Sampling distributions: Name the objects and notation before applying a rule.
Point estimation
Point estimation: Connect the symbolic form to a graph, diagram, table, or example.
Interval estimation
Interval estimation: Compare two methods and record when each one is valid.
Hypothesis tests
Hypothesis tests: Finish with an unfamiliar problem and an independent check.
Likelihood
Likelihood: Name the objects and notation before applying a rule.
Regression
Regression: Connect the symbolic form to a graph, diagram, table, or example.
Bayesian foundations
Bayesian foundations: Compare two methods and record when each one is valid.
Eight original decision checks
These fixed prompts test study and verification choices across the course map. The interactive builder above creates mathematical questions separately.
Question 1: For a sample of 100 with a true proportion of 0.4, what is the standard error of the sample proportion, to three decimal places?
Sampling distributions
0.049
0.240
0.004
0.400
Inspect the reasoning
The standard error of a proportion is the square root of p(1 - p) / n.
p(1 - p) = 0.4 x 0.6 = 0.24.
0.24 / 100 = 0.0024.
The square root of 0.0024 is 0.049.
Best choice: 0.049
Question 2: An estimator has variance 9 and bias 2. What is its mean squared error?
Point estimation
13
11
18
9
Inspect the reasoning
Mean squared error equals variance plus the square of the bias.
The bias squared is 2^2 = 4.
9 + 4 = 13.
An unbiased estimator would have MSE equal to its variance, 9.
Best choice: 13
Question 3: A sample of 100 has a mean of 120 and a known standard deviation of 15. What is the 99% confidence interval, to two decimal places?
Interval estimation
(116.14, 123.86)
(117.06, 122.94)
(105.00, 135.00)
(118.50, 121.50)
Inspect the reasoning
The standard error is 15 divided by the square root of 100, which is 1.5.
For 99% confidence, z = 2.576.
The margin of error is 2.576 x 1.5 = 3.864.
120 - 3.864 = 116.14 and 120 + 3.864 = 123.86.
Best choice: (116.14, 123.86)
Question 4: Test H0: mu = 50 against a two-sided alternative with sample mean 52, sigma = 8, and n = 64. What is the z statistic?
Hypothesis tests
2.0
0.25
16.0
1.5
Inspect the reasoning
The z statistic is (sample mean minus the claimed mean) divided by the standard error.
The standard error is 8 divided by the square root of 64, which is 8 / 8 = 1.
The difference is 52 - 50 = 2.
2 / 1 = 2.0, which gives a two-sided p-value of about 0.046.
Best choice: 2.0
Question 5: Twenty independent Bernoulli trials give 7 successes. What is the maximum likelihood estimate of p?
Likelihood
0.35
0.50
0.29
0.65
Inspect the reasoning
The likelihood is p^7 times (1 - p)^13.
Taking logs gives 7 ln p + 13 ln(1 - p).
Setting the derivative to zero gives 7 / p = 13 / (1 - p), so 7 - 7p = 13p and 20p = 7.
p = 7 / 20 = 0.35, which is just the sample proportion.
Best choice: 0.35
Question 6: A regression has Sxy = 84, Sxx = 28, mean x = 4, and mean y = 20. What are the slope and intercept?
Regression
slope 3, intercept 8
slope 3, intercept 20
slope 0.33, intercept 18.7
slope 2, intercept 12
Inspect the reasoning
The least-squares slope is Sxy divided by Sxx.
84 / 28 = 3.
The line passes through the point of means, so intercept = mean y minus slope times mean x.
20 - 3(4) = 20 - 12 = 8.
Best choice: slope 3, intercept 8
Question 7: A Beta(2, 3) prior is updated with 6 successes and 4 failures. What is the posterior mean, to four decimal places?
Bayesian foundations
0.5333
0.6000
0.4000
0.5000
Inspect the reasoning
The Beta prior is conjugate: add successes to the first parameter and failures to the second.
The posterior is Beta(2 + 6, 3 + 4) = Beta(8, 7).
The mean of Beta(a, b) is a / (a + b).
8 / 15 = 0.5333.
Best choice: 0.5333
Question 8: A sample of 16 has standard deviation 4. What is the estimated standard error of the mean, and how many degrees of freedom does the t procedure use?
Sampling distributions
standard error 1, 15 degrees of freedom
standard error 1, 16 degrees of freedom
standard error 4, 15 degrees of freedom
standard error 0.25, 15 degrees of freedom
Inspect the reasoning
The estimated standard error is s divided by the square root of n.
The square root of 16 is 4, so 4 / 4 = 1.
A one-sample t procedure uses n - 1 degrees of freedom.
16 - 1 = 15.
Best choice: standard error 1, 15 degrees of freedom
After the run
Sort each miss into interpretation, recall, setup, execution, verification, or pacing. Schedule a changed question that tests the corrected decision without reusing the same surface details.
When should I increase difficulty?
Increase it after setup is accurate without a clue and you can explain why the method fits.
When should I use test mode?
Use it once guided sets are stable and you want evidence about selection, pacing, and sustained attention.