ACT Score Calculator

How an ACT Test Score Calculator Works

An ACT test score calculator never sees how many questions you answered correctly. By the time you hold a score report, each section's raw count has already been converted to a scaled score from $1$ to $36$ through equating, a psychometric process that adjusts for the fact that one test form may be slightly harder than another. The same raw count can therefore map to different scaled scores on different test dates.

This is why the calculator works only with scaled scores. Entering something like $65/75$ and expecting a scaled result would be guesswork, because no single raw-to-scale curve exists. Look for the scaled number on your score report, or on the answer-key documentation that comes with an official practice form.

The calculator runs as a deterministic vanilla ES module (js/math.js) entirely in your browser. No section score is sent anywhere, there are no external API calls or tracking, and every result updates reactively as you type. Scaled scores and composites use integer arithmetic. Outputs depend only on the inputs and the selected rules.

Current and Legacy Scoring Rules

The scoring setting selects which sections feed the composite.

Official past score reports are not retroactively modified by ACT, Inc. A legacy composite on your older report remains exactly what it was. The legacy setting exists so students and counselors can compare the two policies without touching any historical record, for instance to see how much a strong or weak Science score moved an old composite.

ACT Composite Score Calculator: Formula and Rounding

The composite is the arithmetic mean of the active section scores, rounded to the nearest whole number. With $N$ active sections and section scores $s_1, \dots, s_N$:

The mean is displayed to three decimal places for transparency. The reported composite, however, is not computed from the displayed decimal. Binary floating-point arithmetic can misrepresent exact halves, so the engine rounds half-up using integers only:

Because the numerator adds exactly half of the denominator before the floor, any mean with a fractional part of $0.5$ or more rounds up, and exact halves ($x.5$) always round strictly upward. A composite of $24.5$ becomes $25$, never $24$.

Worked examples

Current rule, no rounding needed. English $28$, Math $26$, Reading $30$ gives $\text{Total} = 84$ and $\text{Mean} = 84/3 = 28.000$. The composite is $\lfloor (168 + 3)/6 \rfloor = \lfloor 28.5 \rfloor = 28$.

Legacy rule, exact half. English $24$, Math $24$, Reading $25$, Science $25$ gives $\text{Total} = 98$ and $\text{Mean} = 98/4 = 24.500$. The composite is $\lfloor (196 + 4)/8 \rfloor = \lfloor 25 \rfloor = 25$.

Current rule, repeating decimal. If the mean is $29.333$, the integer form still gives the right answer: $\lfloor (176 + 3)/6 \rfloor = \lfloor 29.833 \rfloor = 29$, because the fractional part $0.333$ is below one half.

The same section scores can produce different composites under the two rules. English $30$, Math $30$, Reading $30$, Science $22$ yields $30$ under current ($\text{Total} = 90$, $N = 3$) but $28$ under legacy ($\text{Total} = 112$, $\text{Mean} = 28.000$, $N = 4$).

Superscore Mechanics Across Attempts

The superscore operation takes the maximum scaled score for each active section across $2$, $3$, or $4$ completed attempts, sums those maxima, and applies the same half-up composite formula. For section $j$ across $A$ attempts:

Superscoring does not average attempts. It selects per section, so the result can exceed every single-sitting composite. If two attempts tie on a section, the earliest attempt is credited, which affects only the attempt badge shown in the visual, never the arithmetic.

Worked superscore matrix

Three attempts under the current rule:

$\text{Total} = 29 + 30 + 28 = 87$, $\text{Mean} = 29.000$, and the superscore composite is $\lfloor (174 + 3)/6 \rfloor = 29$. The single-sitting composites are $27$ (attempt 1: $81/3$), $28$ (attempt 2: $84/3$), and $28$ (attempt 3: $85/3 = 28.333$).

The built-in two-attempt preset shows the same effect. Attempt 1 is English $28$, Math $26$, Reading $30$; attempt 2 is English $30$, Math $28$, Reading $27$. The best sections are English $30$ (attempt 2), Math $28$ (attempt 2), and Reading $30$ (attempt 1). $\text{Total} = 88$, $\text{Mean} = 29.333$, composite $29$, compared with $28$ and $28$ for the individual sittings ($84/3$ and $85/3$).

In legacy mode, Science participates in the maxima and $N = 4$. Attempts 2 to 4 are active only when operation is superscore and the attempts setting is at least $2$, $3$, or $4$ respectively, with fields english2..4, math2..4, reading2..4, and science2..4.

Whether a college will use it

The calculator reports what the arithmetic yields, not what an institution will accept. Colleges differ: some superscore ACT results, some consider only the highest single-sitting composite, and some want every sitting reported. Check each school's published testing policy before relying on a superscore figure.

ACT to SAT Conversion: The 2018 Concordance

ACT and SAT use different scales and measure different things, so a comparison is not a unit conversion. It is a concordance, an equipercentile linkage from the 2018 ACT/College Board study that pairs scores sharing the same percentile standing in the study population. It tells you which score on one test sits at roughly the same rank as a given score on the other. It does not predict how you would perform on the other test.

The act-to-sat operation accepts an ACT composite from $9$ to $36$ and returns a single SAT point equivalent plus the concordance range. The sat-to-act operation accepts an SAT total from $590$ to $1600$ in steps of $10$ and returns the concorded ACT composite. Every SAT total inside a published range maps to the same composite: $1300$, $1310$, and $1320$ all map to ACT $28$.

Reference table (Tables A1 and A2)

Single point equivalent versus interval

Each ACT composite has one point equivalent but covers an interval of SAT totals, because SAT totals are far more granular than ACT composites. The interval widths are uneven (the ACT $27$ row spans $1260$ to $1290$, while ACT $32$ spans only $1420$ to $1440$). That unevenness is a property of the study data, not a calculator artifact. When comparing scores for a school or scholarship, the range is usually the more honest figure than the point value.

Bounds and limits

• ACT composites $1$ through $8$ are valid scores, but the 2018 tables publish no concordance row for them. The calculator returns a warning and does not extrapolate.
• SAT totals are accepted from $400$ to $1600$ in increments of $10$. Totals below $590$ have no official equivalent and are rejected.
• The concordance was built on the SAT and ACT as they existed in 2018. It remains the official published table, but treat the output as a comparison of scales, not a prediction of testing outcomes.

What the Calculator Does Not Compute

• Raw scores. Only scaled $1$ to $36$ inputs are valid, for the equating reasons above.
• The Writing test. The optional Writing score (2–12) is never part of the composite.
• Admissions outcomes. There are no acceptance odds, GPA models, Ivy League Academic Index calculations, LSAC admissions figures, or PreACT and WorkKeys scaling.

Outputs, Visualizer, and Export

The reactive SVG chart (act-score-visual-svg) draws horizontal bars on a $0$ to $36$ scale for each active section. In superscore mode each bar carries a badge naming the credited attempt, such as "From attempt 2". In concordance mode it shows comparison cards with the point equivalent and range. The chart supports zoom and pan.

A one-click button copies a plain-text summary of your inputs and results to the clipboard, and the page carries print styling so it can be saved as a PDF for a counselor or a records file.

Presets