Nearly every set of Hardy-Weinberg practice problems is the same three problems in different costumes, and the reason students stall is almost never the algebra. It is that the problem hands you a number and you cannot tell which letter it is.
Hardy-Weinberg practice problems come in three shapes: phenotype counts, the frequency of the recessive phenotype, or an allele frequency given outright. All three run on p + q = 1 and p^2 + 2pq + q^2 = 1, where p and q are allele frequencies and the squared terms are genotype frequencies. Identify which one you were handed; the rest is substitution.
Below, one problem of each shape, carried all the way to a sentence in English. Every line of arithmetic is shown, because the line you skip is the line you get wrong.
What does the Hardy-Weinberg equation assume?
Before any number goes in, the assumptions. The OpenStax population-genetics chapters set out the conditions under which allele frequencies stay constant from one generation to the next: no mutation, no migration into or out of the population, a very large population size, random mating, and no natural selection acting on the trait.
That is a demanding list, and no real population meets all of it exactly. This matters more than it sounds. The equation does not describe an arbitrary real population. It produces the genotype frequencies you would expect if nothing on that list were happening, which is precisely what makes it useful: a measured population that departs from the expectation is telling you that something is happening, and the size of the departure is a separate question you answer with a statistical test rather than by eye. The Hardy-Weinberg equation page sets out the terms; the worked chi-square example is what you reach for when you want to ask whether a departure is bigger than chance would comfortably produce.
So everything below is an expectation. None of it is a description of a real plot of ground.
Problem 1: you are given phenotype counts
In a population of 1,000 flowering plants, red flowers (R) are dominant to white flowers (r). 160 plants have white flowers. Assuming the population is at Hardy-Weinberg equilibrium, how many of the red-flowered plants are heterozygous?
Start by naming what you were given. White is the recessive phenotype, so every white plant is rr, and rr is the genotype whose expected frequency is q^2. That is the entry point, and it is the only one of the three genotypes you can read straight off a phenotype count.
q^2 = 160 / 1,000 = 0.16
q = the square root of 0.16 = 0.4
This is the step that gets skipped. The frequency of the r allele is 0.4, not 0.16. The 0.16 is the frequency of a genotype; the 0.4 is the frequency of an allele. Losing that distinction is the single most common error in these problems, because the wrong answer still looks tidy.
p = 1 - 0.4 = 0.6
Now the genotype frequencies:
p^2 = (0.6)(0.6) = 0.36
2pq = 2 x 0.6 x 0.4 = 0.48
q^2 = 0.16
Check that they sum: 0.36 + 0.48 + 0.16 = 1.00.
Back to counts, since the question asked how many plants:
RR: 0.36 x 1,000 = 360 plants
Rr: 0.48 x 1,000 = 480 plants
rr: 0.16 x 1,000 = 160 plants
And 360 + 480 + 160 = 1,000, which is the check worth doing every time.
The answer is 480 heterozygous red-flowered plants. Said in words: of the 840 plants showing red flowers, 480 of them carry a white allele they are not showing, which is a little over half.
You can verify the allele frequency a second way, without the square root, by counting alleles directly. Each plant is diploid, so 1,000 plants carry 2,000 alleles. The r alleles are two from every rr plant and one from every Rr plant: 2 x 160 + 480 = 320 + 480 = 800. And 800 / 2,000 = 0.4. Same number, arrived at from the other side. If your two routes disagree, the square root is where to look.
Problem 2: you are given how often the recessive phenotype appears
In a large beetle population at Hardy-Weinberg equilibrium, 1 in 400 beetles shows a recessive shell pattern. What proportion of the population carries the recessive allele without showing the pattern?
Same entry point, different wrapping. “Shows the recessive pattern” is the homozygous recessive genotype, so:
q^2 = 1 / 400 = 0.0025
q = the square root of 0.0025 = 0.05
p = 1 - 0.05 = 0.95
The question asks for carriers, which is the heterozygote, which is 2pq:
2pq = 2 x 0.95 x 0.05 = 0.095
p^2 = (0.95)(0.95) = 0.9025
Check: 0.9025 + 0.095 + 0.0025 = 1.0000.
So the expected proportion of carriers is 0.095, or 95 beetles in every 1,000.
The sentence that makes this worth calculating: carriers are expected to be far more common than beetles showing the pattern. Divide one by the other, 0.095 / 0.0025 = 38, so for every beetle showing the recessive pattern there are expected to be about 38 carrying the allele invisibly. A recessive allele can be far more common in the population than its phenotype suggests, and here the allele frequency is 5 percent while the phenotype shows up in a quarter of one percent. That gap is most of what population genetics is for. It also says nothing about any individual beetle, only about what the population is expected to look like in aggregate.
Problem 3: you are given an allele frequency directly
In a population of 500 organisms, the frequency of allele A is 0.7. Assuming Hardy-Weinberg equilibrium, how many individuals of each genotype are expected?
This is the easiest shape and the one people overthink, because there is no square root anywhere. You were handed p.
p = 0.7, so q = 1 - 0.7 = 0.3
p^2 = (0.7)(0.7) = 0.49
2pq = 2 x 0.7 x 0.3 = 0.42
q^2 = (0.3)(0.3) = 0.09
Check: 0.49 + 0.42 + 0.09 = 1.00.
Convert to counts now, at the end, rather than partway through:
AA: 0.49 x 500 = 245
Aa: 0.42 x 500 = 210
aa: 0.09 x 500 = 45
And 245 + 210 + 45 = 500.
These are expected counts, not a prediction about a particular sample of 500 organisms. If you went out and counted and found 230, 220 and 50, you have not refuted anything yet. You have an observation that differs from an expectation, and whether the difference is larger than chance would comfortably produce is a question for a statistical test.
Which shape is this problem?
A short diagnostic that covers most of what you will be handed:
Given a count or proportion of individuals showing the recessive phenotype, that is q^2, and your first move is a square root.
Given a count of individuals showing the dominant phenotype, that is p^2 plus 2pq, so subtract from the total to get the recessive count and you are back in the first case.
Given a frequency described as an allele frequency, that is p or q, and there is no square root.
Given genotype counts for all three classes, you do not need the equation at all to get the allele frequencies. Count alleles the way problem 1 checked itself. Then use the equation to produce expected counts and compare.
The distinction underneath all four lines is the one between a genotype and a phenotype, which is why the genotype and phenotype comparison is worth rereading if the entry point keeps feeling arbitrary. Pedigree charts are the other place that gap trips people up, and pedigree practice problems works three of them in the same order.
When you are working your own assigned numbers rather than these, the useful thing is not the final value but the line where your version and the correct version diverge. Cyllo will take the problem as typed or photographed and carry it line by line in this same shape, showing each substitution so you can find the step where yours went sideways, and stay in the conversation while you ask what changes if the population is not at equilibrium. Photo solving and AI chat run on a subscription; the topic quizzes, the flashcards and the biology reference do not.
One last thing about the word “expected”
Every number in this post is an expectation generated by a model with five assumptions attached. The model is not a claim that the population is at equilibrium. It is the baseline against which you notice that it is not. Free-response questions lean on this hard, which is why the section on how an AP Biology free-response question is built spends so long on the difference between describing a value and justifying a claim about it.
AP is a trademark registered by the College Board, which is not affiliated with, and does not endorse, this site.