AMU MATH302 Week 6 Study Guide: Hypothesis Testing

Use your current classroom sequence

Exact weekly order and assessment labels can differ by section or term. Use the syllabus in your current classroom as the controlling source. The numbered route addresses a common search pattern; its concept map is an editorial study sequence derived from current official scope and a clearly historical archived example, not a claim about your current module.

Topic map and distinctions

A statistical claim must be translated into a parameter statement. The null hypothesis contains the reference equality; the alternative represents the direction or difference the question asks about. Choose the significance level before inspecting the result. A test statistic measures how far the sample evidence lies from the null value in standard-error units under the chosen model. The p-value is the probability, assuming the null model and conditions, of evidence at least as incompatible with that model as observed.

Reject the null when the prespecified rule is met; otherwise fail to reject it. Failure to reject is not proof that the null is true. Identify one-sample versus two-sample structure, then decide whether two samples are independent or whether observations are naturally paired. Paired analysis works with within-pair differences; it is not chosen simply because group sizes happen to match. Conditions and assumptions remain part of the answer. Finally, separate statistical significance from practical importance by examining effect size, interval, units, and consequences.

Concept diagnostics and deeper checks

Build a test from the parameter outward. Write the population quantity, null value, alternative direction, significance level, and sample structure. Only then select the supported test statistic. This order prevents a software menu from defining the scientific question. The null is a reference model for calibration; the alternative is the evidentiary direction. Neither should be rewritten after inspecting a convenient sample pattern.

Pairing is a design property. Repeated measures on the same unit, twins matched by a defensible criterion, or before–after values create meaningful differences. Two independent samples with equal sizes remain independent. In paired work, summarize and test the difference variable; ignoring the link wastes information and can misstate uncertainty. In independent work, inventing pairs manufactures structure that the data do not possess.

A p-value must travel with assumptions and an effect estimate. A smaller value indicates greater incompatibility with the null model under the procedure, not a larger practical benefit or the probability of a hypothesis. Compare the estimated effect and confidence interval with a context threshold. A tiny effect can be statistically detectable in a huge sample, and an important effect can remain uncertain in a small sample. Decision language should preserve both dimensions.

Method and formula selection

The Hypothesis-Test Decision Tree asks three questions before calculation: what is the target, what data structure is present, and what assumption changes the method? Use the decision rows below as a compact selection table. Write the method choice in words before entering values so a function or familiar formula cannot conceal a mismatch.

Fully original fictional worked example

A fictional bicycle cooperative tests whether a new scheduling process changes mean repair time. Forty-two repairs under the old process and 38 separate repairs under the new process are sampled. Because the repair cases are different and not meaningfully matched, the structure is independent two-sample, not paired. The parameter is the difference between population mean repair times represented by the processes.

For a two-sided question, the null states a zero mean difference and the alternative states a nonzero difference. The analysis records the significance level, examines independence and distribution evidence, calculates the test statistic and p-value, and compares the p-value with the fixed threshold.

Even if the result is statistically significant, the cooperative should inspect the estimated difference and interval. A reduction of a few seconds may be operationally negligible; a meaningful reduction with a wide interval may require more evidence. If the samples came from different seasons, confounding could limit a process-effect claim.

Excel check without hiding the reasoning

Excel output should be read by label, not by the smallest decimal on screen. Identify which row or column corresponds to the specified tails and assumptions, confirm group order so the sign has the intended meaning, and record the estimate as well as p-value. Recreate the standard-error structure or compare the reported interval with the test decision. Current classroom directions determine the supported tool and procedure.

Original practice questions and reasoning checks

1. A claim says a population mean exceeds 50. State the directional alternative. Reasoning check: Hₐ: μ > 50; the null retains the equality boundary.

2. The same ten devices are measured before and after calibration. Independent or paired? Reasoning check: Paired, because each after value is linked to its own before value.

3. A p-value is 0.08 and α is 0.05. State the decision. Reasoning check: Fail to reject the null; do not claim the null is proved.

4. What does a small p-value measure? Reasoning check: Evidence incompatibility with the null model under stated assumptions, not the probability that the null is true.

5. A huge sample detects a 0.02-unit difference. What else is needed? Reasoning check: Effect magnitude, interval, units, costs, and context to judge practical importance.

Common wrong-answer patterns

Dangerous paths include placing the sample statistic in the hypotheses instead of a parameter, omitting equality from the null, choosing a one-sided direction after seeing results, treating equal sample sizes as pairing, reading p as the probability the null is true, writing “accept the null,” ignoring assumptions, reporting only the decision without effect magnitude, and equating statistical significance with importance.

How to check your own answer

Underline the claim and write the parameter in words. Fix null, alternative, and significance level before calculation. Draw the sample structure: one group, independent groups, or actual pairs. Check conditions and group order. Confirm the p-value tail matches the alternative. Compare the decision with a confidence interval when appropriate. End with a contextual sentence that distinguishes evidence strength from practical consequence.

Turn one solved example into reusable skill

Using the Hypothesis-Test Decision Tree, finish the fictional example, close the calculation, and reconstruct its decision path from memory. Write the target, data structure, method, assumptions, key substitution, output, and interpretation on separate lines. Then change one condition—such as dependence, sample size, variable type, tail direction, pairing, or distribution shape—and explain whether the same method survives. This contrast practice is more durable than memorizing the displayed numbers.

Create a two-column error log. In the first column, record the earliest decision that failed: translation, classification, model, formula, software input, arithmetic, or interpretation. In the second, write a future check that would catch it. Use the Hypothesis-Test Decision Tree as the organizing label, but express the check in your own words. Rework only original practice or material you are authorized to use; never build the log from uploaded restricted assessments.

Teach the Hypothesis-Test Decision Tree result aloud without looking at the page. A complete explanation names why the method fits, what the output means in context, and one claim the evidence cannot support. If you can compute but cannot explain those three parts, return to the topic map. If you can explain but cannot reproduce the arithmetic, return to the transparent setup and Excel audit. Mastery requires both paths to agree.

Build a miniature formula card only after the reasoning is stable. Put the trigger question above the relationship, define every symbol with units, list the assumptions beside it, and place one reasonableness check below it. On the reverse, write a situation where the relationship should not be used. Attach the card to the Hypothesis-Test Decision Tree rather than to a copied prompt. During review, cover the formula and recover it from the decision structure. This tests understanding while reducing the risk that a familiar-looking question activates the wrong procedure.

Short test-preparation checklist

Before beginning: identify the concept family and rewrite the target in your own words. During work: label inputs, units, distribution or parameter, and assumptions; keep enough precision to reproduce the result. Before finishing: use the Hypothesis-Test Decision Tree, perform an independent numerical or graphical check, and read the interpretation for scope and overclaiming. If current instructions use different notation or software, follow those instructions while preserving the same reasoning trail.

Related MATH302 resources

• MATH302 Statistics Help for AMU Students
• AMU MATH302 Week 5 Study Guide: Confidence Intervals
• AMU MATH302 Week 7 Study Guide: Correlation and Regression
• AMU MATH302 Final Exam Study Guide: Method Selection, Excel, and Answer Checks