11. A number of students in a Grade 9 maths course at a particular high school were randomly divided into two groups. The experimental group received teaching according to the new curriculum while the control group did not. All the students were given a test of computational skill (out of a total of 5 points) after the course and the results were as follows: for the experimental group, the sample size was 27, the sample mean was 1.19 and the sample variance was 74. For the control group, the sample size was 25, the sample mean was 3.54 and the sample variance was 80. Assume that population variances are equal. Construct a 99% confidence interval for the difference between the population mean test scores of the new and old curriculums. (μ_1-μ_2)

1 answer

To construct a 99% confidence interval for the difference between the population mean test scores of the new and old curriculums (i.e., \( \mu_1 - \mu_2 \)), we can follow the steps below. Here, we will define the groups as follows:

  • Group 1: Experimental group (new curriculum)
  • Group 2: Control group (old curriculum)

Step 1: Gather sample statistics

From the problem, we have:

  • Experimental group (Group 1)

    • Sample size (\( n_1 \)) = 27
    • Sample mean (\( \bar{x_1} \)) = 1.19
    • Sample variance (\( s_1^2 \)) = 74
  • Control group (Group 2)

    • Sample size (\( n_2 \)) = 25
    • Sample mean (\( \bar{x_2} \)) = 3.54
    • Sample variance (\( s_2^2 \)) = 80

Step 2: Calculate the pooled standard deviation

Since we assume that the population variances are equal, we can calculate the pooled variance (\( s_p^2 \)):

\[ s_p^2 = \frac{(n_1 - 1)s_1^2 + (n_2 - 1)s_2^2}{n_1 + n_2 - 2} \]

Calculating the pooled variance: \[ s_p^2 = \frac{(27 - 1) \times 74 + (25 - 1) \times 80}{27 + 25 - 2} \]

\[ s_p^2 = \frac{26 \times 74 + 24 \times 80}{50} \]

\[ = \frac{1924 + 1920}{50} = \frac{3844}{50} = 76.88 \]

Thus, the pooled standard deviation (\( s_p \)) is: \[ s_p = \sqrt{76.88} \approx 8.76 \]

Step 3: Calculate the standard error of the difference in means

The standard error (SE) for the difference in means is given by:

\[ SE = s_p \sqrt{\frac{1}{n_1} + \frac{1}{n_2}} \]

Calculating the SE: \[ SE = 8.76 \sqrt{\frac{1}{27} + \frac{1}{25}} = 8.76 \sqrt{0.037037 + 0.04} = 8.76 \sqrt{0.077037} \approx 8.76 \times 0.277 \]

\[ SE \approx 2.43 \]

Step 4: Find the critical value

For a 99% confidence interval with degrees of freedom (\( df = n_1 + n_2 - 2 = 27 + 25 - 2 = 50 \)), we look up the t-value in the t-table. For \( df = 50 \) and a two-tailed test at the 99% confidence level, the critical t-value (approximately) is: \[ t_{0.005, 50} \approx 2.676 \]

Step 5: Calculate the confidence interval

The confidence interval for the difference in means can be calculated using:

\[ (\bar{x_1} - \bar{x_2}) \pm t_{\alpha/2} \cdot SE \]

First, calculate the difference in sample means: \[ \bar{x_1} - \bar{x_2} = 1.19 - 3.54 = -2.35 \]

Applying the values we have:

\[ -2.35 \pm 2.676 \cdot 2.43 \]

Calculating the margin of error: \[ 2.676 \cdot 2.43 \approx 6.52 \]

Now we can construct the confidence interval: \[ (-2.35 - 6.52, -2.35 + 6.52) = (-8.87, 4.17) \]

Final Result

The 99% confidence interval for the difference between the population mean test scores of the new and old curriculums is:

\[ \boxed{(-8.87, 4.17)} \]