IQ vs the GED: what a credential test measures, and what the research found

0 The short answer

No IQ score corresponds to a GED score: the GED is a four part achievement credential that is passed at 145 on each subject, on a scale that runs from 100 to 200, and a passing score says that a candidate reached a standard, not where the candidate ranks in ability. The best known research, by Heckman, Humphries and Mader, found that people who go on to earn a GED look similar to high school graduates on an achievement test but fall short of them in noncognitive skills such as persistence, and a 2019 study that used the WAIS-IV in a small clinical sample found GED holders about seven IQ points below graduates and no different from dropouts. The two findings use different people, different measures and different designs, so they do not contradict each other, and neither converts a GED into an IQ.

1 What Does the GED Test Measure, and How Is It Scored?

The GED is a battery of four subject tests that certifies high school level achievement, and its scores describe performance against a standard, not a position on an ability scale. The GED Testing Service page on the test lists the four subjects and their times as read on October 13, 2026: Mathematical Reasoning (115 minutes), Reasoning Through Language Arts (150 minutes, including 45 minutes for a written essay), Social Studies (70 minutes) and Science (90 minutes). Adding those times gives 425 minutes, which is our arithmetic, since the page states no total. The question formats are multiple choice plus drag and drop, fill in the blank, select an area and drop down items, and the language arts test adds one extended response. The page also says candidates do not have to take all four tests at once. In their review of the research, Heckman, Humphries and Mader describe the battery as an eight hour subject based test that has been revised three times, and they say the content moved away from factual recall toward interpretation and analysis of information.

The scoring is a pass or fail standard per subject. The GED Testing Service scores page says the test has a passing score of 145 for each subject and that a candidate needs 145 or higher on all four subjects to earn the high school equivalency. A GED Testing Service presentation on interpreting scores, dated July 20, 2023, gives the range as 100 to 200 points per content area and names four performance levels: not passing at 100 to 144, high school equivalency at 145 to 164, college and career ready at 165 to 174, and college and career ready plus credit at 175 to 200. The scores page describes the two higher levels as showing readiness for college level courses, with possible waivers from placement testing, and skills that could qualify for up to 10 college credit hours. It does not say how any of the three cut points was determined, and it does not mention norming. The same presentation explains the older scoring: the 1988 series asked for a total of 225 points across five tests, an average of 45, and the 2002 series asked for 2,250 points, an average of 450, with a minimum of 410 on each test.

Two features of this design matter for any comparison with IQ. First, a passing score is a threshold, so everyone from the candidate who barely reached 145 to the candidate who scored 160 holds the same credential, while an IQ report places each person on a continuous distribution with a percentile and a confidence interval. Second, the unit is different. A subject score of 145 is not a standard score with a mean of 100 and a standard deviation of 15 in any population the GED Testing Service describes on these pages. The page on what IQ actually measures describes the construct an IQ battery targets, and the section below on the CHC model places the GED against it.

2 Is There an IQ Equivalent to a Passing GED Score?

No published source gives an IQ equivalent for a GED score, and the number 145 is the clearest trap, because an IQ of 145 and a GED subject score of 145 have nothing in common. On the standard IQ scale with a mean of 100 and a standard deviation of 15, an IQ of 145 is three standard deviations above the mean, about the 99.9th percentile (our arithmetic on the normal curve), which is the territory the page on IQ 145 describes. A GED subject score of 145 is the lowest score that passes. The coincidence of the two numbers invites the mistake of reading one as the other. The GED Testing Service pages read for this article contain no conversion, and the Heckman review does not rely on one: it measures ability with an achievement test, the AFQT, and compares groups on that test.

Three problems would defeat a conversion even if someone tried to build one. The first is the reference group. The GED Testing Service presentation says the percentile ranks printed on the 2002 series transcript were those of graduating high school seniors, not of adult candidates, and that passing surpassed 40 percent of those seniors. Graduating seniors are a selected group that excludes dropouts and is not the adult population, so a percentile among them cannot be moved onto the population scale of an IQ test. The page on why a percentile depends on its reference group explains the same problem for other tests. A lookup that treated the 40th percentile as a population percentile would return an IQ of about 96 on a standard deviation of 15 scale (our arithmetic), and that number would be wrong in two ways at once, because the group is wrong and the series is no longer current.

The second problem is the construct. The achievement tests that the research uses as a bridge, such as the AFQT discussed below, correlate with IQ but are not interchangeable with it, and the page on IQ versus the ASVAB shows why even a close relative of IQ cannot be converted one for one. The third is the decision rule. Passing is a yes or no outcome from four separate tests with no averaging across subjects (the scores page states that a high score on one cannot offset a low score on another), so even in principle a GED outcome carries less information about ability than a continuous score does. What the credential does tell you is that the holder met a stated standard in four content areas. The sections that follow turn to what research has learned about the people who earn it.

3 How Was the GED Passing Standard Set, and Why Did It Change?

The GED passing standard was set against graduating high school seniors and has been moved several times, so it describes a benchmark that changed, not a fixed amount of ability. In their review, Heckman, Humphries and Mader report an estimate that the original test could be passed on a first try by 80 percent of graduation bound high school seniors, a figure that includes a high chance of passing by guessing, and that after three increases in difficulty only 60 percent of current graduation bound seniors were estimated able to pass the entire test on a first try. They also report that the 1997 increase in the national minimum difficulty moved that share from 66 percent to 60 percent. These are the authors' reports of estimates from GED Testing Service sources and a secondary source, and the figures describe the 2002 era test, not the current four test battery.

The current four test battery has been in use since January 2014, and in January 2016 the GED Testing Service lowered the passing score on each subject from 150 to 145, according to reporting by Education Week dated January 26, 2016. The article says the company added the two higher benchmarks at 165 and 175 at the same time. It gives the company's stated reasoning: students who passed the tougher test did better in college than high school graduates, with fewer needing remedial coursework. The article reports Oregon community college figures for 2014 to 2015: 15 percent of GED passers needed reading or writing remediation, against 47.5 percent of high school diploma holders, and 39 percent against 62 percent in math. A company spokesman is reported as saying the aim was a cut score on par with the average graduating high school senior. The company projected that 100,000 more people could pass one or more subjects and 25,000 could become eligible for the credential by passing all four.

Read as measurement, that history teaches three things. A passing standard is a judgment about what should count as enough, so it can be moved by policy without any change in the people who take the test, which is the opposite of how a normed IQ score behaves. The reference group is again graduating seniors, so the standard is relative to a group that is a subset of the population. And the evidence offered for the 2016 change was college remediation among passers, which is a criterion outcome and says nothing about IQ. The Standards for Educational and Psychological Testing address this directly in Standard 5.21, which asks that when score interpretations involve cut scores, the rationale and procedures used to establish them be documented clearly. The page on how IQ tests are made separates that kind of standard setting from the norming that gives an IQ score its meaning.

4 Who Takes the GED, and Why Is the Group Hard to Compare?

GED recipients are not a random slice of the population or of dropouts, so any comparison with graduates compares people who differ in many ways besides the credential. The credential began in the military: Hewitt, Cody, Marker and Loring recount that the GED was established in 1942 so that service members unable to finish high school could demonstrate high school level proficiency for college placement, and that eligibility reached civilians in New York in 1947. They also point out that GED stands for General Educational Development, and that the common misreading as a General Equivalency Diploma has probably helped the credential be treated as the same thing as a diploma. Heckman and colleagues report the scale of the program in the abstract of their review: in 2008 almost 500,000 dropouts passed the test, about 12 percent of all high school credentials issued that year.

Motives vary. The review cites GED Testing Service data showing that in 2008, 60 percent of test takers reported further education as a reason for taking the test, and of that 60 percent, 20 percent planned a four year college, 28 percent a two year college and 22 percent a technical or trade program. It documents that the audience has widened over time through adult education programs and prison education programs, and it argues that the test now serves a more diverse group than the motivated, disciplined group it first targeted. That change in composition is the central difficulty. A study of GED recipients in 1980 and a study in 2005 sample different mixes of people, and a coefficient estimated in one era may not describe the other.

Selection into the credential also runs in a direction that matters for ability. People who leave school and then sit an eight hour test are people who left for a reason and then chose to certify. Some left to work or care for family, some left because of academic trouble, some left after trouble with behavior or health. The review's estimates reflect this mix: the people who eventually certify differ from the people who do not, and differ from graduates, in both ability and traits other than ability. The page on average IQ by education level lists education groups in a table, and a group label such as "high school or GED" inside such a table combines these different routes into one row, which is one reason a group average cannot stand in for any individual.

A last feature of the group is the age at which testing happens. A person who earns a GED at 17 and a person who earns one at 45 took the same test under the same standard but arrive with different schooling, different work histories and different time since formal study. The research below handles this by adjusting for schooling, and the adjustments matter to the result, so it is worth reading how they were done before drawing a conclusion about ability.

5 What Did Heckman, Humphries and Mader Find About the Cognitive Ability of GED Recipients?

Heckman, Humphries and Mader concluded that GED recipients are about as capable as high school graduates on a scholastic achievement test, and they stressed that this holds for a measure of achievement, adjusted for schooling. Their review, circulated as IZA Discussion Paper 4975 in May 2010 and as NBER Working Paper 16064, appeared in the Handbook of the Economics of Education in 2011. Its abstract says the GED establishes cognitive equivalence on one measure of scholastic aptitude, and its introduction says that GEDs are as smart as ordinary high school graduates as measured by a scholastic achievement test. The same introduction says that, as a group, GEDs fail to perform at the level of high school graduates in life outcomes.

The measure is the Armed Forces Qualification Test (AFQT), which the review describes in a footnote as an achievement test of numerical operations, arithmetic reasoning, paragraph completion and word knowledge. It was given in 1979 to the NLSY79 sample, then aged 14 to 22, and in 1999 to the NLSY97 sample, then aged 14 to 18. Because respondents took it at different ages and with different amounts of schooling, the authors adjusted each score for the level of schooling at the time of the test, conditioning on final years of completed schooling and on how the person left high school (dropout, GED or diploma). That yields two estimates per person: a pre-8th-grade score, which measures ability before dropout decisions, and a post-schooling score, which measures ability as people enter the labor market.

The findings are about those estimated distributions. Before high school, individuals who eventually sort into GED certification have higher cognitive ability than dropouts and are very similar to terminal high school graduates. At the end of schooling, terminal GEDs sit between dropouts and high school graduates but are more similar to graduates. The authors read the separation as evidence that the credential signals something real, and they also say it shows the need to control for academic ability earned through school attendance, which should not be credited to the GED. Their introduction states the logic plainly: GEDs have higher achievement test scores than dropouts in part because they complete more years of high school.

Three cautions apply before reading this as "GED holders have normal IQs". The AFQT is an achievement test, and the review notes in a footnote that other work shows it is predicted by both cognitive and noncognitive traits, so it is not a pure ability measure. The distributions are model adjusted estimates, not raw scores. And the finding concerns where GED bound students stood on a test as teenagers and young adults, compared with graduates and dropouts, so it says nothing about the level of any one person. The page on IQ and the ASVAB explains why the AFQT percentile cannot be translated into an IQ, which keeps the review's finding from being turned into a number.

6 Where Do GED Recipients Fall Short, if Not on Test Scores?

The review locates the shortfall in noncognitive skills and in outcomes that persist after test scores are controlled, which is why Heckman calls the GED a mixed signal. The review traces the argument to Cameron and Heckman's 1993 paper, The Nonequivalence of High School Equivalents, in the Journal of Labor Economics, and describes its finding as follows: on average the GED provides no benefit to male test takers after controlling for years of completed schooling or AFQT scores, and later work replicates the finding at later ages. The review adds estimates from Heckman and LaFontaine using later waves of the NLSY79: after correcting for selection and controlling for AFQT scores, male GEDs earn on average 1 percent less per hour than dropouts while terminal high school graduates earn 3.6 percent more, and female GEDs earn 1.7 percent more than dropouts while graduates with no college earn 10.6 percent more. High school graduation keeps a wage premium after the AFQT is controlled, which the authors read as graduates possessing a valued trait that an achievement test does not capture.

The mixed signal idea is credited in the review to Heckman and Rubinstein, whose 2001 paper on noncognitive skills and the GED program appeared in the American Economic Review. The review says dropping out is associated with negative social traits, among them criminal behavior, divorce, risky behavior and job turnover, that statistical studies often leave uncontrolled, and that a test of cognitive ability does not measure these traits. In the review's figures, GEDs resemble or fare worse than dropouts on teenage behavior measures, and estimated noncognitive ability is below that of uncredentialed dropouts for all GEDs except those with some college. The review adds that the gradient of the GED's wage effect is steeper with respect to noncognitive ability than cognitive ability, especially in the bottom deciles.

The same pattern shows up in what happens after the test. The review cites a GED Testing Service study that followed 1,000 randomly selected people who passed after the 2002 increase in difficulty and found that 31 percent ever enrolled in a postsecondary institution and that 77 percent of those who enrolled did so for only a single semester. In the review's words, GEDs go to college at higher rates than dropouts but few finish more than one semester. This is a reading of one cohort of the older test series, so it is not a statement about the current battery.

None of this proves that a low score on a trait caused the outcome. The authors write that noncognitive deficits explain the lack of success of many GEDs, but the studies are observational. What the research does support is narrower: test equivalence on an achievement measure does not imply equivalence in persistence, and persistence is something an IQ test does not measure either. The page on what IQ does and does not capture says the same about the limits of any ability score.

7 Does the GED Carry Labor Market Value? The Case For and Against

The evidence on the credential's labor market value is mixed: the average effect looks small or null, while one well known study finds a signal at the passing margin for some groups. The review's abstract says the literature finds minimal value for the certificate in labor market terms. The strongest counterweight is by Tyler, Murnane and Willett, Estimating the Labor Market Signaling Value of the GED, in the Quarterly Journal of Economics in 2000. Their working paper abstract says they used differences among states in GED passing standards as natural experiments, analyzed test scores merged with Social Security earnings for 16 to 21 year old dropouts who took the test in 1990, and found that earning a GED raised the 1995 earnings of young white dropouts near the passing margin by 10 to 19 percent. They found no statistically significant gain for young nonwhite dropouts.

That result and the Heckman review are not in direct conflict, because they answer different questions. Tyler and colleagues estimate an effect at the margin for low scoring candidates who wanted the credential. The review notes that they argue their estimate is consistent with Cameron and Heckman's finding for the average test taker, and that they claim high signaling benefits exist for low skilled takers at that margin while being absent for the general population of test takers. The review also reports critiques. Rubinstein, in a 2003 discussion, argued that differences in study effort across states would bias their estimate upward, although the review notes he presented no direct empirical evidence, and that if higher standards discourage low ability people from taking the test, his conclusion could be reversed. Jepsen and colleagues, using Missouri data, showed that retaking behavior creates a discontinuity at the passing standard that can distort comparisons between certifiers and dropouts.

What should a reader take from a dispute among economists? Mostly that "does the GED pay" has no single answer, because the answer depends on whom you ask about. For a person near the cut score in a state with a high standard, a signal may exist. For the typical recipient, the review finds the average gain minimal. A third strand comes from the test provider itself: the Oregon community college figures that the company offered when it lowered the standard in 2016 show lower remediation among GED passers than among diploma holders. That comparison is not adjusted for who selects into community college, so it is a description, not an estimate of what the credential does.

None of these studies measures IQ, and none attempts to. They measure earnings, wages, enrollment and persistence. A reader looking for the "IQ of GED holders" should therefore not treat these outcome studies as ability studies. The page on IQ and income covers the link between ability and earnings from the other side, and the entanglement of schooling, ability and earnings it describes is the same one that makes the GED's own labor market value so hard to isolate.

8 What Did a 2019 Study With an Actual IQ Test Find?

A 2019 study that gave the WAIS-IV to adult neurology patients found that GED holders scored about seven Full Scale IQ points below high school graduates and no higher than patients who left school without a GED. The study is by Hewitt, Cody, Marker and Loring, General Educational Development (GED) and Educational Attainment Equivalency for Demographically Adjusted Norms, in Archives of Clinical Neuropsychology in 2019. Its purpose was practical: neuropsychologists adjust test norms for years of education, and no consensus exists on how many years a GED should count for, so the authors asked whether GED holders and graduates differ on the Test of Premorbid Function (TOPF) and the WAIS-IV Full Scale IQ (FSIQ).

The authors reviewed 2,058 neuropsychology records from January 1, 2012 to December 15, 2017. They identified 205 patients who had both the TOPF and the WAIS-IV and could be placed in one of three education groups: left high school without a GED, left high school and later earned a GED, or completed high school with no further formal education. They excluded patients with evidence of cognitive decline, patients with reported learning disabilities or special education (31), and patients whose ages fell outside the range for the WAIS-IV demographic correction. The final sample was 151 people: 36 with no certificate, 31 with a GED and 84 with a high school diploma. The mean age was 40.29 years, and 73.5 percent were referred for evaluation for epilepsy surgery, with the rest referred by a general neurology service.

The results were consistent across the main measures. Group differences were significant for FSIQ (partial eta squared of 0.14) and for the TOPF (0.16). The FSIQ means in the text were 84.11 for high school graduates, 76.90 for the GED group and 74.72 for the group with no certificate, and the post hoc comparisons showed graduates above both other groups while the GED group and the no certificate group did not differ. The gap between graduates and the GED group is about 7.2 points (our arithmetic on the reported means). The pattern held for the Verbal Comprehension, Perceptual Reasoning and Working Memory indices. On the Processing Speed Index, graduates outscored the group with no certificate but did not differ from the GED group.

The authors state three limits themselves. The data are archival and retrospective. The sample includes many programmatic evaluations for possible epilepsy surgery, and clinical details such as the age at seizure onset were not recorded, so a systematic group difference cannot be ruled out. And some patients with epilepsy may have left school because of seizures rather than by choice. They write that future research is needed to see whether the findings generalize to other neurological conditions and to cognitively normal people without neurologic disease. Their conclusion is limited to the question they asked: the GED is not equivalent to a standard 12 year education when characterizing educational background for demographically adjusted norms. A study of 151 patients cannot say what the average IQ of GED holders in the general population is, and the page on how IQ tests are made explains what a representative norm sample would need that this one did not.

9 Why Can GED Holders Look Equal to Graduates in One Study and Lower in Another?

The studies differ in the measure, the timing, the sample and the question, and each difference can move the result without any of the studies being wrong. Table 1 sets the main designs side by side.

Table 1. Five designs that bear on the GED and IQ. Results are as reported by each source; none of the studies converts a GED score into an IQ.

The first difference is the measure. Heckman and colleagues use an achievement test, adjusted and estimated at two points in life; Hewitt and colleagues use a measured WAIS-IV score in adulthood. The second is timing: the review's strongest equivalence claim concerns the pre-8th-grade estimate, before the dropout decisions, while the 2019 study measures people in their forties after decades of different lives. The third is the sample. The cohorts in the review are nationally representative youth samples, the 2019 sample is a clinical one with exclusions, and every group in it has a mean FSIQ below 100, a reminder that patients referred for neurological evaluation are not a cross section of adults.

The fourth difference is schooling itself, and it is the one that most changes how to read the pattern. In a meta-analysis in Psychological Science, Ritchie and Tucker-Drob pooled 142 effect sizes from 42 data sets with over 600,000 participants, drawn from three kinds of quasiexperimental design, and found consistent evidence of beneficial effects of education on cognitive abilities of approximately 1 to 5 IQ points for an additional year, persisting across the life span. If that average applied to a person who left school two years before graduating, it would amount to about 2 to 10 points, which is our arithmetic and is shown only for scale, since individual cases vary and the meta-analysis does not study GED holders. The point is conceptual. Part of any gap between a GED holder and a graduate in adult IQ can come from the school years missed, which is exactly what the review's schooling adjustment tries to remove, and it can do so only for the achievement test it adjusts.

Selection works in the other direction. The review shows that the people who later certify were similar to graduates in estimated ability before high school, so the adult gap is not simply an ability gap that existed from the start. The honest summary is that the two sets of findings fit a single story: GED bound students began near graduates on achievement, left school earlier, accumulated fewer years of schooling and, in a clinical sample, scored lower as adults. How much of the later gap is schooling, how much is noncognitive, and how much is the clinical population remains unresolved by any study read for this page.

10 Is the GED an Achievement Test or an Ability Test, in CHC Terms?

The GED is an achievement test: it samples acquired knowledge and skill in four content areas, which in the Cattell-Horn-Carroll (CHC) framework sits mostly in the knowledge and academic skill domains, not in fluid reasoning, working memory or processing speed. The GED Testing Service pages describe subjects and formats and do not use CHC language, so the following classification is ours. Mathematical Reasoning draws on quantitative knowledge (Gq), the language arts test on reading and writing ability (Grw) and comprehension knowledge (Gc), and the social studies and science tests on general knowledge (Gc) together with reasoning applied to that content. A battery such as the one the CHC model page describes samples broad abilities that are less tied to a specific curriculum: fluid reasoning (Gf), visual processing (Gv), working memory (Gwm) and processing speed (Gs), as well as Gc and Gq.

That difference has practical consequences. An achievement test is designed to reflect instruction, so scores should rise with teaching and study, which is the purpose of preparation courses. The review reports that the median test taker in 1980 had studied for 20 hours and in 1989 for 30 hours, and that the share who studied more than 100 hours grew from 11.8 percent to 24.2 percent in that period. The authors interpret the preparation times as suggesting little or no human capital formation, but the point here is simpler: the construct is trainable in the sense that content can be learned. The page on intelligence versus knowledge and the page on fluid versus crystallized intelligence draw this boundary for IQ batteries, and the page on aptitude versus intelligence draws it for admissions tests.

The 2019 study shows what a battery can add. In the WAIS-IV indices that Hewitt and colleagues report, GED holders trailed graduates on verbal comprehension, perceptual reasoning and working memory but not on processing speed. A single GED score cannot produce that kind of profile, because it does not isolate abilities. Profiles also vary within a person for ordinary reasons, and the page on index score differences shows how large gaps between index scores are in general, which is a caution against reading a group pattern as an individual one. The page on verbal versus performance IQ covers how older reports split ability scores.

None of this makes the GED a poor test. An achievement test does what a credential needs: it shows that a person reached a defined level of skill in defined subjects. The error is asking it to answer a question about general ability, which it was not built or validated to answer. The page on IQ and academic achievement covers the meta-analytic relationship between the two, and a relationship that falls short of identity is the reason a conversion would mislead.

11 Why Do Researchers and Clinicians Treat a GED Differently From a Diploma?

Researchers and clinicians separate the GED from a diploma because lumping them together changes results, and the authors who studied it show how. Hewitt and colleagues list three common ways neuropsychologists translate a GED into years of education when adjusting test norms: counting it as the 12 years of a standard high school education, using the number of school years actually completed before the GED, or assigning it a lower fixed equivalent such as 10 years. Their hypothetical example, drawn from published demographic corrections, shows why the choice matters. A 76 year old man who left school in ninth grade and later earned a GED could land at the 4th, 5th and 8th percentiles on three tasks if the GED counts as 12 years, and at the 16th percentile on each if the corrections use 8 years. The example is the authors' illustration, not a finding, and it shows that a credential label can change an interpretation that should rest on the whole picture.

The same lumping distorts statistics. Heckman and colleagues argue that treating the GED as equivalent to a diploma gives false signals about progress. Counting GEDs as dropouts, they report, the African American male high school graduation rate in 2000 is at the same level as in 1960, and they argue that a substantial part of the measured convergence of Black and White high school attainment is fueled by prison issued GEDs. These are the authors' claims about measurement, and they concern how a credential is counted, not the ability of any group of people. They also argue that the availability of the GED induces some students to drop out. The review cites estimates from natural experiments, the 1997 increase in the minimum difficulty and California's introduction of a GED credential in 1974, which it reports as a 1.3 percent decrease in dropout rates where the passing standard rose and a 3.1 percent fall in high school graduation rates when the GED became available.

For a reader interested in IQ, the practical lesson is that any statistic labeled "high school" may or may not include GED holders, and the label hides the difference the research found. A study that groups GEDs with graduates can understate the differences in later outcomes, and a study that groups them with dropouts can overstate them. Hewitt and colleagues note that population studies often group GED recipients with graduates. The page on average IQ by education level is a useful reference for the general pattern of ability across education groups, with the caution that group averages describe groups.

It is also worth saying what these arguments do not claim. Nothing in the review says GED holders are less intelligent as people, and the authors' headline on cognition is the opposite: on an achievement test, those who certify look like graduates. Their concern is with persistence, motivation and reliability, and with the use of a credential as a policy shortcut. Where those concerns are described as traits, they are group averages from observational data, and the page on the standard deviation of 15 is a reminder that any group difference sits inside much larger variation among individuals. A GED holder can be a person at any point in the ability distribution.

12 What Do the Testing Standards Say, and What Should a Reader Do With a GED Score?

Testing standards treat a score interpretation as valid only for the use the evidence supports, so a GED score read as a measure of IQ fails that test. The Standards for Educational and Psychological Testing, published jointly in 2014 by the American Educational Research Association, the American Psychological Association and the National Council on Measurement in Education, begin with the overarching Standard 1.0: each intended score interpretation for a specified use should be clearly set out and supported with appropriate validity evidence. The comment to Standard 1.1 says it is incorrect to use the unqualified phrase "the validity of the test" and that no test permits interpretations valid for all purposes or in all situations. Standard 1.2 asks for a rationale for each intended interpretation for a given use, with a summary of the evidence. A GED score has a rationale and evidence for certifying high school level skills. It has none, in any source read for this page, for ranking general ability. Standard 12.2 adds that when an educational test serves multiple purposes, evidence of validity, reliability and fairness should be provided for each intended use, and Standard 5.21 asks that the rationale and procedures for establishing cut scores be documented.

The APA Guidelines for Psychological Assessment and Evaluation, approved in March 2020, say in the rationale of Guideline 4 that validity is not a unitary property of a test and that validity evidence is evaluated in the context of settings and purposes. Guideline 8 asks psychologists to be aware of the need for an appropriate normative comparison, and Guideline 11 asks them to recognize the nature of and relationship among individual, cohort and group differences. Read together, they say what this page has argued: choose the instrument that fits the question, use the right comparison group, and do not read a group pattern as an individual result.

What should a reader do? If you hold a GED and wonder about your ability, the credential tells you nothing either way, and it never settled the question in any direction. If a clinician asks about your education, report the years you attended and the credential you hold, because the clinician decides how to use them. If you want a measured profile of abilities, an assessment by a licensed psychologist is the route for any clinical, school accommodation or legal purpose, and for admission to a high IQ society the society's own accepted tests apply. An online normed test is a different, lighter tool.

We sell a paid assessment, so treat this paragraph as a disclosure and check it against the rest of the page. ACIS reports a Full Scale IQ and six primary indices on the standard scale with a mean of 100 and a standard deviation of 15, each with a percentile and a 95 percent confidence interval, against adult norms for ages 16 to 90. Prices read on the home page on October 13, 2026 can change. All are one time payments with no subscription: Quick is 15 dollars for 6 subtests in 3 domains and takes about 45 minutes, Optimized is 30 dollars for 13 subtests in 5 domains and takes about 110 minutes, and Full Scale is 50 dollars for all 20 subtests in six domains and takes about 175 minutes. Breaks are allowed, there is a 5 day quality guarantee, a purchased form can be completed within 30 days, and the free trial needs no card. The test is online and unsupervised, it is not a clinical or diagnostic instrument, it is not for hiring, school accommodations or admission to high IQ societies, and it is offered in English only. Its Vocabulary and Information subtests and its Mathematical Achievement subtest draw on learned knowledge, so schooling can shape those scores as it shapes a GED. Its other subtests sample reasoning, spatial processing, working memory and speed. The technical manual documents the instrument, and the page on IQ versus UCAT shows the same distinction between an admissions aptitude test and an ability test for another high stakes exam.

13 Sources Behind This Page

Every figure on this page traces to one of the sources below, or is labeled as our own arithmetic. The arithmetic is ours: the 425 total minutes of the four GED tests, the 99.9th percentile of an IQ of 145 and the IQ of about 96 for the 40th percentile on a standard deviation of 15 scale, the 7.2 point gap between GED holders and graduates, and the 2 to 10 point illustration. The Heckman, Humphries and Mader review was read as the IZA discussion paper of May 2010, and the articles by Cameron and Heckman and by Heckman and Rubinstein are described here as that review reports them. Pages from the GED Testing Service and Education Week were read on October 13, 2026 and can change.

• American Educational Research Association, American Psychological Association and National Council on Measurement in Education. Standards for Educational and Psychological Testing. AERA, 2014, Standards 1.0, 1.1, 1.2, 5.21 and 12.2.
• American Psychological Association, APA Task Force on Psychological Assessment and Evaluation Guidelines. APA Guidelines for Psychological Assessment and Evaluation, approved by the APA Council of Representatives in March 2020, Guidelines 4, 8 and 11.
• Heckman J J, Humphries J E and Mader N S. The GED. IZA Discussion Paper 4975, Institute for the Study of Labor, May 2010.
• Heckman J J, Humphries J E and Mader N S. The GED. NBER Working Paper 16064, National Bureau of Economic Research, June 2010, revised December 2011; published in Hanushek E A, Machin S and Woessmann L (eds), Handbook of the Economics of Education, volume 3, North-Holland, 2011, pages 423 to 484.
• Cameron S V and Heckman J J. The nonequivalence of high school equivalents. Journal of Labor Economics, 1993, volume 11, issue 1, part 1, pages 1 to 47.
• Heckman J and Rubinstein Y. The importance of noncognitive skills: Lessons from the GED testing program. American Economic Review, 2001, volume 91, issue 2, pages 145 to 149.
• Tyler J H, Murnane R J and Willett J B. Estimating the labor market signaling value of the GED. The Quarterly Journal of Economics, 2000, volume 115, issue 2, pages 431 to 468; working paper abstract: NBER Working Paper 6391, February 1998.
• Hewitt K C, Cody M W, Marker C D and Loring D W. General Educational Development (GED) and educational attainment equivalency for demographically adjusted norms. Archives of Clinical Neuropsychology, 2019, volume 34, issue 8, pages 1340 to 1345.
• 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.
• GED Testing Service. The GED test: subjects, times and question formats. ged.com, read on October 13, 2026.
• GED Testing Service. GED test scores: passing score and score levels. ged.com, read on October 13, 2026.
• Faucette D, GED Testing Service. Interpreting GED test scores for placement. Presentation, July 20, 2023.
• Gewertz C. Passing score lowered on new GED exam. Education Week, January 26, 2016.

14 Frequently Asked Questions