Multi-Focus on Practices 1 and 2

Build and evaluate a probability simulation: FRQ practice

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: 2.3 Estimating Probabilities Using Simulation, 2.4 Introduction to Probability, 2.7 Independent Events and Unions of Events.

Version 1

A campus network reports that each independently monitored device has probability 0.08 of needing a restart during a maintenance window. A team monitors 4 devices.

  1. (a) State the probability question and define one simulation trial.
  2. (b) Describe a complete simulation using random numbers.
  3. (c) The simulation produced 118 successful trials out of 400. Compute the estimate, compare it with the exact probability, and discuss variability.
Show the rubric and model answers

Part (a): 2 points

  • States the event being estimated. Estimate the probability that at least one of the 4 monitored devices needs a restart.
  • Defines one trial as a full set of devices. One trial simulates all 4 devices and records whether at least one restart occurs.

Part (b): 4 points

  • Uses a chance device with the stated restart probability. For each device, generate a random number from 000 to 999 and label the first 80 values as a restart.
  • Generates independent outcomes for all devices. Generate 4 separate random numbers in each trial so one device's outcome does not set another's.
  • Defines a trial success correctly. Count a trial as a success when one or more generated device outcomes are restarts.
  • Repeats the simulation many times. Repeat the full trial 400 times and divide the number of successful trials by 400.

Part (c): 4 points

  • Computes the exact complement probability. P(at least one) = 1 - (0.92)^4 = 0.2836.
  • Computes the simulation estimate. 118 successes in 400 trials gives 0.2950.
  • Compares the estimate with the exact value. The estimate 0.2950 is close to the exact value 0.2836; the gap is plausible simulation variation.
  • Explains how more trials affect variability. A larger number of trials makes the simulation estimate less variable across repeated runs, though it does not force an exact match.

Version 2

A campus network reports that each independently monitored device has probability 0.12 of needing a restart during a maintenance window. A team monitors 4 devices.

  1. (a) State the probability question and define one simulation trial.
  2. (b) Describe a complete simulation using random numbers.
  3. (c) The simulation produced 247 successful trials out of 600. Compute the estimate, compare it with the exact probability, and discuss variability.
Show the rubric and model answers

Part (a): 2 points

  • States the event being estimated. Estimate the probability that at least one of the 4 monitored devices needs a restart.
  • Defines one trial as a full set of devices. One trial simulates all 4 devices and records whether at least one restart occurs.

Part (b): 4 points

  • Uses a chance device with the stated restart probability. For each device, generate a random number from 000 to 999 and label the first 120 values as a restart.
  • Generates independent outcomes for all devices. Generate 4 separate random numbers in each trial so one device's outcome does not set another's.
  • Defines a trial success correctly. Count a trial as a success when one or more generated device outcomes are restarts.
  • Repeats the simulation many times. Repeat the full trial 600 times and divide the number of successful trials by 600.

Part (c): 4 points

  • Computes the exact complement probability. P(at least one) = 1 - (0.88)^4 = 0.4003.
  • Computes the simulation estimate. 247 successes in 600 trials gives 0.4117.
  • Compares the estimate with the exact value. The estimate 0.4117 is close to the exact value 0.4003; the gap is plausible simulation variation.
  • Explains how more trials affect variability. A larger number of trials makes the simulation estimate less variable across repeated runs, though it does not force an exact match.

Version 3

A campus network reports that each independently monitored device has probability 0.16 of needing a restart during a maintenance window. A team monitors 4 devices.

  1. (a) State the probability question and define one simulation trial.
  2. (b) Describe a complete simulation using random numbers.
  3. (c) The simulation produced 411 successful trials out of 800. Compute the estimate, compare it with the exact probability, and discuss variability.
Show the rubric and model answers

Part (a): 2 points

  • States the event being estimated. Estimate the probability that at least one of the 4 monitored devices needs a restart.
  • Defines one trial as a full set of devices. One trial simulates all 4 devices and records whether at least one restart occurs.

Part (b): 4 points

  • Uses a chance device with the stated restart probability. For each device, generate a random number from 000 to 999 and label the first 160 values as a restart.
  • Generates independent outcomes for all devices. Generate 4 separate random numbers in each trial so one device's outcome does not set another's.
  • Defines a trial success correctly. Count a trial as a success when one or more generated device outcomes are restarts.
  • Repeats the simulation many times. Repeat the full trial 800 times and divide the number of successful trials by 800.

Part (c): 4 points

  • Computes the exact complement probability. P(at least one) = 1 - (0.84)^4 = 0.5021.
  • Computes the simulation estimate. 411 successes in 800 trials gives 0.5138.
  • Compares the estimate with the exact value. The estimate 0.5138 is close to the exact value 0.5021; the gap is plausible simulation variation.
  • Explains how more trials affect variability. A larger number of trials makes the simulation estimate less variable across repeated runs, though it does not force an exact match.

Keep going