From a single four-cell table, the PU520 Unit 3 exercise example derives relative risk, an odds ratio and a risk difference, then states which measure the cohort design permits reporting. Searches like "pu 520 unit 3 assignment example", "pu520 unit 3 sample" and "pu520 unit 3 example" land here.
What a finished PU520 Unit 3 association measures exercise looks like
The grid sits at the top of the first page, exposure in the rows and outcome in the columns, with margins totaled: 56 of 800 near the corridor developed hypertension, against 42 of 1,200 farther away. Beneath it, labeled parts work through each measure. Risk in the exposed is 7.0 per 100 and in the unexposed 3.5, giving a relative risk of 2.0 with a 95 percent confidence interval of 1.35 to 2.95. The odds ratio, 2.08, follows from the cross-product and sits close to the relative risk because the outcome is uncommon. A risk difference of 3.5 per 100 and an attributable fraction among the exposed of 50 percent come next. The final part asks what would change if the same four numbers had come from a case-control study, and answers it.
How a PU520 Unit 3 example is structured
The exercise is ordered so that each measure is introduced by the question it answers. Relative risk comes first because the data are from a cohort that followed people forward, which makes incidence in each group directly measurable. The odds ratio follows, and the example explains in two sentences why it runs slightly higher than the relative risk and why the gap would widen if hypertension were common. The confidence interval is calculated on the log scale and back-transformed, and the working appears so the width can be checked. Absolute measures arrive after the ratios, since a public health reader needs to know how many cases the exposure adds, not only how much it multiplies them. The case-control question closes the exercise and carries the unit's main idea: the design, not the arithmetic, decides which measure may be reported.
Grid layout stated first
Rows hold exposure and columns hold outcome, with cells labeled a through d. The convention is stated because every formula in the exercise depends on it, and a transposed grid is a common silent error.
Relative risk with its interval
Risk in each row, the ratio, then the log-scale standard error and a 95 percent interval from 1.35 to 2.95. Because the interval stays above 1.0, the example says sampling variation alone is an unlikely account of the association.
Odds ratio beside it
The cross-product, ad over bc, gives 2.08. One paragraph explains the rare outcome approximation and shows how the two measures would separate if a third of each group had become cases.
What the exposure adds
The risk difference of 3.5 per 100 over five years is restated as roughly 28 extra diagnoses among the 800 exposed residents, a number a housing or transportation planner can weigh.
Same numbers, different design
Had cases and controls been sampled, only the odds ratio could be reported, since the investigator would have fixed how many cases entered the study. The exercise ends on that point.
Where marks go in PU520 Unit 3
Graders on association exercises usually split credit between calculation and interpretation, and both are easy to lose. Calculation losses cluster around the grid: cells transposed, a relative risk computed as a ratio of odds, or an interval built on the raw scale instead of the log. Interpretation losses are subtler. A relative risk read as though it were the risk difference, so that 2.0 becomes two percent more, is a misreading many sections mark outright, and calling the corridor a cause from one cohort draws a comment. Papers also drop credit when they report only ratios and leave out the absolute difference, or when an odds ratio from a case-control design is presented as a relative risk. The example's closing design question answers that last criterion.
Get a PU520 Unit 3 example written to your instructions
For the PU520 Unit 3 exercise, the desk needs the grid or dataset your section supplied, the questions in their original order and the rubric. The first custom sample costs nothing, arrives in 24-48 hours, and reads each measure off your own four cells, with the design question answered for whichever study type produced your data.
PU520 Unit 3 questions, answered
When should the exercise report an odds ratio instead of a relative risk?
When the design fixed the number of cases, as a case-control study does, because incidence cannot be calculated from data sampled on the outcome. In a cohort or a cross-sectional survey both measures are available, and the relative risk is usually easier for readers to grasp. The example reports both from its cohort grid and says which one leads.
Why is the confidence interval built on the log scale?
Ratio measures are skewed: they cannot fall below zero but can rise without limit. Taking the natural log makes their sampling distribution roughly symmetric, so the familiar plus-or-minus calculation works there and the endpoints are converted back afterward. That is why the interval of 1.35 to 2.95 sits unevenly around 2.0, and why many graders check for it.
Does a relative risk of 2.0 prove the freight corridor raises blood pressure?
No. It shows an association in one composite cohort. Traffic noise, air pollution, household income and housing age could all travel with distance from the corridor, and the exercise names them as candidates. Later PU520 units typically return to exactly this problem of rival explanations, so a Unit 3 answer that flags it without resolving it is doing what the course asks.