on Sampling Distributions - I (12th Standard Statistics, Samacheer Kalvi), beyond the ones printed in the textbook — each with the correct option highlighted and a clear, worked explanation. Free to read in English and Tamil.
Q1
A random sample of 16 observations from a normal population gives \( \bar{x} = 53 \) and \( s = 4 \). The value of the test statistic \( t_0 \) for testing \( H_0 : \mu = 50 \) is
- A. 0.75
- B. 12
- C. 0.1875
- D. 3Correct
Explanation. \( t_0 = \dfrac{\bar{x}-\mu_0}{s/\sqrt{n}} = \dfrac{53-50}{4/\sqrt{16}} = \dfrac{3}{1} = 3 \). Dividing by s alone gives 0.75; multiplying by \( \sqrt{n} \) only in the numerator without dividing by s gives 12.
Q2
Two independent small samples of sizes \( m = 10 \) and \( n = 8 \) are used to test the equality of two normal population means with equal unknown variances. The degrees of freedom of the \( t \) statistic are
- A. 17
- B. 9
- C. 16Correct
- D. 18
Explanation. The two-sample statistic follows the t-distribution with \( m+n-2 = 10+8-2 = 16 \) degrees of freedom, as in Example 2.3, where \( t_{16,0.005} = 2.921 \) was used.
Q3
For two independent samples of sizes \( m = 10 \) and \( n = 8 \), \( \sum (x_i-\bar{x})^2 = 108 \) and \( \sum (y_j-\bar{y})^2 = 38 \). The pooled variance \( s_p^2 \) is
- A. 9.125Correct
- B. 8.11
- C. 17.4
- D. 73
Explanation. \( s_p^2 = \dfrac{(m-1)s_x^2+(n-1)s_y^2}{m+n-2} = \dfrac{108+38}{10+8-2} = \dfrac{146}{16} = 9.125 \). Dividing by \( m+n = 18 \) gives 8.11; 17.4 adds the two sample variances 12 and 5.4.
Q4
In a paired \( t \)-test on 9 pairs of observations, the mean of the differences is \( \bar{d} = 2 \) and their standard deviation is \( s = 3 \). The calculated value \( t_0 \) is
- A. 18
- B. 2Correct
- C. 6
- D. 0.67
Explanation. \( t_0 = \dfrac{\bar{d}}{s/\sqrt{n}} = \dfrac{2}{3/\sqrt{9}} = \dfrac{2}{1} = 2 \), with \( n-1 = 8 \) degrees of freedom. Using n instead of \( \sqrt{n} \) gives 6; ignoring \( \sqrt{n} \) gives 0.67.
Q5
A sample of 11 observations from a normal population with unknown mean has variance \( s^2 = 18 \). The value of \( \chi_0^2 \) for testing \( H_0 : \sigma^2 = 12 \) is
- A. 16.5
- B. 1.5
- C. 6.67
- D. 15Correct
Explanation. \( \chi_0^2 = \dfrac{(n-1)s^2}{\sigma_0^2} = \dfrac{10 \times 18}{12} = 15 \), with 10 degrees of freedom. Using n = 11 instead of \( n-1 \) gives 16.5; inverting the ratio gives 6.67.
Q6
In a contingency table with total frequency \( N = 400 \), a cell lies in a row with total 120 and a column with total 150. Its expected frequency under the hypothesis of independence is
- A. 500
- B. 18000
- C. 45Correct
- D. 135
Explanation. \( E_{ij} = \dfrac{R_i \times C_j}{N} = \dfrac{120 \times 150}{400} = \dfrac{18000}{400} = 45 \). Forgetting to divide by N leaves 18000; averaging the two totals gives 135.
Q7
In a \( 2 \times 2 \) contingency table, \( a = 30, b = 10, c = 10, d = 30 \) and \( N = 80 \). Using \( \chi_0^2 = \dfrac{N(ad-bc)^2}{(a+b)(c+d)(a+c)(b+d)} \), the value of \( \chi_0^2 \) is
- A. 20Correct
- B. 0.25
- C. 0.025
- D. 40
Explanation. \( ad-bc = 900-100 = 800 \); \( \chi_0^2 = \dfrac{80 \times 800^2}{40 \times 40 \times 40 \times 40} = \dfrac{51200000}{2560000} = 20 \), with 1 degree of freedom. Leaving out N gives 0.25.
Q8
A Poisson distribution is fitted to a frequency distribution, its mean being estimated from the data. After pooling the classes with expected frequency less than 5, there are 7 classes. The degrees of freedom of the chi-square statistic are
- A. 7
- B. 5Correct
- C. 6
- D. 4
Explanation. Degrees of freedom \( = k-1-s = 7-1-1 = 5 \), since one parameter, the mean m, is estimated. In Example 2.12, 4 classes after pooling gave \( 4-1-1 = 2 \).
Q9
The observed frequencies of three classes are 50, 30 and 20 (total 100). Under \( H_0 \) they are in the ratio 2 : 2 : 1. The value of \( \chi_0^2 = \sum \dfrac{(O_i-E_i)^2}{E_i} \) is
- A. 2.5
- B. 0
- C. 200
- D. 5Correct
Explanation. Expected frequencies are \( \tfrac{2}{5} \times 100 = 40 \), 40 and \( \tfrac{1}{5} \times 100 = 20 \). \( \chi_0^2 = \dfrac{100}{40}+\dfrac{100}{40}+0 = 5 \). The sum of \( (O_i-E_i) \) is always 0; 200 is \( \sum (O_i-E_i)^2 \).
Q10
The degrees of freedom of the chi-square statistic for testing independence of attributes in a contingency table with 3 rows and 4 columns are
- A. 5
- B. 7
- C. 6Correct
- D. 12
Explanation. Degrees of freedom \( = (m-1)(n-1) = (3-1)(4-1) = 2 \times 3 = 6 \). 12 is the number of cells and 5 comes from \( (3-1)+(4-1) \).
Q11
Four unbiased coins are tossed 160 times and a binomial distribution with \( p = \tfrac{1}{2} \) is fitted. The expected frequency of getting no head is
- A. 10Correct
- B. 40
- C. 80
- D. 32
Explanation. \( P(0) = {}^4C_0 (1/2)^4 = \tfrac{1}{16} \). Expected frequency \( = N \times P(0) = 160 \times \tfrac{1}{16} = 10 \). In Example 2.11, five coins tossed 640 times gave \( 640 \times 0.03125 = 20 \).
Q12
W.S. Gosset published his research on the t-distribution under the pseudonym
- A. Oxford
- B. StudentCorrect
- C. Biometrician
- D. Brewer
Explanation. The brewery in Ireland where Gosset worked allowed him to publish only under the pseudonym "Student", so the distribution is called Student's t-distribution.
Q13
Consider the following statements about the \( t \)-distribution:
1. It is symmetrical with mean zero.
2. The normal curve is higher in the middle than the \( t \) curve.
3. The \( t \)-distribution has less area in the tails than the normal distribution.
Which of these are correct?
- A. 1 and 3 only
- B. 2 and 3 only
- C. 1, 2 and 3
- D. 1 and 2 onlyCorrect
Explanation. The t-distribution is symmetrical with mean zero and the normal curve is higher in the middle. But t has a greater spread sideways, so there is more area in its tails; statement 3 is false.
Q14
Assertion (A): When the sample size is less than 30, the large sample procedures based on the standard normal distribution cannot be applied.
Reason (R): For small samples, the sampling distributions of test statistics are far from normal.
- A. Both A and R are true and R is the correct explanation of ACorrect
- B. Both A and R are true but R is not the correct explanation of A
- C. A is true but R is false
- D. A is false but R is true
Explanation. The chapter introduction states that for n < 30 the sampling distributions of test statistics are far from normal, which is exactly why the t-distribution is used instead of the standard normal.
Q15
Which of the following is a requirement for applying the chi-square test of independence of attributes?
- A. The total frequency N should be reasonably large, say greater than 50Correct
- B. The sample size should be less than 30
- C. The population variance must be known
- D. Every observed cell frequency must be equal
Explanation. The book notes that N should be reasonably large (greater than 50) and that no expected cell frequency should be less than 5; small cells are pooled with neighbouring cells.
Q16
In the \( t \)-test for a normal population mean, the alternative hypothesis is \( H_1 : \mu < \mu_0 \). The rejection rule at level \( \alpha \) is
- A. \( t_0 < -t_{n,\alpha} \)
- B. \( t_0 < -t_{n-1,\alpha} \)Correct
- C. \( t_0 > t_{n-1,\alpha} \)
- D. \( |t_0| \ge t_{n-1,\alpha/2} \)
Explanation. For a left-tailed alternative the critical value is \( -t_{n-1,\alpha} \) and \( H_0 \) is rejected when \( t_0 < -t_{n-1,\alpha} \). The rule \( |t_0| \ge t_{n-1,\alpha/2} \) belongs to the two-sided alternative.
Q17
Match the test with the degrees of freedom of its test statistic:
(a) t-test for a single mean - (1) \( (m-1)(n-1) \)
(b) t-test for two independent means - (2) \( k-1-s \)
(c) chi-square test of independence - (3) \( n-1 \)
(d) chi-square test of goodness of fit - (4) \( m+n-2 \)
- A. (a)-3, (b)-1, (c)-4, (d)-2
- B. (a)-4, (b)-3, (c)-2, (d)-1
- C. (a)-3, (b)-4, (c)-2, (d)-1
- D. (a)-3, (b)-4, (c)-1, (d)-2Correct
Explanation. Single mean: n-1; two independent means: m+n-2; independence in an m x n contingency table: (m-1)(n-1); goodness of fit with k classes and s estimated parameters: k-1-s.
Q18
Karl Pearson first examined whether observed data support a given specification, the chi-square goodness of fit test, in a paper published in
- A. 1906
- B. 1937
- C. 1900Correct
- D. 1911
Explanation. Pearson's chi-square paper appeared in 1900. In 1911 he founded the world's first university statistics department at University College, London; 1906 relates to Gosset's stay in his laboratory.