Multi-Focus on Practices 2, 3, and 4

Connect an experiment to a two-proportion interval: FRQ

3 versions of one AP Statistics free-response question, each with a 10-point rubric and a model answer for every point. Write your response on paper before you open the rubric.

These are StatsLearn originals written in the AP style, not released exam questions. Topics: 1.13 Experimental Design, 3.10 Constructing a Confidence Interval for the Difference Between Two Population Proportions, 3.11 Justifying a Claim Based on a Confidence Interval for the Difference Between Two Population Proportions.

Version 1

A health program recruits 290 volunteers and randomly assigns 140 to text reminders and 150 to email reminders. In the text group, 87 attend a follow-up visit. In the email group, 72 attend. Use z* = 1.96.

  1. (a) Explain what random assignment permits and what volunteer recruitment does not permit.
  2. (b) Calculate both sample proportions and check the large-count condition.
  3. (c) Construct the confidence interval for the difference in population proportions.
  4. (d) Interpret what the interval supports in this study.
Show the rubric and model answers

Part (a): 2 points

  • Explains the purpose of random assignment. Random assignment uses chance to distribute other variables between reminder groups, supporting a causal comparison for these volunteers.
  • States the scope of inference correctly. The design supports causation for the volunteers, but volunteer recruitment does not support generalization to a wider population.

Part (b): 3 points

  • Computes the text-group proportion. p-hat1 = 87/140 = 0.6214.
  • Computes the email-group proportion. p-hat2 = 72/150 = 0.4800.
  • Checks the large-count condition for both groups. The four counts 87, 53, 72, and 78 are each at least 10.

Part (c): 4 points

  • Uses the observed difference and computes the unpooled standard error. The interval is centered at 0.1414, and SE = 0.0578.
  • Computes the margin of error. ME = 1.96(0.0578) = 0.1133.
  • Computes the lower endpoint. Lower endpoint = 0.0281.
  • Computes the upper endpoint. Upper endpoint = 0.2548.

Part (d): 1 point

  • Uses the interval to make the supported conclusion. Because the entire interval is above zero, the data support a higher follow-up rate for text reminders among these volunteers.

Version 2

A health program recruits 370 volunteers and randomly assigns 180 to text reminders and 190 to email reminders. In the text group, 115 attend a follow-up visit. In the email group, 94 attend. Use z* = 1.96.

  1. (a) Explain what random assignment permits and what volunteer recruitment does not permit.
  2. (b) Calculate both sample proportions and check the large-count condition.
  3. (c) Construct the confidence interval for the difference in population proportions.
  4. (d) Interpret what the interval supports in this study.
Show the rubric and model answers

Part (a): 2 points

  • Explains the purpose of random assignment. Random assignment uses chance to distribute other variables between reminder groups, supporting a causal comparison for these volunteers.
  • States the scope of inference correctly. The design supports causation for the volunteers, but volunteer recruitment does not support generalization to a wider population.

Part (b): 3 points

  • Computes the text-group proportion. p-hat1 = 115/180 = 0.6389.
  • Computes the email-group proportion. p-hat2 = 94/190 = 0.4947.
  • Checks the large-count condition for both groups. The four counts 115, 65, 94, and 96 are each at least 10.

Part (c): 4 points

  • Uses the observed difference and computes the unpooled standard error. The interval is centered at 0.1442, and SE = 0.0510.
  • Computes the margin of error. ME = 1.96(0.0510) = 0.0999.
  • Computes the lower endpoint. Lower endpoint = 0.0443.
  • Computes the upper endpoint. Upper endpoint = 0.2440.

Part (d): 1 point

  • Uses the interval to make the supported conclusion. Because the entire interval is above zero, the data support a higher follow-up rate for text reminders among these volunteers.

Version 3

A health program recruits 450 volunteers and randomly assigns 220 to text reminders and 230 to email reminders. In the text group, 145 attend a follow-up visit. In the email group, 118 attend. Use z* = 1.96.

  1. (a) Explain what random assignment permits and what volunteer recruitment does not permit.
  2. (b) Calculate both sample proportions and check the large-count condition.
  3. (c) Construct the confidence interval for the difference in population proportions.
  4. (d) Interpret what the interval supports in this study.
Show the rubric and model answers

Part (a): 2 points

  • Explains the purpose of random assignment. Random assignment uses chance to distribute other variables between reminder groups, supporting a causal comparison for these volunteers.
  • States the scope of inference correctly. The design supports causation for the volunteers, but volunteer recruitment does not support generalization to a wider population.

Part (b): 3 points

  • Computes the text-group proportion. p-hat1 = 145/220 = 0.6591.
  • Computes the email-group proportion. p-hat2 = 118/230 = 0.5130.
  • Checks the large-count condition for both groups. The four counts 145, 75, 118, and 112 are each at least 10.

Part (c): 4 points

  • Uses the observed difference and computes the unpooled standard error. The interval is centered at 0.1460, and SE = 0.0459.
  • Computes the margin of error. ME = 1.96(0.0459) = 0.0900.
  • Computes the lower endpoint. Lower endpoint = 0.0561.
  • Computes the upper endpoint. Upper endpoint = 0.2360.

Part (d): 1 point

  • Uses the interval to make the supported conclusion. Because the entire interval is above zero, the data support a higher follow-up rate for text reminders among these volunteers.

Keep going