What was George Land’s original methodology and sample for the creativity tests attributed to NASA?
This fact-check may be outdated. Consider refreshing it to get the most current information.
Executive summary
George Land developed a divergent-thinking assessment originally used to help identify creative engineers and scientists for NASA; the instrument asked participants to generate multiple solutions to problems and was designed to measure divergent thinking rather than IQ or domain expertise [1] [2]. Land and collaborator Beth Jarman later applied that same test to large samples of children and adults—most widely reported as 1,600 children aged about 4–5 drawn from Head Start cohorts—and reported dramatic declines in “creative genius” scores with age, but the details and provenance of the NASA commission and the sampling have been inconsistently reported across secondary sources [3] [4] [5].
1. Origins: a practical test for NASA’s problem-solvers
Multiple accounts state NASA commissioned George Land (with Beth Jarman later credited) to create a tool to distinguish highly creative engineers and scientists, and that the instrument was therefore built with practical selection needs in mind rather than as a laboratory measure of general creativity [1] [6] [2]. Sources characterize Land as a “research scientist” who devised a test to identify imaginative problem-solving among technical staff for organizational use; those descriptions form the basis of the popular story that the test was “the NASA test” [1] [7].
2. Methodology: a divergent‑thinking task, scored for fluency and originality
Accounts agree the test measured divergent thinking—the capacity to produce many different ideas or uses for an object or problem—by asking respondents to generate multiple solutions or novel uses, with scoring focused on quantity and novelty of responses rather than convergent correctness [1] [2]. Several write-ups describe the tasks as “come up with as many ideas as possible” prompts (e.g., uses for a fork) and indicate scoring thresholds that Land labeled as “genius” levels, although publicly available summaries do not reproduce the full scoring rubric or psychometric details [8] [2].
3. Sample: the oft‑cited 1,600 children and Head Start linkage
The most common claim across reports is that Land administered the test to roughly 1,600 children aged about 4–5—frequently described as Head Start participants—and later re-tested the same cohort at ages 10 and 15, yielding steep declines in high scorers [3] [4] [5]. Many secondary sources repeat “1,600 children, ages 4–5” verbatim [3] [8] [9], and some explicitly link the sample to Head Start enrollment [4] [5], but the available reporting here does not include the original study documentation to confirm sampling procedures, representativeness, or attrition in the follow‑ups.
4. Longitudinal claims and the headline numbers
Land’s popular narrative reported that about 98% of the 4–5 year olds scored at a “creative genius” level, that this fell to roughly 30% at age 10 and 12% at 15, and that only about 2% of adults scored at that level—figures widely reproduced in TED talks and organizational literature [8] [10] [5]. Those dramatic ratios are central to the “we are born creative geniuses” meme, but the sources available are secondary retellings that summarize Land’s claims rather than publishing original data, statistical methods, or the exact scoring cutoffs used in those categorical claims [8] [11].
5. Controversies, limits, and what the reporting omits
Reporting assembled here flags consistent gaps: the original test items, scoring rubric, psychometric validation, sampling frame, and raw longitudinal data are not presented in the secondary sources, so independent verification of the headline percentages is not possible from these accounts alone [1] [12] [4]. Some commentators advise skepticism, noting later creativity research finds different lifespan patterns and that Land’s strong conclusions about education causing creativity loss rest on correlational inferences rather than controlled experiments—points acknowledged in critiques that call for caution in treating the story as definitive [13] [4].
6. Bottom line: what can be stated with confidence
It is well documented in multiple summaries that Land’s instrument measured divergent thinking and that he applied it first with NASA and later to large samples of children and adults, reporting steep declines in high scorers with age; however, the concrete methodological details (precise test items, scoring algorithm, sampling procedures, and original data) are not available in the sources reviewed here, so the exactness of the oft‑cited percentages cannot be independently confirmed from this reporting alone [1] [3] [4]. Readers seeking definitive validation should consult Land & Jarman’s original publications or archived NASA records, which these secondary summaries reference but do not reproduce [12].