Average IQ in Australia: which sample and which scale?

0 The short answer

Australian IQ results must be read with the test's norm group. The WAIS-5 A&NZ uses Australian and New Zealand norms collected in 2023 and 2024; a combined scale centered near 100 does not prove that each country separately has a mean of exactly 100. Other figures, including 99.20 in a historical compilation and 104.45 among online test takers, come from different evidence. This page traces those samples, explains their limits and distinguishes them from PISA and adult skills results. None of these national summaries tells you an individual Australian's IQ.

1 What a Norm Centered at 100 Actually Means

A test norm centered at 100 supplies a reference point; it does not guarantee that every later sample, country subgroup or age cohort has exactly that mean. In a conventional deviation IQ system, performance is converted to a scale with a mean near 100 and a standard deviation near 15 in the reference population. The page on how IQ scores are normed explains the conversion, and the page on the standard deviation of 15 explains the unit. For a combined Australian and New Zealand reference frame, centering the combined distribution does not separately fix the Australian and New Zealand means. Estimating either mean requires appropriate subgroup data. Estimating a contemporary national population requires a suitable sampling design, date and definition of the score.

For most of the history of testing in Australia the norm samples were foreign, and the closest thing to a local anchor was British. Raven's Progressive Matrices, the test behind the great majority of samples in every international table and now sold in the edition described on the page about Raven's 2, was standardized in Britain in 1938, 1979, 1992 and 2007. John Raven's account of that history in Cognitive Psychology records that the Australian Council for Educational Research, through de Lemos's Australian re-standardization of the Standard Progressive Matrices, and the New Zealand Council for Educational Research, through its 1984 norms supplement, both reported results closely corresponding to the 1979 British norms. In plain terms, Australian and New Zealand children of the mid 1980s scored where British children of 1979 scored, This historical correspondence is evidence about those standardization samples; it is not an identity that forces every later Australian or New Zealand sample to score 100.

The Wechsler scales told the same story in a different form until 2025. Pearson's own page for the WASI-II, read on September 20, 2026, states that the WAIS-IV Australian and New Zealand Language Adapted Edition and the WASI-II both use U.S. norms. An Australian adult tested on the WAIS-IV A&NZ therefore received a position within an American standardization sample, Such results use a US reference frame and can be studied, but they should not be described as scores based on a local adult norm sample. The WAIS-5 A&NZ changed that, as the next section describes, and it is the reason the Australian case now differs from the British one examined on the page about average IQ in the UK, where the fourth edition kept American norms after a validation study.

2 The Instrument Australian Psychologists Use, and Whose Norms It Carries

Since 2025 the adult Wechsler scale sold in Australia reports a position among Australians and New Zealanders tested in 2023 and 2024, which makes its 100 an Australian and New Zealand mean rather than an American one. Pearson's product page for the WAIS-5 A&NZ, read on September 20, 2026, gives a publication date of 2025, an age range of 16 years 0 months to 90 years 11 months, a completion time of 45 minutes for the seven subtest Full Scale IQ or 60 minutes for the ten primary index subtests, and describes 2023 to 2024 Australian and New Zealand norms, co-normed with the WMS-5 A&NZ. It lists five primary index scores and fifteen ancillary index scores, and it states that the test was normed on individuals whose primary language is English. Pearson's own announcement of the edition adds that the normative samples were collected in 2023 and 2024 from a sample matched to the most current census on five demographic variables. The size of the standardization sample is not printed on either page, and this page does not guess it.

The sample score report Pearson publishes for the edition, dated August 27, 2025, shows what the norms do in practice. Subtest scaled scores are reported both against the examinee's own age band and against a reference group aged 20 years 0 months to 34 years 11 months, the composites carry a 95 percent confidence interval, and each composite prints its own standard error, 3.67 points for the Full Scale IQ of the 70 year old examinee in the sample. The page on what the WAIS-5 is sets out the structure of the American edition, and the page comparing the WAIS-4 and WAIS-5 explains what changed between editions. The complete kit was listed at 3,500 Australian dollars on the day of reading, a price that will change.

The edition it replaces is still on sale. Pearson's page for the WAIS-IV A&NZ Language Adapted Edition, read the same day, gives a publication date of 2008, an age range of 16 to 90 years 11 months, an administration time of 67 minutes and a qualification level of C. It explains that the test was language adapted for Australia and New Zealand, and it announces that the WAIS-IV scoring software will no longer be supported from 2027. What it does not say, and what the WASI-II page says for it, is that its norms are American. An Australian adult who holds a WAIS-IV A&NZ report therefore holds a position within the American standardization sample of the fourth edition, and the same adult tested today on the WAIS-5 A&NZ holds a position within an Australian and New Zealand sample of 2023 and 2024. The editions differ in norms and other test features. A difference between their scores may reflect those changes, administration or person related variation; it cannot be attributed to a change in the person alone.

Children were scored locally earlier than adults. Pearson's page for the WISC-V A&NZ describes an updated Australian and New Zealand standardization sample for ages 6 years 0 months to 16 years 11 months, published in 2016, and the WASI-II page records that the fourth edition of the children's scale already used Australian norms. Australian Mensa's list of accepted tests, read on September 20, 2026, names the WAIS-IV and the Australian standardization of the WISC-V among the instruments whose results it accepts at or above the 98th percentile when administered by a psychologist with general registration, and had not yet added the WAIS-5 on the day of reading.

3 The Six Samples Behind 99.20

Becker's Australian figure rests on six rows of Raven's data, four of them children, none of them scored against an Australian norm, and the two that anchor it are the 1986 Australian standardization of the Standard Progressive Matrices scored against British tables. The national table lists 5,962 people across the six samples, a mean age of about 19, a lowest sample value of 93.60 and a highest of 105.83. The largest row, recorded from the Raven manual, is the 1986 Australian re-standardization sample, split by Becker into about 3,333 children aged 8 to 15 scored against the 1979 British norms, at 101.75 before correction and 100.28 after, and about 667 young people aged 16 to 17 scored against the 1992 norms, at 104.57 before correction and 105.83 after. A Queensland sample of 640 children aged 5 to 10, tested in 1980 on the Coloured Progressive Matrices and scored against the 2007 British norms, comes in at 92.33 before correction and 98.00 after, the gap being Becker's allowance for the 27 years of norm drift between testing and norming. Raven's dominates such compilations because matrices are the standard example of a culture fair intelligence test, built to lower the load of language, although no test is fully free of culture.

The remaining three rows are recent and small. The Victorian normative study by Cotton and colleagues, listed by Becker as 615 children aged 6 to 12 tested in 2005 and scored against the 2007 norms, gives 93.18 before and 93.60 after correction, the lowest row in the file. Two adult samples on the Advanced Progressive Matrices, scored against the 1992 British norms, complete the set: 455 adults aged 16 to 68 from Waschl and colleagues' Adelaide study at 99.69 before and 96.33 after correction, and 252 mid-level managers from Birney and colleagues at 101.75 before and 98.39 after. The corrections subtract points from the recent adult samples because the 1992 norms are older than the testing, and add points to the old child samples because their norms are younger than the testing, which is the Flynn effect applied in reverse.

Weighting the six rows by their listed sample sizes, our arithmetic on the dataset file gives a mean of 99.59 after correction, which is Becker's own sample weighted value, and an unweighted mean of 98.74, which is his unweighted one. His quality weighted figure is 99.54, his school assessment conversion gives 98.85, and the combination of the two, the 99.20 that appears in rankings, is the average of a Raven's mean and a schooling mean. The two 1986 standardization rows, which hold two thirds of the people, average 101.21 after correction by the same arithmetic, which is what Raven's own remark about the Australian norms corresponding closely to the British ones would predict. The remaining variation can reflect sample selection, test differences, norm dates and the adjustment model. This audit does not identify how much each factor contributed. The two decimal places show the calculation's output, not national measurement precision.

The comparison countries in the same file put the figure in context. Taiwan stands at 106.59, Japan at 106.39, Singapore at 106.17, Hong Kong at 105.35, China at 102.84 and South Korea at 102.64; Germany at 100.61, the United Kingdom at 99.66, Canada at 99.62 and New Zealand at 99.32; Australia at 99.20 and the United States at 97.50, by our reading of the same sheet. The British figure examined on the page about average IQ in the UK rests on samples that are the norm samples themselves, and the American figure examined on the page about average IQ in the USA rests on 57 rows of very unequal quality. The ordering depends on the included studies and transformations. Similarity to a British norm does not establish that an Australian sample was recruited to reproduce that score. Nor can departures from the norm be dismissed as sampling noise without estimating the relevant uncertainty and evaluating selection.

4 The Self-Selected Online Table: 104.45 and Fifth Place

The figure most Australians meet first is 104.45, fifth in the world, and it describes the 4,245 people who chose to sit one website's test in 2025, not the country. The International IQ Test country ranking, last updated on January 1, 2026 and read on September 20, 2026, pools 1,212,714 people who took the same test on the site during 2025, fixes the mean of the whole pool at 100 with a standard deviation of 15, and excludes countries with fewer than 100 takers. On that table Australia ranks fifth of 137 countries at 104.45 from 4,245 participants, behind South Korea at 106.97, China at 106.48, Japan at 106.30 and Iran at 104.80. New Zealand ranks ninth at 102.35 from 1,184 takers, Canada thirteenth at 102.09 from 7,685, the United Kingdom fourteenth at 101.57 from 7,310, the United States eighteenth at 101.04 from 22,605, and Germany thirty-eighth at 99.32 from 34,350.

The table's own previous year column is the best argument against reading it as a population figure. Australia's 2024 entry was 102.57 from 9,626 takers; in 2025 the takers fell by more than half and the mean rose by 1.88 points. The United Kingdom moved from 99.68 to 101.57 and the United States from 99.74 to 101.04 over the same year, while the site reports that in 84.8 percent of countries the mean shifted by less than 2 points. A mean that moves two points when the sample changes composition is a mean of the sample, and the ACIS page on average IQ by country uses this dataset precisely because it is one selection process applied to every country, with the warning printed beside the tool that no country in it was sampled by probability and none of its figures is age normed against that country's population.

What the table can support is narrow. Australian takers of this test scored about three points above British and American takers of the same test, A hypothetical independent sample with SD 15 and n = 4,245 would have a standard error of about 0.23 points and a simple 95 percent margin near 0.45 points. That arithmetic does not establish the site's actual sampling uncertainty and excludes selection, dependence and measurement bias. What it cannot support is the sentence usually built on it, that Australians are among the five most intelligent nations. The review of the International IQ Test examines the site on its own terms, and the page on free versus validated IQ tests explains why a self-selected mean is not a norm. A state or city ranking of Australian IQ built from the same kind of data, of which several circulate, is the same selection effect broken down by postcode.

5 Australia, New Zealand, the UK and the US Across Every Frame

Comparisons change with the reference frame, participants, test and date. The table places several published figures beside one another so the reader can see that dependence. It does not establish that two countries are identical or that a difference is statistically reliable. A shared test makes a comparison more interpretable only when administration, language, population coverage and sampling also support it. PISA uses its own achievement scale and inference procedures; its values are not converted here into IQ. Likewise, a small difference in a historical compilation is not evidence of a precisely measured national distinction.

Two of those rows deserve a sentence each. The Becker row for New Zealand rests on only two samples, the 1984 New Zealand standardization of the Standard Progressive Matrices, 3,174 children aged 8 to 15 scored against the 1979 British norms at 100.74 after correction, and the Dunedin birth cohort of 644 children on the WISC-R at 99.09, so the New Zealand sample mean of 100.60 is a statement about 3,818 children with a mean age of ten. The PISA row is the only one in which every country was sampled by probability, and the OECD's summary of PISA 2025, from which the New Zealand, British and American figures are taken, flags New Zealand and the United States with a caution because one or more sampling standards were not met. The Australian brief states the comparison directly: Australia's science score was the same as those of the United Kingdom, Canada, New Zealand, Finland and the United States, its reading score was below New Zealand's and the same as the British, Canadian and American scores, and its mathematics score was below the British and Canadian scores, the same as New Zealand's and above the American and German scores.

Australia and New Zealand can in principle be compared using a shared instrument and appropriate samples. Sharing a norm frame does not eliminate country specific means: just as two age matched groups can differ on the same scale, two countries can have different observed means within a combined reference frame. What this page lacks is a current representative adult comparison that establishes those subgroup means and their uncertainty. Becker's combined values differ by 0.12 points and the online participant means by 2.10 points, but those are results of their particular designs. The sibling pages on average IQ in Canada and average IQ in Germany examine the same distinction between a norm reference and a population estimate.

6 Average IQ in Australia by Age

Age norming compares performance with an age appropriate reference group; it does not require every current Australian age group to have an observed mean of exactly 100. The page on average IQ by age explains why raw task performance and age adjusted standing are different. A 25 year old and a 70 year old can receive the same age adjusted IQ after achieving different raw scores. The WAIS-5 A&NZ report also distinguishes an age comparison from a young adult reference comparison. Readers should check which column they are interpreting rather than treating one as an error in the other.

A cross-sectional analysis of American WAIS-5 norms illustrates how conclusions depend on the comparison. Scoring age groups against a common younger reference reveals age related differences that age specific norms account for. That study concerns American normative data, so this page does not transfer its precise estimates to Australia or treat a cross-sectional difference as one person's longitudinal decline. Cohort, education and selection can accompany age in a cross-sectional comparison. The page on fluid versus crystallized intelligence explains why different cognitive tasks can have different age patterns without assigning a fixed trajectory to every adult.

Australia's age structure enters the national figure through education and cohort rather than through the norms. The Australian Bureau of Statistics' Education and Work, Australia, May 2025, released on November 19, 2025, reports in its table of people aged 15 to 74 with a bachelor degree or above by age group and sex that 34 percent of Australians in that age range held a bachelor degree or above, and that those aged 25 to 34 were the most likely to, at 47 percent. These figures describe educational attainment in the specified age groups; they do not establish a monotonic trend for every successive cohort or an IQ difference of a given size. The sources reviewed here do not provide a current representative IQ mean for every adult decade on a common scale. Age norms would not prevent such a study, but a suitable sample and a clearly specified reference would be required.

7 Average IQ in Australia by Education

Education is one contributor to test performance, alongside other developmental, environmental and measurement factors. The Australian Bureau of Statistics' Education and Work release for May 2025 reports that 64 percent of people aged 15 to 74 had attained a non-school qualification, that 34 percent held a bachelor degree or above, that 89 percent of those aged 20 to 24 had attained Year 12 or a Certificate III or above, and that more than 3.2 million people were studying. Those are the population's schooling figures, and they matter for any IQ table because schooling raises test performance. Ritchie and Tucker-Drob's meta-analysis of the effect of education on intelligence, across 142 effect sizes from 42 datasets and more than 600,000 people, found gains of about one to five IQ points for each additional year of education, persisting across the lifespan and present on every broad category of ability studied. Those study estimates are not a formula for predicting Australia's national change from its degree share. Educational attainment can also reflect selection and cohort composition. Renorming changes the score reference; it does not identify the cause of any underlying performance change.

The gradient within Australia is measured on stated scales rather than on IQ. The ABS Programme for the International Assessment of Adult Competencies, Australia, 2011 to 2012, the OECD survey of adult skills, placed 37.9 percent of Australians aged 15 to 74 at literacy level 3, 14.5 percent at level 4 and 1.2 percent at level 5, which is 53.6 percent at level 3 or above by our addition, with 10.4 percent at level 1 and 3.7 percent below it. In numeracy the shares were 31.3, 10.9 and 1.4 percent at levels 3, 4 and 5, or 43.6 percent at level 3 or above, with 15.3 percent at level 1 and 6.5 percent below it. The ABS states that it has no plans for a future PIAAC release and that Jobs and Skills Australia's Foundation Skills Study takes over the measurement, which means that Australia did not report in the 2023 cycle from which the British and American adult figures on other pages of this site are drawn. The ACIS page on average IQ by education shows the online version of the gradient, a difference of 5.76 points between the no diploma and four year degree groups in one self-selected table, and the page on IQ and academic achievement explains the individual level relationship that both kinds of table are gesturing at.

8 The Flynn Effect in Australia

Australian studies contribute to the evidence on changes across cohorts, but trends depend on the task, population and period. Meta-analyses summarize that variation rather than establishing one universal rate. Trahan and colleagues' meta-analysis of 285 studies and 14,031 people found a mean gain of 2.31 standard score points per decade, with a 95 percent confidence interval of 1.99 to 2.64, and 2.93 points per decade on modern Wechsler and Stanford-Binet tests. Pietschnig and Voracek's century of global IQ gains, across 271 samples, almost four million people and 31 countries from 1909 to 2013, put the annual gain at 0.41 points for fluid tests, 0.30 for spatial, 0.28 for full scale and 0.21 for crystallized, stronger for adults than for children and slowing in recent decades. Flynn's original analysis of 14 nations in 1987 reported gains of 5 to 25 points in a single generation, largest on culturally reduced tests of the Raven's kind.

The Australian record consists of the standardization history and two studies. The first item is the correspondence Raven recorded between the 1986 Australian re-standardization and the 1979 British norms, which places Australian children of 1986 at the level of the 1979 British tables rather than the 1938 ones. The second is Cotton and colleagues' normative study of the Coloured Progressive Matrices in Victoria, published in Personality and Individual Differences in 2005, which presented percentile ranks for six age levels from a sample of 618 primary school children aged 6.00 to 11.92 years, reported good inter-item consistency and split-half reliability at each level, and concluded that the item count could be reduced without loss of reliability. Local normative studies of this kind are needed because imported norms age, which is the practical face of the Flynn effect. The third item is Nettelbeck and Wilson's study in Intelligence in 2004, titled The Flynn effect: smarter not faster, which compared Australian children tested in 1981 and 2001 on the same measures and found that word knowledge on the Peabody Picture Vocabulary Test had risen by about five points across the 20 years while inspection time, a measure of the speed of taking in visual information, had not changed at all.

Becker's corrections make the arithmetic of the effect visible in the Australian rows. The Queensland children of 1980 scored 92.33 against the 2007 British norms and 98.00 after he added 5.67 points for the 27 years of drift; the Adelaide adults of 2016 scored 99.69 against the 1992 norms and 96.33 after he subtracted 3.36 points for the 24 years in the other direction. The page on the Flynn effect explains why a historical group trend is not an automatic correction for an individual score, and the page on matrix reasoning explains why the fluid tests on which the gains were largest are also the ones on which a norm year matters most. A population mean can rise on a fixed reference scale even though a newly normed scale is centered again. The vocabulary and inspection time comparison concerns those particular tasks and cohorts; it does not prove that every Australian gain occurred in knowledge or that speed never changed.

The most recent Australian trend runs the other way, on a different scale. PISA in Brief Australia 2025 reports that between 2022 and 2025 the science mean did not change, the reading mean fell by 7 points and the mathematics mean fell by 9 points, and that over the longer run the science mean is 18 points below its 2006 level, reading 37 points below 2000 and mathematics 46 points below 2003, with more low performers and fewer high performers in each subject. Those are declines in measured achievement among fifteen year olds on the PISA scale, not declines in IQ, They do not by themselves establish an adult IQ trend. A reader who wants the two trends reconciled should note their dates: the Raven's gains are documented up to 2005, the PISA declines from 2000 onwards, A future edition could help study trends only if the comparisons account for changed tasks, samples, administration and norms. Its release alone would not identify the direction or cause of a national change.

9 What Representative Australian Data Actually Measure

Australia has representative measurements of cognitive skills at fifteen and, until 2012, in adulthood, and they are more informative than any IQ table because they say who was tested, how they were sampled and what the score means. PISA 2025, whose Australian brief De Bortoli and Underwood published for ACER on September 8, 2026, tested 14,658 students from 766 schools, representing about 275,000 Australian fifteen year olds, who were aged between 15 years 3 months and 16 years 2 months at the time and mostly in Year 10. Australia scored 509 in science against an OECD average of 482, with seven countries significantly above it and 78 below; 491 in reading against 461, with nine above and 77 below; and 478 in mathematics against 463, with twelve above and 71 below. Sixty percent of Australian students reached the National Proficient Standard in science against 49 percent across the OECD. The brief reports every difference only where it is statistically significant, and it states the sampling frame, the exclusions and the weighting.

The adult measurement is older. The ABS PIAAC release of 2011 to 2012 is the last representative survey of Australian adult literacy and numeracy on an international scale, and the figures above, 53.6 percent at literacy level 3 or above and 43.6 percent at numeracy level 3 or above, are on the survey's own five level scale. The Household, Income and Labour Dynamics in Australia survey, a representative household panel, added three brief cognitive measures in its 2012 wave, Backwards Digit Span, the Symbol Digit Modalities Test and a 25 item version of the National Adult Reading Test, described in Wooden's working paper for the Melbourne Institute, and it reports them as raw scores. These measures should be interpreted as reported in the study. Constructing a broad cognitive composite or an IQ scale would require a justified model, norming method and validation; a responsible study can undertake that work, but this page does not silently substitute such a conversion for the published scores.

Skills and cognitive ability can be related without becoming interchangeable. A literacy score also depends on learned language and reading experience; a mathematics result reflects its particular task and educational context. The what IQ measures guide explains why construct coverage matters, while the page on IQ and income distinguishes associations across people from predictions about one person. A historical compilation may transform school assessment results into its chosen index, but the transformation remains a modeling decision. PISA itself has not thereby become an IQ test. Representative surveys are valuable because their own definitions, populations and uncertainty can be examined on their own terms.

10 Within Australia: State, City, Language and Background

The differences that are documented within Australia are differences by state and territory, by location, by socioeconomic background and by language at home, each measured on the PISA scale by a probability sample, which is more than can be said for any regional IQ ranking. In science the Australian Capital Territory scored 537, New South Wales and Victoria 513, Queensland 512, South Australia 505 and Western Australia 501, while Tasmania at 484 and the Northern Territory at 465 were not significantly different from the OECD average. In reading the range ran from 517 in the Australian Capital Territory to 445 in the Northern Territory, and in mathematics from 502 to 436. ACER's summary of the results reports that students in major city schools, three quarters of the sample, outperformed students in regional schools by 24 points in science, 27 in reading and 30 in mathematics.

The socioeconomic gradient is the largest gap in the report. Students in the highest quarter of socioeconomic background scored 549 in science, 529 in reading and 519 in mathematics against 469, 455 and 436 for the lowest quarter, gaps of 80, 74 and 83 points, and the share reaching the National Proficient Standard in mathematics ran from 31 percent in the lowest quarter to 66 percent in the highest. Girls and boys scored the same in science, at 509 each; girls led in reading, 507 to 475; boys led in mathematics, 486 to 469. Eighty-five percent of students spoke English at home, and they scored 514, 496 and 478 in the three subjects against 500, 482 and 486 for students who spoke another language, a gap in science and reading and none in mathematics. Half the students were Australian born to Australian born parents, 35 percent were first generation and 14 percent foreign born, and first generation students scored higher than Australian born students in every subject.

Those figures describe a country in which the national mean hides a great deal, and they are the reason a single Australian IQ is less informative than the table above it. The ABS release on Australia's population by country of birth, for June 30, 2025, counts 8.8 million overseas born residents, 32.0 percent of a population of 27.6 million, with India, England, China and New Zealand the largest sources. A norm sample matched on selected demographic variables may improve coverage, but that does not prove equal coverage of every language or background group; an online table includes whoever turned up; and Becker's rows, drawn from Queensland in 1980, Victoria in 2005 and Adelaide in 2016, include none of it in any controlled way. The page on average IQ by state shows how the same selection problem plays out in the American tables, and the socioeconomic correlations in the report say nothing about any one student, which is the point the last sections make.

11 The Evidence by Design

Read by design rather than by headline, the Australian evidence sorts into a local norm sample that defines the scale, six compiled study rows, including standardization and other samples, on foreign norms, one self-selected online table, and three representative surveys on scales of their own, and the third column of the table below is the one to keep.

The practical comparison is between evidence designs. A local norm sample defines a test reference for its intended population. A historical compilation aggregates studies and applies transformations. A participant website describes the people who took part under its conditions. A probability survey estimates defined outcomes for a target population, subject to its coverage and sampling limits. These are different sources of evidence, not steps in one universal ranking of national intelligence. A careful reader can still learn from each source once the question and limits are kept explicit.

What the Evidence Supports, Stated Narrowly

The evidence supports several qualified statements, rather than one definitive Australian IQ. The WAIS-5 A&NZ supplies a regional adult norm frame, while its predecessor used a different reference. Becker's compilation returns 99.20 under its combined method and 99.59 under the sample size weighting discussed above; these are outputs from the included studies and adjustments. The online table returns 104.45 for 4,245 participants, whose participation mechanism limits population inference. PISA describes school age performance on achievement scales, and the older PIAAC release describes adult skills. Differences across those figures cannot be interpreted as one country simultaneously possessing several contradictory underlying IQs, because the definitions, samples and dates differ.

The sources do not establish a precise, contemporary adult IQ ranking for Australia or a claim that Australia and New Zealand must have identical population means. They do not identify genetic causes of national differences or justify conclusions about a particular person from nationality. A state or city table needs its own sampling and validation evidence; a postcode attached to a voluntary online response does not provide that evidence. The page on average IQ explains the role of the reference population, and the page on what a good IQ is asks the separate individual question. Keeping these limits explicit preserves useful information instead of discarding all group research or overstating it.

12 What This Does Not Say About You

A national summary does not establish where an individual Australian falls within a cognitive score distribution. Under an approximately normal model with mean 100 and SD 15, about 68 percent fall between 85 and 115 and about 95 percent between 70 and 130. Those are model proportions, not a verified current Australian distribution or an individual prediction. A regional reference frame also has age, language and administration conditions to consider. A score of 130 sits at the 98th percentile of the reference population, and that percentile is the threshold Australian Mensa applies: on the day of reading, September 20, 2026, its supervised test carried an administrative fee of 75 Australian dollars, or 60 with a concession, was open to candidates aged 10 to 75, and could be replaced by evidence of a qualifying score on an approved test administered by a registered psychologist. The page on the Mensa IQ test explains why the same standing reads as 148 on a scale with a standard deviation of 24, the page on the IQ score chart shows the whole distribution, and the percentile calculator converts any score on any stated scale.

The direction of inference matters as much as the arithmetic. An online table built from 4,245 takers does not describe a reader who did not take that test. A Victorian percentile table from 2005 describes primary school children of that year. A PISA score describes fifteen year olds in a sampling frame, and a PIAAC level describes adults in a household survey of 2011, and neither is a number any individual holds. The page on score versus percentile explains why the same standing looks different on different scales, and any single score is an interval rather than a point, as the standard error printed on every WAIS-5 A&NZ composite reminds the reader.

For an Australian adult who wants a number that describes them, the position this page has applied to every dataset applies again: ask whose norms, whose sample and what the standard error is. A clinical administration of the WAIS-5 A&NZ reports a position within the 2023 to 2024 Australian and New Zealand sample, with the standard error of each composite printed on the report; a WAIS-IV A&NZ report from before 2025 reports a position within American norms. We sell a competing paid assessment, so treat the rest of this paragraph as a disclosure and check it against the rest of the page. An ACIS report scores a person against an English speaking adult reference frame of 3,243 records aged 16 to 90, described in the technical manual, whose version 1.4 Full Scale analysis reports composite reliability .9886 and SEM 1.6029 for the documented score definition, and prints the percentile and confidence interval of every index. That frame is not an Australian national sample, the test is taken online without supervision, it is not a clinical diagnosis, it is not accepted by Mensa for admission, it produces no signed report and it is English only; the page on reliability and validity states what each of those limits means for a score. Any of those instruments, read with its limits, is a measurement of the person. The national figure, from any source, is not, and the form, version and intended use remain essential to interpreting any personal result. Published reliability from one configuration must not be transferred automatically to every revised form.

13 Sources Behind This Page

Every figure above is traceable to one of the following, and each is linked at the point where it is used. Dataset values for Australia, New Zealand and the comparison countries were read from the NIQ dataset file downloaded from View on IQ on September 20, 2026, and the calculations described as ours were performed on that file; the Adelaide and manager samples in it are Waschl and colleagues 2016 and Birney and colleagues 2017. Product details and prices were read on Pearson's Australian pages on the same day, including the WASI-II page that states the WAIS-IV A&NZ uses U.S. norms, the WISC-V A&NZ page, the WAIS-5 A&NZ announcement and the WAIS-5 A&NZ sample score report, and they will change. Mensa conditions were read on Australian Mensa's joining page and accepted tests page, the population figures on the ABS country of birth release, the HILDA measures in Wooden 2013, and the 14 nation figure in Flynn 1987.

• Becker D. The NIQ dataset, version 1.3.5, released October 25, 2023. viewoniq.org, read September 20, 2026.
• Lynn R and Vanhanen T. IQ and the Wealth of Nations. Praeger, 2002. doi.org.
• International IQ Test. Average IQ by country 2026, ranking of 1,212,714 takers in 2025, last updated January 1, 2026. international-iq-test.com, read September 20, 2026.
• Raven J. The Raven's Progressive Matrices: Change and stability over culture and time. Cognitive Psychology, 2000, volume 41, issue 1, pages 1 to 48.
• Pearson Clinical Assessment AU&NZ. Wechsler Adult Intelligence Scale, Fifth Edition, Australian and New Zealand (WAIS-5 A&NZ), product page with the 2023 to 2024 norms, timing, prices and FAQ. pearsonclinical.com.au, read September 20, 2026.
• Pearson Clinical Assessment AU&NZ. Wechsler Adult Intelligence Scale, Fourth Edition, Australian and New Zealand Language Adapted Edition (WAIS-IV A&NZ), product page. pearsonclinical.com.au, read September 20, 2026.
• Cotton S M, Kiely P M, Crewther D P, Thomson B, Laycock R and Crewther S G. A normative and reliability study for the Raven's Coloured Progressive Matrices for primary school aged children from Victoria, Australia. Personality and Individual Differences, 2005, volume 39, issue 3, pages 647 to 659.
• Nettelbeck T and Wilson C. The Flynn effect: Smarter not faster. Intelligence, 2004, volume 32, issue 1, pages 85 to 93.
• Trahan L H, Stuebing K K, Fletcher J M and Hiscock M. The Flynn effect: A meta-analysis. Psychological Bulletin, 2014, volume 140, issue 5, pages 1332 to 1360.
• Pietschnig J and Voracek M. One century of global IQ gains: A formal meta-analysis of the Flynn effect (1909 to 2013). Perspectives on Psychological Science, 2015, volume 10, issue 3, pages 282 to 306.
• De Bortoli L and Underwood C. PISA in Brief Australia 2025: Student performance. Australian Council for Educational Research, September 8, 2026.
• OECD. PISA 2025: Insights and Interpretations, summary tables of country scores in science, reading and mathematics. oecd.org, September 2026.
• Australian Bureau of Statistics. Education and Work, Australia, May 2025, including the table of people aged 15 to 74 with a bachelor degree or above by age groups and sex. abs.gov.au, released November 19, 2025.
• Australian Bureau of Statistics. Programme for the International Assessment of Adult Competencies, Australia, 2011 to 2012. abs.gov.au, released October 9, 2013.
• Ritchie S J and Tucker-Drob E M. How much does education improve intelligence? A meta-analysis. Psychological Science, 2018, volume 29, issue 8, pages 1358 to 1369.
• Winter E L, Dale B A, Maharjan S, Lando C R, Larsen C M, Courville T and Kaufman A S. Cognitive aging revisited: A cross-sectional analysis of the WAIS-5. Journal of Intelligence, 2025, volume 13, issue 7, article 85.

The Standards for Educational and Psychological Testing (AERA, APA and NCME, 2014) and the APA Ethical Principles of Psychologists and Code of Conduct, Assessment standards provide the interpretive framework: evidence must support the use, the reference group and limitations must be stated, and assessment results require appropriate explanation. Citing these standards does not imply their authors endorse ACIS or another product.

14 Frequently Asked Questions