Name: _______________________________ K E Y
Student ID: _______________________________

MATH102 11SP Midterm ch7-10
B [ answers in web view ] Total points: 70

Let the words of my mouth and the meditation of my heart
Be acceptable in Your sight, O LORD, my Rock and my Redeemer.
-- Psalm 19:14
  1. Drug company "Faizer" claims their antidepressant is preferred by 30 out of 40 physicians. In response, drug company "Vonartis" claims their antidepressant is preferred by 55 out of 60 physicians. Is Vonartis' drug significantly more preferred than Faizer's?
    1. State the null and alternative hypotheses, both in words and in notation. [3]
      H0: binomial proportion pV ≤ pF
      HA: binomial proportion pV > pF
    2. What statistical test is be appropriate to test the hypothesis?
      Should it be 1-tailed or 2-tailed? [2]
      Comparing two independent proportions, 1-tailed.
    3. Run the test (either p-value or classical approach) and draw a conclusion. [4]
      SEV = sqrt(pq/n) = sqrt(0.917*0.083/60) = 0.03568
      SEF = sqrt(pq/n) = sqrt(0.75*0.25/40) = 0.068465
      SEp1-p2 = sqrt(0.917*0.083/60 + 0.75*0.25/40) = 0.07721
      z = ( (p'V - p'F) - 0 ) / SE = (0.917 - 0.75) / 0.07721 = 2.1588
      P-value: p = 0.0156
      Classical: z* = 1.65
    4. Interpret your conclusion in the context of the two drug companies.
      Please use complete English sentences. [2]
    5. What assumptions did you rely upon in conducting the test? Are the assumptions met? Why? [2]
      Random sampling of physicians (this is a big assumption here!), np = 30 > 5, nq = 10 > 5, np = 55 > 5, nq = 5 ≥ 5 (borderline!).
  2. Human beta-endorphin (HBE) is a hormone secreted by the pituitary gland under conditions of stress (like exams!). Suppose we wish to determine whether blood concentration of HBE (pg/mL) is different for men who exercise regularly as compared with men who do not exercise regularly.
    1. State the null and alternative hypotheses, both in words and in notation. [3]
      [H0: μex = μno: HBE levels are the same for both groups.
      HA: μex ≠ μno: HBE levels are different for the two groups.
    2. What statistical test is be appropriate to test the hypothesis?
      Should it be 1-tailed or 2-tailed? [2]
      t-test on independent groups, 2-tailed.
    3. Data for this experiment are given below. Sketch boxplots for the data, on a common axis (number line). [4]
      Mean:SD:
      Exercisers: 60 58 63 49 51 43 54 546.9282
      Non-exercisers: 41 37 51 60 28 35 4211.6276
    4. Run the test (either p-value or classical approach) and draw a conclusion. [4]
      SEE = 2.6186, SEN = 4.7469, SE = 5.4213.
      mean diff = 12, so t = 2.2135
      df = min(n1, n2) - 1 = 5 (real df = 6.45).
      P-value: two-tailed: 0.05 < p < 0.10 (real p = 0.0778).
      Classical: t* = 2.5706
    5. Interpret your conclusion in the context of the original research question.
      Please use complete English sentences. [2]
      Fail to reject H0, HBE levels for the two groups are not significantly different.
    6. What assumptions did you rely upon in conducting the test? [2]
      Random sampling on both groups; HBE levels are normally distributed in both groups; variance of HBE levels is similar in both groups.
  3. Does HBE concentration in men increase after they exercise?
    Data from a study of 6 men are below.
    MeanSD
    Before: 425539 504947 47 5.762
    After: 475544 514954 50 4.195
    1. State the null and alternative hypotheses, both in words and in notation. [3]
      H0: μd ≤ 0 (presuming d = after - before; you could also subtract in the other order), HBE concentration does not increase after exercise.
      HA: μd > 0, HBE concentration does increase after exercise.
    2. What statistical test is be appropriate to test the hypothesis?
      Should it be 1-tailed or 2-tailed? [2]
      Dependent t-test on pairwise differences. 1-tailed.
    3. Run the test (either p-value or classical approach) and draw a conclusion. [4]
      mean diff = 3.00, SD of diffs = 3.033,
      SEd = 1.238, so t = 2.423. df=5, one-tailed.
      P-value: 0.02 < p < 0.05 (real p = 0.03)
      Classical: t* = 2.015.
    4. Interpret your conclusion in the context of the original research question.
      Please use complete English sentences. [2]
      Reject H0, HBE concentration rises after exercise.
    5. What assumptions did you rely upon in conducting the test? [2]
      Random sampling of men; change in HBE concentration is normally distributed.
  4. A factory needs to ensure that the widgets it produces have variance no more than 2.5mm2. An inspector from corporate headquarters randomly selects 41 widgets from the factory, to check if the factory is within specifications. Those 41 widgets have a variance of 3.6mm2 in length.
    1. State the null and alternative hypotheses, both in words and in notation. [3]
      H0: variance σ2 ≤ 2.5
      HA: variance σ2 > 2.5
    2. What statistical test is be appropriate to test the hypothesis?
      Should it be 1-tailed or 2-tailed? [2]
      χ2 (chi-squared), 1-tailed.
    3. Run the test (either p-value or classical approach) and draw a conclusion. [4]
      χ2 = 57.6, df = 40.
      P-value: 0.025 < p < 0.05 (actual p = 0.03527)
      Classical: χ2* = 55.8
    4. Interpret your conclusion in the context of the original research question.
      Please use complete English sentences. [2]
      Reject H0: the factory is out of spec.
    5. What assumptions did you rely upon in conducting the test? [2]
      Random sampling; length of widgets is normally distributed.
  5. "Faizer" and "Vonartis" also produce competing blood glucose monitors. An independent lab obtains one glucose monitoring device from each company. A single sample of blood is tested 15 times in each company's glucose monitoring device. Faizer's device yields a standard deviation of 0.35 mmol/L; Vonartis' device yields a standard deviation of 0.22 mmol/L. Is there a difference in the precision of the two companies' devices?
    1. State the null and alternative hypotheses, both in words and in notation. [3]
      H0: σF = σV, or σF2 / σV2 = 1
      HA: σF ≠ σV, or σF2 / σV2 ≠ 1
    2. What statistical test is be appropriate to test the hypothesis?
      Should it be 1-tailed or 2-tailed? [2]
      F-test comparing two variances, 2-tailed.
    3. Run the test (either p-value or classical approach) and draw a conclusion. [5]
      F = 0.352 / 0.222 = 2.5310, df = (14, 14)
      2-tailed p-value: p ≅ 0.10 (actual p = 0.0934)
      Classical: F*(12,14,0.025) = 3.05, F*(15,14,0.025) = 2.95
    4. Interpret your conclusion in the context of the two drug companies.
      Please use complete English sentences. [2]
      Fail to reject H0: no significant difference in precision between the two companies' blood glucose monitors.
    5. What assumptions did you rely upon in conducting the test? [2]
      Random sampling of blood, normal distribution of blood glucose levels.