Why alternating series lose marks on IB Math AA

The two questions an IB examiner is really asking

Whenever a series appears in Section B of IB Math AA Paper 2, the mark scheme is built around two layered questions, even when the stem only seems to ask one. The first is the convergence question: does the infinite sum tend to a finite limit? The second is the absoluteness question: does the series of absolute values also converge? These are not stylistic add-ons. A candidate who argues that a series converges by the alternating series test but ignores the absolute-value question is leaving one to two marks on the table, because the IB mark scheme awards the second method mark for any valid test applied to Σ|aₙ|, then a third mark for a clear conclusion that names 'absolute' or 'conditional'.

Think of the two-question structure as a sequence: first, prove convergence by any valid route. The alternating series test, the integral test, the comparison test, the ratio test, and the root test are all acceptable here, and an examiner will accept whichever you execute cleanly. Second, examine the absolute-value series Σ|aₙ| separately and run a fresh test on it. If that second test shows convergence, the original series is absolutely convergent, which is a stronger statement and one the IB mark scheme often rewards with a final 'hence' mark. If the absolute series diverges, but the original converged conditionally via the alternating series test or a comparison, the verdict is conditional convergence.

In practice, the question that the IB examiner is really asking is a discrimination question, not a calculation question. Two candidates can each correctly identify a convergent series, but only the candidate who correctly distinguishes 'converges conditionally' from 'converges absolutely' picks up the final reasoning mark. Examiners at IB marking sessions are trained to read for the word 'absolute' or 'conditional' in the concluding sentence. If a candidate writes 'the series converges because of the alternating series test' and stops, the second judgement is missing and the last method mark is not awarded. The phrasing you write at the end of a series question is, in the IB mark-scheme sense, half the answer.

A seven-test triage map for AA HL series questions

Most AA HL series items on Paper 2 are designed so that exactly one test is the cleanest path, and the examiner's job is to check whether you picked it. Walking into the question without a triage plan is one of the most common ways candidates lose marks on this topic, because they apply the ratio test to a series where the comparison test would have taken one line, or they default to the integral test on a non-monotonic function and waste four minutes. The triage below is the order I would walk a student through on the whiteboard.

Ratio test, first scan. If the general term aₙ contains factorials, n-th powers of constants, or products of consecutive integers, the ratio test is almost always the cleanest tool. Compute |aₙ₊₁ / aₙ|, simplify, and read the limit. A limit L < 1 means absolute convergence; L > 1 means divergence; L = 1 is inconclusive and forces a second test.

Root test, second scan. If aₙ involves an expression raised to the n-th power, especially with an n inside an nth root, the root test is faster than the ratio test because it skips the cancellation step. Compute lim nth-root(|aₙ|) and interpret it the same way as the ratio test.

Comparison test, third scan. If aₙ is a positive rational expression in n — a polynomial divided by a polynomial, or a square root, or 1/ln(n) — the comparison test against a p-series or geometric series is the workhorse. The IB mark scheme accepts direct comparison and limit comparison; the limit comparison is usually less work.

Alternating series test, only for sign-changers. If aₙ contains a factor of (-1)ⁿ or (-1)ⁿ⁺¹, the alternating series test becomes the primary tool for the original series. But the absolute-value test still has to be run separately, and that is where most candidates lose the final mark.

Integral test, only for monotonic positive terms. The integral test applies when aₙ = f(n) for a positive, continuous, decreasing function f. On Paper 2 it appears less often than the other four tests, but it is the only test that simultaneously gives convergence and, in the better-marked questions, the value of the sum.

p-series and geometric recognition. A quick shorthand: Σ 1/nᵖ converges iff p > 1. Σ a·rⁿ converges iff |r| < 1. Spotting these two patterns in disguise can save a candidate two to three minutes per item.

Divergence test as a triage exit. If lim aₙ ≠ 0, stop. The series diverges, and no further test is needed. This is the most underused time-saver in the AA HL toolkit.

When a candidate is on the floor in the exam room, the question is not 'which test do I know' but 'which test exits fastest.' For most AA HL Paper 2 items, the divergence test is checked first because it takes 15 seconds. If that fails, the test chosen is whichever one the form of aₙ matches most cleanly. Train the triage as a reflex, not as a decision tree you re-derive every time.

Absolute versus conditional: the alternating series trap

Conditional convergence is the most mis-marked concept in the AA HL series syllabus, and the reason is that the alternating series test is a 'convergence only' test. It does not address Σ|aₙ|. A candidate who applies the alternating series test to a series like Σ (-1)ⁿ / √n, concludes that it converges, and walks away has answered half the question. The second half is: what about Σ 1/√n? This is a p-series with p = 1/2, and since p ≤ 1, it diverges. So the original series is conditionally convergent — it converges, but not absolutely.

The IB mark scheme will typically allocate the marks as follows on a 6-mark conditional convergence item: one mark for stating the alternating series conditions (aₙ positive, decreasing, tending to zero); one method mark for verifying each condition; one mark for the conclusion 'converges by the alternating series test'; one method mark for setting up the absolute-value series; one method mark for applying a test to the absolute series (here, p-series with p = 1/2); and one final mark for the explicit conclusion 'the series converges conditionally.' That final conclusion is the one candidates most often forget, and it is the one mark the examiner cannot award without the word 'conditional' or its equivalent.

There is a deeper subtlety the AA HL paper occasionally tests. A series can converge absolutely, which is a strictly stronger condition than conditional convergence. A series that converges absolutely also converges conditionally in a trivial sense (because if Σ|aₙ| converges, the original converges), but the IB mark scheme distinguishes the two cases explicitly. A candidate who writes 'the series converges conditionally' when Σ|aₙ| also converges will lose the final reasoning mark. The defensive habit is to run the absolute-value test, then match the conclusion to the result: if Σ|aₙ| converges, write 'converges absolutely'; if Σ|aₙ| diverges but the original converges, write 'converges conditionally.'

Worked example: ratio test on a factorial series

Consider a Paper 2-style item: determine the convergence and absolute convergence of Σ n² / 2ⁿ from n = 1 to infinity. This is a series where the ratio test is the natural tool because of the 2ⁿ denominator and the polynomial numerator. The first scan: does lim aₙ exist and equal zero? Here aₙ = n² / 2ⁿ, and the limit is zero, so the divergence test does not help; the test is inconclusive, and we move to the ratio test.

Compute aₙ₊₁ / aₙ = ((n+1)² / 2ⁿ⁺¹) · (2ⁿ / n²) = ((n+1)² / n²) · (1/2) = (1 + 1/n)² · (1/2). The limit as n → ∞ is 1 · (1/2) = 1/2. Since 1/2 < 1, the ratio test gives absolute convergence. Notice the short-circuit: we did not need to test the original series separately, because the ratio test was applied to |aₙ| = aₙ (all terms are positive), so absolute convergence and convergence are decided in one go.

The IB mark-scheme language for this item: 'aₙ₊₁ / aₙ → 1/2 < 1, so by the ratio test the series converges absolutely.' The candidate should write the limit step, the inequality, and the conclusion in that order, with the word 'absolutely' explicitly stated. A candidate who writes 'converges' but omits 'absolutely' loses the final mark on a 6-mark question. The same shape applies to the root test, the comparison test, and the integral test, and examiners at IB marking sessions read for the same adjective every time.

Worked example: alternating series with a trick in the absolute tail

Now consider Σ (-1)ⁿ / (n + ln n) from n = 1. The first decision is whether to attempt the alternating series test, and the answer is yes because of the (-1)ⁿ factor. The conditions: aₙ = 1/(n + ln n) is positive, decreasing (eventually — for n ≥ 1, both n and ln n are increasing, so the denominator is increasing, hence aₙ is decreasing), and lim aₙ = 0. The alternating series test therefore gives convergence. So far, so straightforward. The trap is in the absolute-value series: Σ 1/(n + ln n). For large n, n + ln n behaves like n, so by limit comparison with the harmonic series Σ 1/n, which diverges, Σ 1/(n + ln n) also diverges. The conclusion is conditional convergence.

Two marking points to notice. First, the candidate must compute lim aₙ explicitly — examiners at the IB mark scheme often allocate a method mark for the limit calculation, and writing 'aₙ → 0 because n dominates ln n' is acceptable shorthand. Second, the limit comparison step must be written out, even if it feels obvious: compute lim (1/(n + ln n)) / (1/n) = lim n/(n + ln n) = 1, and since the harmonic series diverges and the limit is a positive finite constant, the absolute series diverges. The conclusion: 'the series converges conditionally.'

Comparison and limit comparison: getting the constant right

Comparison tests are the most error-prone of the seven tools on AA HL Paper 2, not because the logic is hard, but because the inequality direction has to be defended in writing. A candidate who writes '1/n² < 1/n, so Σ 1/n² converges by comparison' is making a true statement about the inequality but a true-but-useless statement for the conclusion, because the majorising series Σ 1/n diverges. The IB mark scheme requires the candidate to compare with a series whose convergence or divergence is already known.

Limit comparison is usually cleaner. Given a positive-term series Σ aₙ, pick a benchmark series Σ bₙ — usually a p-series or geometric — such that lim aₙ/bₙ = c, with 0 < c < ∞. If c is finite and positive, then Σ aₙ and Σ bₙ share the same convergence behaviour. The trap is the constant. If c = 0, the comparison is inconclusive. If c = ∞, also inconclusive. Only 0 < c < ∞ actually drives the conclusion. Candidates who skip the constant check lose the final reasoning mark.

A common AA HL Paper 2 item type is the form Σ (polynomial in n) / (polynomial in n, higher degree), where the limit comparison reduces to identifying the leading-term ratio. For example, Σ (3n² + 1) / (n³ + n) compares cleanly to Σ 1/n, since the leading-term ratio is lim (3n² / n³) / (1/n) = lim 3n²/n² = 3. With c = 3, and the harmonic series diverging, the conclusion is divergence. The candidate who spots the leading-term ratio and writes the limit comparison cleanly earns all four method marks on a typical 6-mark question.

Mark-scheme language and the IB-specific traps

The IB mark scheme is more prescriptive about language than most candidates realise. The mark scheme does not award the final 'hence' mark for vague phrasing like 'it converges' or 'the series is fine.' Three rules of thumb protect the final mark. First, name the test in the same sentence as the conclusion — 'by the ratio test, the series converges absolutely' is the gold-standard form. Second, write the conclusion as a separate sentence, not a clause. The examiner reads the final sentence as the answer, and a buried conclusion inside a run-on sentence can be missed in a fast-marking session. Third, mirror the IB syllabus terminology: 'absolute convergence,' 'conditional convergence,' 'diverges by the divergence test,' and 'p-series with p = 1/2' are the words that appear in the mark scheme.

There are also three IB-specific traps that catch AA HL candidates across cohorts. The first is the alternating harmonic-type series, where the absolute series diverges by p-test (p = 1) and the original converges by alternating series test. The second is the factorial ratio test, where a candidate simplifies |aₙ₊₁ / aₙ| incorrectly because they forget to track the absolute value inside the factorial — but factorials are non-negative, so the absolute value is often a no-op. The third is the integral test on a function that is not eventually decreasing; the IB mark scheme explicitly requires the function to be positive, continuous, and decreasing, and skipping one of these conditions forfeits a method mark.

Common pitfalls and how to avoid them

Candidates lose marks on AA HL series questions in five predictable patterns. The first is applying a test whose conditions are not met — running the integral test on a non-monotonic function, or the alternating series test on a series that does not alternate. The defensive habit is to state the three conditions of the test explicitly before applying it; the IB mark scheme awards a method mark for condition-checking, so writing the conditions is its own point. The second pattern is forgetting the absolute-value test entirely. The defensive habit is to ask, on every series question, 'have I run a test on Σ|aₙ| yet?' and to add it as a paragraph even when the ratio test has already implied absolute convergence.

The third pattern is misreading the ratio test limit. A limit of exactly 1 is inconclusive, and many candidates panic and conclude divergence or convergence. The IB mark scheme gives no marks for a wrong conclusion on L = 1, so the right move is to switch tests. The fourth pattern is arithmetic error in the ratio test, usually from mishandling the n-th term of a polynomial expression. A candidate who computes (n+1)² / n² as 1 + 1/n² rather than 1 + 2/n + 1/n² is in trouble, and the defensive habit is to expand binomial coefficients explicitly. The fifth pattern is time. Series items in Section B of Paper 2 are typically worth 6 to 8 marks and should take 8 to 12 minutes. A candidate who spends 18 minutes on a single series item has spent the budget of two items, and the IB mark scheme does not award extra credit for thoroughness when other questions are left blank.

The single best defensive habit is the closing sentence. Write the conclusion in a complete sentence, name the test, name the verdict (absolute, conditional, divergent), and stop. Then re-read the question stem to make sure the conclusion answers what was actually asked. Candidates who build this habit typically gain 1 to 2 marks per Paper 2 attempt relative to peers of equal ability who do not.

How this topic fits into a Paper 2 preparation plan

Series and convergence sit in Section B of the AA HL Paper 2, alongside differential equations, probability, and calculus options. In most cohorts, two to three of the six Section B questions are calculus-heavy, and at least one of them is a series or convergence item. The preparation plan that works for most AA HL candidates is to drill seven tests, time-boxed, and then to do past-paper series items under timed conditions. A good target is 8 minutes per series item on a first attempt and 6 minutes on a re-attempt. A candidate who cannot break 10 minutes per item is over-writing, usually by stating too many of the test conditions or re-deriving a test from scratch on the floor of the exam room.

The IB scoring system in the AA HL course awards a final grade on a 1–7 scale, and Section B carries roughly 40% of the marks on Paper 2. A candidate who reliably scores 6 to 7 marks on every series item is converting a difficult topic into a guaranteed 6 to 7 points per paper, which is roughly the difference between a band 5 and a band 6 final grade. The preparation strategy that compounds fastest is past-paper drilling on series items, followed by error-log review of the two or three mistakes per paper that came from test-mis-selection rather than arithmetic. Most candidates who plateau at a band 5 have the arithmetic; what they lack is the triage reflex.

Conclusion and next steps

Absolute and conditional convergence is one of the few AA HL topics where the difference between a band 5 and a band 6 final score comes down to a small number of well-defined habits: choose the right test, run the absolute-value test separately, name the verdict explicitly, and time-box the item to 8 minutes. A candidate who builds these four habits into a Paper 2 preparation plan will treat Section B series items as a guaranteed point-scoring zone rather than a minefield. The natural next step is a focused drill on the seven-test triage, with a stopwatch and a stack of AA HL past-paper series items, then a structured error log of the test-mis-selection patterns. TestPrep Europe's diagnostic assessment on AA HL series and convergence is a productive starting point for candidates building a sharper preparation plan around this specific question type.

Frequently asked questions

What is the difference between absolute and conditional convergence in IB Math AA HL?

A series converges absolutely when the sum of the absolute values of its terms Σ|aₙ| also converges. A series converges conditionally when the original series converges but the absolute-value series Σ|aₙ| diverges. The IB mark scheme distinguishes the two cases explicitly, and the final reasoning mark is awarded only when the candidate writes the correct adjective — 'absolute' or 'conditional' — in the conclusion.

Why does the alternating series test not prove absolute convergence?

The alternating series test is a 'convergence only' tool. It certifies that a sign-changing series converges when the absolute values aₙ are positive, decreasing, and tend to zero, but it says nothing about Σ|aₙ|. To establish absolute convergence, a separate test — usually the ratio test, comparison test, or p-series recognition — must be run on the absolute-value series.

Which convergence test should I use first on AA HL Paper 2?

Scan the form of aₙ. If aₙ contains factorials or n-th powers, try the ratio test. If aₙ is an expression raised to the n-th power, try the root test. If aₙ is positive and rational, try comparison or limit comparison. If aₙ contains a (-1)ⁿ factor, use the alternating series test on the original and a separate test on the absolute-value series. Run the divergence test first as a 15-second exit.

What is the L = 1 case for the ratio test, and how should I handle it?

When lim |aₙ₊₁ / aₙ| = 1, the ratio test is inconclusive. The series may converge or diverge, and the test alone does not decide. The correct response on AA HL Paper 2 is to switch tests: try the comparison test, the integral test, or the root test, depending on the form of aₙ. The IB mark scheme does not award a conclusion mark when the ratio test gives L = 1 and the candidate guesses.

How many minutes should I spend on a series question in Paper 2?

A typical AA HL series item in Section B is worth 6 to 8 marks and should take 8 to 12 minutes on a first attempt. A candidate who cannot break 10 minutes per item is usually over-writing, often by re-deriving test conditions or stating too many justifications. A timed drill of past-paper series items, with a 10-minute cap per question, is the fastest way to build the pacing reflex.