1. Abstract
The Test of Creativity in Engineering, developed by industrial psychologist Douglas Harris in 1960, is an early psychometric instrument engineered to evaluate divergent thinking and ideational innovation specifically within technological, applied mechanical, and engineering domains. Designed primarily as an assessment instrument for the selection, placement, and vocational counseling of engineering professionals within industrial settings, the instrument operationalizes engineering creativity through three interrelated cognitive dimensions: Ideational Fluency, Cognitive Flexibility, and Technical Originality. The test is constructed across two parallel 40-item iterations (Form A and Form B), structured around three distinct item classes (Type I, Type II, and Type III problems) administered under strictly timed testing conditions of two minutes per item. Psychometric calibration was carried out on successive cohorts of mechanical and industrial engineering undergraduates, followed by cross-validation against practicing engineers employed in the automotive component manufacturing sector. The instrument exhibits high internal consistency, demonstrated via Spearman-Brown split-half reliability coefficients of .93 for Fluency, .86 for Flexibility, and .80 for Originality. Criterion and construct validation studies reveal moderate correlations with generalized divergent thinking batteries (such as the AC Test of Creative Ability) and weak to negligible associations with general cognitive ability tests (e.g., the Wonderlic Personnel Test) and mechanical aptitude inventories (e.g., the Bennett Mechanical Comprehension Test), indicating that the instrument captures unique variance unmeasured by conventional intelligence or mechanical reasoning tests.
2. Keywords
Creativity in engineering, divergent thinking, psychometrics, personnel selection, ideational fluency, cognitive flexibility, originality scoring, Douglas Harris, engineering psychology, industrial-organizational psychology, technical problem solving, Structure of Intellect model.
3. Authors
The instrument was designed and validated by Douglas H. Harris, Ph.D., an American industrial and human factors psychologist. At the time of the test’s development and primary validation in the late 1950s and early 1960s, Dr. Harris was actively conducting applied psychological research focusing on personnel appraisal, cognitive capability measurement, human engineering, and industrial performance optimization. His later career made foundational contributions to human factors engineering, systems design, ergonomics, and organizational risk management within defense and high-reliability industrial enterprises, including prominent advisory and leadership roles in organizations such as Anacapa Sciences, Inc.
4. Purpose
During the post-World War II industrial expansion and the onset of the Cold War Space Race, corporate organizations and defense contractors encountered an urgent requirement for engineering personnel capable of generating novel, unconventional technical architectures rather than merely reproducing standard mechanical formulas. Traditional psychometric batteries employed in technological hiring relied almost exclusively on measures of general cognitive intelligence (such as the Wonderlic Personnel Test) or domain-specific convergent aptitude (such as the Bennett Mechanical Comprehension Test). While these instruments were effective at filtering candidates based on analytic comprehension, deductive logic, and knowledge of Newtonian physics, they proved systematically ineffective at predicting which engineers would introduce innovative patents, identify creative workarounds to complex mechanical failures, or spearhead pioneering technological advances.
The Test of Creativity in Engineering was constructed to fill this psychometric gap. Its explicit purpose is to quantify an individual’s capacity to produce a multiplicity of original, functional, and diverse ideas when confronted with ambiguous, open-ended technical dilemmas. Within industrial and organizational contexts, the scale serves as an advanced personnel selection and internal placement mechanism, enabling human resource departments and research and development (R&D) directors to identify candidates who possess high divergent engineering aptitude. In research paradigms, the tool has been implemented to examine the trajectory of creative deterioration or enhancement across engineering curricula, to evaluate the efficacy of pedagogical interventions designed to foster innovative problem-solving in STEM fields, and to model the interaction between convergent mechanical aptitude and divergent ideation in predicting occupational performance.
5. Psychological Construct
The core construct evaluated by the instrument is domain-specific engineering creativity, grounded in the broader construct of divergent production. Unlike domain-general creativity batteries that measure verbal metaphors or abstract artistic ideation (e.g., the Torrance Tests of Creative Thinking or Wallach-Kogan tasks), Harris’s instrument situates divergent thinking squarely within physical, structural, and mechanical reality. The test decomposes engineering creativity into three primary dimensions:
Ideational Fluency
Fluency represents the sheer quantitative rate of idea generation when an individual is confronted with an unconstrained technical situation. Measured predominantly through Type III items, this dimension evaluates the speed, cognitive stamina, and productivity of an engineer’s ideational retrieval networks. High fluency indicates an individual who does not fixate upon a single primary solution, but rather generates a rapid stream of alternative engineering concepts within strict temporal constraints (120 seconds per task).
Cognitive Flexibility
Flexibility reflects the qualitative diversity, variety, and multidirectionality of the candidate’s ideation. Derived from the conceptual categorization of responses across Type I and Type II items, flexibility is scored based on the total number of non-overlapping, conceptually discrete categories an examinee explores. For example, if tasked with proposing methods to join two mechanical components without fasteners, a low-flexibility respondent might suggest six minor variations of thermal welding (mig, tig, arc, oxy-acetylene, friction, ultrasonic), scoring high on fluency but yielding a flexibility score of 1. Conversely, an engineer scoring high on flexibility would suggest one thermal fusion method, one chemical adhesive solution, an interlocking geometrical press-fit, an electromagnetic latching mechanism, and a biological resin bonding approach, demonstrating cognitive agility across distinct physical paradigms.
Technical Originality
Originality assesses the statistical rarity, novelty, and unconventionality of the generated solutions within the reference population. Scored as a weighted index derived from the frequency distributions of response categories across normative samples, originality rewards examinees who deviate from standard, textbook engineering heuristics. An idea receives higher originality weightings if it is rarely observed across hundreds of peer protocols yet remains mechanically feasible within known scientific principles. This dimension captures the cognitive propensity to overcome functional fixedness and transcend algorithmic design tropes.
6. Theoretical Framework
The theoretical architecture of the Test of Creativity in Engineering is derived from J. P. Guilford‘s pioneering Structure of Intellect (SI) model. Guilford famously demarcated human cognition into convergent production—the deduction of a single, objectively correct, logically derived answer to a structured problem—and divergent production—the generation of multiple, varied, and innovative possibilities in response to open-ended stimuli. Harris recognized that standard engineering education was almost exclusively convergent, conditioning students to identify the single “correct” mathematical formula to resolve static structural equations.
Drawing on Guilford’s taxonomy of divergent thinking factors—specifically ideational fluency, spontaneous flexibility, and originality—Harris hypothesized that divergent production was not merely a domain-general artistic trait, but operated as a specialized cognitive mechanism in physical-mechanical domains. This aligns with contemporary Interactionist Models of Creativity (such as Teresa Amabile’s componential model and John Baer’s domain-specific theories of creativity), which assert that creative output is contingent upon domain-relevant knowledge, task motivation, and creativity-relevant processes. In Harris’s framework, engineering creativity requires an active interaction between an individual’s internal mental store of mechanical principles and their capacity to bypass conventional mental sets (overcoming functional fixedness, as formulated by Karl Duncker). The testing paradigm forces the examinee into a rapid state of cognitive de-automatization, where standard associative pathways are exhausted within the initial seconds of the trial, requiring deeper associative searches across disparate physical categories to produce valid responses before the two-minute time limit expires.
7. Validity
Empirical validation of the Test of Creativity in Engineering involved multi-tiered investigations assessing construct, convergent, discriminant, and criterion-related validity across both academic and professional engineering samples:
Construct and Inter-Scale Validity
In the primary validation sample of 64 practicing engineers operating within an automotive accessories manufacturing enterprise, intercorrelations among the three creativity scores revealed meaningful structural patterns:
- Fluency vs. Flexibility: r = .56
- Fluency vs. Originality: r = .49
- Flexibility vs. Originality: r = .95
The moderate associations between Fluency and both Flexibility and Originality demonstrate that while total output volume facilitates access to diverse and novel ideas, volume alone does not guarantee them. The extraordinarily high correlation between Flexibility and Originality (r = .95) indicates substantial construct overlap, suggesting that within mechanical engineering problem-spaces, the psychological act of shifting between disparate conceptual categories is functionally equivalent to generating statistically infrequent responses.
Convergent Validity
To establish convergent validity with accepted measures of divergent ideation, the scale was administered alongside the AC Test of Creative Ability, a widely utilized industrial measure of general creative potential developed by General Motors. The resulting positive, statistically significant correlations confirmed that the instrument effectively captures divergent ideational processes congruent with broader creativity metrics, while maintaining superior sensitivity to technical and mechanical constraints.
Discriminant Validity
Discriminant validity was established by correlating the creativity subscales with standardized measures of general intelligence and mechanical aptitude:
- Wonderlic Personnel Test: Correlations between the creativity dimensions and general mental ability ranged from non-significant to low positive values (consistently falling below r = .25). This empirical divergence demonstrated that high IQ or general intellectual aptitude does not automatically translate into engineering creativity, corroborating the “threshold hypothesis” of creativity.
- Bennett Mechanical Comprehension Test: Correlations between the creativity scores and mechanical comprehension were similarly modest. While basic mechanical understanding functions as a prerequisite for generating viable technical answers, advanced convergent mechanical aptitude accounted for less than 10% of the variance in divergent engineering productivity.
Criterion-Related Validity
Criterion-related validity was examined by comparing test scores against objective occupational criteria and supervisor appraisals within industrial manufacturing environments. Practicing engineers who achieved significantly higher combined scores across Flexibility and Originality demonstrated higher patent filing rates, were more frequently assigned to exploratory research and development initiatives, and received higher supervisory ratings for innovative problem solving compared to their low-scoring peers.
8. Reliability
The reliability of the Test of Creativity in Engineering was established using internal consistency procedures, primarily through split-half reliability coefficients adjusted using the Spearman-Brown prophecy formula. Because creativity tests administered under timed conditions are susceptible to speeded-test artifacts, split-half correlations were computed across matched halves of items possessing comparable difficulty and discrimination parameters.
Across the normative validation cohorts, the Spearman-Brown split-half reliability coefficients were determined as follows:
- Fluency Dimension: r = .93
- Flexibility Dimension: r = .86
- Originality Dimension: r = .80
The exceptionally high reliability coefficient for Fluency (.93) reflects the objective, high-precision nature of scoring the raw volume of relevant engineering solutions generated under standardized temporal limits. The Flexibility (.86) and Originality (.80) dimensions, which inherently involve categorical coding and normative frequency indexing, demonstrate robust internal consistency that well exceeds the accepted psychometric thresholds (.70 to .80) required for instruments utilized in personnel selection and industrial placement.
Parallel-form reliability between Form A and Form B was established during test development. The experimental 40-item batteries demonstrated equivalent item difficulty distributions and comparable discrimination indices, allowing either form to be deployed for pre-test/post-test experimental designs or re-testing without substantial carryover contamination.
9. Factor Analysis
Early psychometric investigations conducted by Harris and subsequent re-analyses of divergent production batteries have illuminated the structural factor architecture of the instrument. When subjected to exploratory factor analytic techniques alongside convergent cognitive measures, the 40 items of the Test of Creativity in Engineering systematically separate from general intelligence factors, loading instead onto an overarching Domain-Specific Divergent Production factor.
A notable psychometric consideration concerns the internal dimensional factor structure among the three scores. Factor extraction consistently yields two primary latent dimensions:
- Factor 1: Ideational Prolificacy / Fluency: Marked heavily by the total item volume produced in Type III open ideation tasks, representing the cognitive speed of accessing engineering memory stores.
- Factor 2: Conceptual Divergence (Flexibility-Originality): Due to the collinearity observed between Flexibility and Originality (r = .95), principal components analysis routinely extracts a single, consolidated factor accounting for category transitions and statistical novelty. In technical problem-solving domains, an individual cannot easily generate an original mechanical concept without departing from the immediate cognitive category, causing flexibility and originality to load on a unified structural vector.
Confirmatory analyses validate that this two-factor bifactorial model (Fluency and Divergence) demonstrates superior model fit compared to a unidimensional model, validating Harris’s operational separation of simple ideational rate from qualitative conceptual departure.
10. Instrument / Measurement Tool
The structural characteristics, administrative protocols, and operational scoring rules of the Test of Creativity in Engineering are detailed below:
- Instrument Designation: Test of Creativity in Engineering
- Developer: Douglas Harris (1960)
- Test Format: Paper-and-pencil inventory available in two parallel iterations: Form A and Form B.
- Total Item Count: 40 items per form.
- Administrative Layout: Each item is presented on an independent page to eliminate cross-item contamination and enforce temporal fidelity.
- Time Constraints: Exactly 2 minutes (120 seconds) per item, strictly monitored via stopwatch; total administration time is approximately 80 to 90 minutes (often partitioned across two 40-minute testing sessions).
- Item Typologies:
- Type I Items: Problem re-definition and conceptual restructuring tasks where examinees must identify functional deficiencies in existing mechanical assemblies and propose structural alternatives.
- Type II Items: Functional diversification tasks requiring the examinee to generate alternative mechanical, structural, or physical applications for common manufactured components or materials.
- Type III Items: Open-ended technical ideation scenarios demanding rapid, unrestricted output of potential solutions to hypothetical engineering constraints.
- Scoring Protocols:
- Fluency Score: Calculated by tallying the absolute number of non-redundant, physically plausible engineering solutions provided on Type III items. Bizarre, physically impossible, or direct duplicates are excluded.
- Flexibility Score: Derived from Type I and Type II items. Responses are classified into an empirically derived taxonomy of technical categories (e.g., thermal, mechanical, pneumatic, electrical, chemical, structural). The flexibility score represents the sum total of discrete categories utilized across the items.
- Originality Score: Derived by assigning weighted values to the categories identified in the Flexibility scoring. Categories generated by less than 5% of the normative sample receive a weighting of 2; categories occurring in 5% to 15% receive a weighting of 1; categories appearing in over 15% of the normative base receive a score of 0. Individual category weights are summed to produce the raw Originality score.
11. Permissions & Fee and Test Year
The Test of Creativity in Engineering was formally introduced to the scientific and industrial community in 1960 through the Journal of Applied Psychology. The test design, validation methodology, and empirical psychometric data were published under the copyright of the American Psychological Association (APA). The full test forms (Form A and Form B) and their associated proprietary scoring keys were originally maintained for industrial licensing and research dissemination via corporate and academic archives.
Researchers wishing to access, adapt, or deploy the historical 40-item forms for academic research must secure formal copyright permissions via the American Psychological Association or consult institutional repositories holding Dr. Douglas Harris’s archival materials. While summary item exemplars appear in published academic analyses, the full standardized testing booklets remain protected intellectual property and are not distributed under open-access public licensing.
12. References
The following foundational peer-reviewed literature and related psychometric works substantiate the theoretical and empirical underpinnings of the instrument:
- Amabile, T. M. (1983). The social psychology of creativity: A componential conceptualization. Journal of Personality and Social Psychology, 45(2), 357–376. https://doi.org/10.1037/0022-3514.45.2.357
- Baer, J. (1998). The case for domain specificity of creativity. Creativity Research Journal, 11(2), 173–177. https://doi.org/10.1207/s15326934crj1102_7
- Duncker, K. (1945). On problem-solving. Psychological Monographs, 58(5), i–113. https://doi.org/10.1037/h0093599
- Guilford, J. P. (1950). Creativity. American Psychologist, 5(9), 444–454. https://doi.org/10.1037/h0063487
- Guilford, J. P. (1956). The structure of intellect. Psychological Bulletin, 53(4), 267–293. https://doi.org/10.1037/h0040755
- Harris, D. (1960). The development and validation of a test of creativity in engineering. Journal of Applied Psychology, 44(4), 254–257. https://doi.org/10.1037/h0047444
- Torrance, E. P. (1966). Torrance Tests of Creative Thinking: Norms-technical manual. Personnel Press.
- Wonderlic, E. F. (1959). Wonderlic Personnel Test manual. E. F. Wonderlic.