Frontiers of Architectural Research (2015) 4, 318–329

H O S T E D B Y Available online at www.sciencedirect.com

http://dx.doi.2095-2635/& 2(http://creativ

nCorrespondE-mail addPeer review

www.elsevier.com/locate/foar

RESEARCH ARTICLE

Does the smartest designer design better?Effect of intelligence quotient on students’design skills in architectural design studio

Sajjad Nazidizajin, Ana Tomé, Francisco Regateiro

CERIS, DECivil, Instituto Superior Técnico, Universidade de Lisboa, Lisboa 1049-001, Portugal

Received 14 October 2014; received in revised form 5 August 2015; accepted 13 August 2015

KEYWORDSArchitectural designstudio;Intelligence quotient(IQ);Design education;Human factors;Design thinking

org/10.1016/j.foar.2015015 The Authors. Produecommons.org/licenses

ing author. Tel.: +35121ress: sajjad.nazidizaji@under responsibility of

AbstractUnderstanding the cognitive processes of the human mind is necessary to further learn about designthinking processes. Cognitive studies are also significant in the research about design studio. The aimof this study is to examine the effect of designers intelligence quotient (IQ) on their designs.The statistical population in this study consisted of all Deylaman Institute of Higher Educationarchitecture graduate students enrolled in 2011. Sixty of these students were selected via simplerandom sampling based on the finite population sample size calculation formula. The students’ IQwas measured using Raven’s Progressive Matrices. The students’ scores in Architecture Design Studio(ADS) courses from first grade (ADS-1) to fifth grade (ADS-5) and the mean scores of the designcourses were used in determining the students’ design ability. Inferential statistics, as well ascorrelation analysis and mean comparison test for independent samples with SPSS, were alsoemployed to analyze the research data.Results indicated that the students’ IQ, ADS-1 to ADS-4 scores, and the mean scores of the students’design courses were not significantly correlated. By contrast, the students’ IQ and ADS-5 scores weresignificantly correlated. As the complexity of the design problem and designers’ experienceincreased, the effect of IQ on design seemingly intensified.& 2015 The Authors. Production and hosting by Elsevier B.V. This is an open access article under theCC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

.08.002ction and hosting by Elsevier B.V./by-nc-nd/4.0/).

8418344.ist.utl.pt (S. Nazidizaji).Southeast University.

1. Introduction

Design studio is considered the core of the design curriculum(Demirbaş and Demirkan, 2003). Researchers have describeddesign studio as the center of architecture education (Schön,1985; Ochsner, 2000; Vyas et al., 2013). Starting from an ill-defined problem (Schön, 1983), the development of ideas

This is an open access article under the CC BY-NC-ND license

319Effect of intelligence quotient on students’ design skills

and solutions is evaluated through different types of critique(Oh et al., 2013). These procedures are common in all designstudios. Social interaction and interpersonal interactionsamong design studio participants, including student–studentand student–tutor, are significant. The importance of colla-boration (Vyas et al., 2013), teamwork, and decision makingin design studio has been studied as well (Yang, 2010).Architecture students should also develop a set of designthinking (Dorst 2011) and creative skills (Demirkan andAfacan, 2012), which are increasingly prioritized in work-places and in society as a whole. A set of problem solvingskills is among the abilities that design studio studentsrequired to manage a growing spectrum of new complexranges of problems and situations caused by societal changesin students’ future careers. Learning theories in design studiohave also been discussed (Demirbaş and Demirkan, 2007).

Considering the importance of design thinking in thedesign process (Dorst, 2011) and in design studio (Oxman,2004), researchers have emphasized the necessity to under-stand the cognitive processes of the human mind to enhancethe understanding of design thinking (Oxman, 1996, Nguyenand Zeng, 2012) and to view design as a high-level cognitiveability. Design cognition studies are conducted via experi-mental and empirical methods (Alexiou et al., 2009).

According to Gregory and Zangwill (1987), “Design gen-erally implies the action of intentional intelligence”. Mean-while, Cross (1999) introduced the natural intelligenceconcept in design with the assumption that design itself isa special type of intelligence. Papamichael and Protzen(1993)discussed the limits of intelligence in design. Otherstudies emphasized the significance of spatial ability as atype of intelligence in graphic-based courses (Potter andvan der Merwe, 2001; Sorby, 2005; Sutton and Williams,2010a). Furthermore, Allison (2008) concluded that spatialability is crucial in learning and problem solving.

However, effective and measurable predictive mentalfactors, and tools that can influence the design process indesign studio are insufficiently studied.

Raven’s Matrices tests are originally developed to mea-sure the “eduction” (from the Latin word educere, whichmeans “to draw out”) of relations(Mackintosh and Bennett,2005); moreover, these tests are some of the best indicatorsof the g factor (Snow and Kyllonen, 1984, Kunda et al..2013). The g factor assesses the positive correlations amongvarious cognitive abilities and implies that individual per-formances on a certain type of cognitive task could becompared with those on other types of cognitive tasks(Kamphaus et al., 1997).

Raven tests directly measure two major elements of thegeneral cognitive ability (g), namely, (1) “eductive ability”,which is the capacity to “make meaning out of confusion”,easing the manner of dealing with complexity; and (2)“reproductive ability”, which is the capacity to process,remember, and recreate explicit information, and thosewho communicate interpersonally(Raven, 2000).

Raven tests have been extensively applied in researchand in practice, and a vast “pool of data” has been accu-mulated thus so far (Raven, 2000).Given the independenceof language skills in Raven tests, the three versions of thesetests (Advanced, Colored, and Standard ProgressiveMatrices) have been among the most widely applied intelli-gence tests(Brouwers et al., 2009).

The current study reflects a hypothesis of the correlationbetween students’ intelligence quotient (IQ) and designabilities in architectural design studio. The IQ indicator isbased on Raven’s Progressive Matrices applied to the sampleof Deylaman Institute of Higher Education architecturestudents enrolled in 2011. The architecture design skillindicator is obtained according to scores during the firstyear of Architecture Design Studio (ADS-1) to the final year(ADS-5). This study initially considers a theoretical frame-work that includes six components, namely, (1) a designstudio in architecture education; (2) design thinking indesign studio; (3) a cognitive approach in design; (4) spatialability and design studio; (5) design, problem solving and IQ;and (6) creativity, design, and IQ. Subsequently, hypothesesare formed. Descriptive and inferential statistics are empl-oyed to test the hypotheses using the SPSS software.

2. Theoretical framework

2.1. Design studio in architecture education

According to the “learning by doing” philosophy(Schön,1983), design studio is widely recognized as an indispensa-ble component of the design curriculum(Shih et al., 2006)and as the heart of architectural education(Oh et al., 2013).

Demirbaş and Demirkan (2007) regarded design studio asthe core of the design curriculum, and noted that all othercourses in the curriculum should be related to design studio.Demirbaş and Demirkan (2003) contended that design studiois related to design problems sociologically and to designeducation relations with other disciplines epistemologically.

By bridging mental and social abilities, Rüedi (1996)viewed design as a “mediator” between invention (mentalactivity) and realization (social activity). Design is an open-ended problem-solving process, and the functions of des-ign theories support designers’ cognitive abilities (Verma,1997). Hence, design studio helps in the free exchange ofideas (Tate, 1987)through an information process that maybe assumed as a social and organizational method for bothtutors and students (Iivari and Hirschheim, 1996).

Regarding the significance of designers’ experiencescompared with regulations and facts (Demirkan, 1998), adesign studio in architectural education is the first environ-ment where the initial experiences for future professionscan be obtained (Demirbaş and Demirkan, 2003).

Schön (1985) concluded that the design studio learningprocess starts with ill-defined problems and is developedthrough the “reflection-in-action” approach. In design stu-dio, the knowledge learnt in different courses should beapplied in the design process to determine an optimalsolution for the design of an ill-defined problem. In designeducation, teaching and learning methods are intended tobalance critical awareness and the creative process(Demirbaş and Demirkan, 2007). Schön (1983) also empha-sized that the studio-based learning and teaching methodcan be extended to other professional educations in otherdisciplines. In design studio, students communicate withone another, and receive comments from other students anda tutor (Kvan and Jia, 2005), which is a process calledcritique. Oh et al. (2013) reviewed different types of criti-ques in design studio.

Figure 1 Connections between cognitive approaches of design and intelligence approaches.

S. Nazidizaji et al.320

Aside from the importance of the cognitive ability ofdesign studio students, the other influential factors indesign studio are social and interpersonal communication(Cross and Cross, 1995), encounter with open-ended pro-blems (Schön, 1985), and collaborative design (Vyas et al.,2013). Oxman (2004) introduced the concept of think mapsto teach design thinking in design education. Additionally,the significance of emotion and motivation was consideredby Benavides et al. (2010). Demirkan and Afacan (2012)analyzed creativity factors in design studio. The majorfactors that influenced design studio performance werereviewed in our previous work (Nazidizaji et al., 2014),including social interaction and collaboration. The impor-tance of emotional intelligence for design studio studentswas also investigated.

2.2. Design thinking in design studio

After Rowe (1987) used the term “design thinking” in his 1987book, the term has been widely used and has been a part ofthe collective "consciousness of design researchers” (Dorst,2011). Design thinking has received increasing atte-ntion and popularity in the research about the cognitiveaspects of design as a base for design education (Oxman,2004), and has been considered a new paradigm for addressingdesign problems in different disciplines (Dorst, 2011).

Oxman (1995) classified different types of design thinkingstudies into seven categories, namely, (l) design methodol-ogy; (2) design cognition; (3) design for problem solving;(4) psychological aspects of mental activities in design;(5) collaboration, which is the social and educational aspectof design; (6) artificial intelligence in design; and (7) com-putational methods, models, systems, and technology.

Oxman (2001) suggested that the cognitive aspect ofdesign thinking should be regarded as a key educationalobjective in design education. The two major broad direc-tions of this study are experimental and empirical appr-oaches. Empirical approaches that include protocol analysisin certain special design processes are repeatedly applied.These studies are normally related to the clarification ofthinking processes in specific activities that formulateproblems and generate solutions (Cross, 2001).

Schön (1985) highlighted the importance of design think-ing. He also emphasized the significance of empiricalresearch and cognitive studies in improving design peda-gogy. In investigations on design teaching, cognitive studies

are significant because these studies encourage a clearapproach in the design pedagogy development (Schön,1985) (Eastman et al. 2001).

2.3. Cognitive approach in design

Design is typically regarded as a high-level cognitive ability, andnumerous computational and empirical studies have focused ondesign cognition (Alexiou et al., 2009). Oxman (1996)concludedthat the potential importance of the relationship betweencognition and design has received increasing attention amongdesign investigators. Design has been viewed as probably oneof the most intelligent human behaviors. Design has a solidconnection to cognition. Cognition is the study about all forms ofhuman intelligence, including vision, perception, memory,action, language, and reasoning. Expanding the knowledge abouthuman cognitive processes is necessary for understanding thenature of the mind, and consequently the nature of designthinking. In other studies, Nguyen and Zeng (2012) emphasizedthe importance of understanding design cognitive activities todevelop design technologies and provide an effective design.Additionally, studies about creativity and design cognition haveconcluded that different types of design cognition in the designprocess affect the outcomes of both low and high creativity (Lu,2015).

Kim and Maher (2008) considered the following two majorapproaches for the cognitive approach of design:

1)

The symbolic information-processing approach (SIP), whichwas introduced by Simon, emphasizes the rational problem-solving process of designers, with more attention providedto both design problems and designers (Eastman, 1969; Akin,1990; Goel, 1994).

2)

The situativity approach (SIT), which was introduced bySchon, focuses on the situational context of designers andon the environment of designers (Schön, 1983; Bucciarelli,1984).

Figure 1 shows the connections between the ideas ofcognitive approaches of design with intelligence appr-oaches.

Certain cognitive tests are available, such as the spatialability test, which measure the spatial cognition of des-igners.

321Effect of intelligence quotient on students’ design skills

2.4. Spatial ability and design studio

As a cognitive ability, spatial ability is one of the mostsignificant components related to designers. The idea of spatialability denotes a sophisticated process that designers exten-sively employ in the design activity (Sutton and Williams,2010b). Spatial ability in the design field is vital for bothproblem solving and learning, regardless if a problem is notparticularly spatial (Allison, 2008). This feature shows thatspatial ability is significant in design education. Spatial abilitymay be defined as “the ability to generate, retain, retrieve,and transform well-structured visual images” (Lohman, 1996).Additionally, spatial ability has been defined as “the ability tounderstand the relationships among different positions in spaceor imagined movements of two- and three-dimensionalobjects” (Clements, 1998).

The literature (Potter and van der Merwe, 2001; Sorby,2005; Sutton and Williams, 2010) shows the significance ofspatial ability in graphics-based courses, and the implica-tions of poor skills on career choices and success rates.Spatial cognition for designers transpires by constructinginternal or external representations, in which the repre-sentations can serve as cognitive support to informationprocessing and memory (Tversky, 2005).

According to Schweizer et al. (2007), certain evidenceproposes that the performance required to complete Raven’sMatrices test also depends on spatial ability somehow. Suttonand Williams (2007) defined spatial ability as the performanceon tasks that require three aspects, namely, (1) the mentalrotation of objects; (2) ability to understand how objectsappear at different angles; and (3) understanding of howobjects relate to one another in space. Sutton and Williams(2007) introduce delight tests for measuring spatial cognitions;one of these tests is Raven’s Matrices test. The authorscontended that the Raven test does not strictly measurespatial ability but can be considered an on verbal ability testthat recognizes the forms of spatial concepts.

Moreover, Guttman (1974) pointed out the concernsrelated to the validity of Raven’s Matrices test with thegenetic analysis of spatial ability, whereas Schweizer et al.(2007) investigated the discriminant and convergent validityof Raven’s Matrices while considering spatial abilities andreasoning. Schweizer et al. (2007) reinvestigated the rela-tionship between spatial ability measured and Raven’sAdvanced Progressive Matrices; four scales that representvisualization, reasoning, closure, and mental rotation werealso applied to a sample of N=280 university students. Theresults indicated the existence of convergent validity.Meanwhile, Lohman (1996) concluded that the hierarchicalmodels of human abilities provide g statistical and logicalpriority over spatial ability measures and that Raven’sMatrices tests are some of the best measures of g.

In relation to the gender differences in spatial ability andspatial activities (Newcombe et al., 1983), assessments of thespatial nature of tasks are positively correlated with themasculinity evaluated, and with greater male participationthan that of females.

The meta-analysis results (Linn and Petersen, 1985)suggest the following points: (a) gender differences arisein certain types of spatial ability but not in other types;(b) large gender differences exist only on “mental rotation”

measures; (c) minor gender differences exist on spatialperception measures; and (d) gender differences that existcan be detected across a life span. In relation to theinfluence of age on spatial ability (Salthouse, 1987), olderadults perform at lower accuracy levels than young adultsdo in each experiment.

2.5. Design, problem solving, and intelligencequotient

Problem solving in design has attracted significant interestamong design researchers since the 1960s. Most “designmethodology” works have been influenced by the premisethat design establishes a “natural” type of problem solving.Nevertheless, this assumption has been insufficiently exp-lored (Goel, 1994).

Alexiou et al. (2009) indicated that few ambiguities existedon whether design can be a special type of problem solving or atotally distinct style of thinking; the distinguishing features ofdesign were more or less established in general.

One approach for differentiating design is related toproblem space and solution space ideas. Problem spacesignifies a collection of requirements, whereas solutionspace presents a couple of constructions that satisfy theserequirements. In problem-solving theory, “problem space isa representation of a set of possible states, a set of ‘legal’operations, as well as an evaluation function or stoppingcriteria for the problem-solving task” (e.g., Ernst andNewell, 1969; Newell and Simon, 1972).

Moreover, Resnick and Glaser (1975) argued that anessential part of intelligence is the ability to solve problemsand a careful study of the problem-solving behavior, parti-cularly many of the psychological processes that compriseintelligence. Bühner et al. (2008) reviewed ideas correlatedto problem solving and intelligence.

Although problem solving abilities have been initiallyassumed to be possibly independent from intelligence, correla-tions among these constructs have been frequently demon-strated (Rigas et al., 2002; Kröner et al., 2005; Süß, 1996).

Rigas et al. (2002) applied the Kühlhaus and NEWFIREscenarios to evaluate problem-solving performance. Thecorrelations between intelligence and problem-solvingscores (Advanced Progressive Matrices, APM, Raven, 1976)were r=0.43 (Kühlhaus) and r=0.34 (NEWFIRE) when cor-rected for attenuation.

Leutner (2002) conducted two experiments related to theeffect of domain knowledge on the correlation betweenproblem solving and intelligence, and concluded that “Withlow domain knowledge, the correlation is low; with increas-ing knowledge, the correlation increases; with furtherincreasing knowledge, the correlation decreases; finally,when the problem has become a simple task, the correlationis again low”.

2.6. Creativity, design, and intelligence quotient

Creativity in design has been discussed through the pro-blem–solution co-evolution(Dorst and Cross, 2001) andbased on the topic itself (Sarkar and Chakrabarti, 2011;Demirkan and Afacan, 2012). Creativity is vital for the

S. Nazidizaji et al.322

design of all types of artifacts. Assessing creativity canhelp recognize innovative products and designers, and canimprove both design and products (Sarkar and Chakrabarti,2011). Creativity is a natural part of the design process, whichhas been frequently categorized through a “creative leap” thatoccurs between solution and problem space (Demirkan, 2010).Given the complex nature of creativity, consensus is lackingregarding the definition of creativity that completely covers theconcept and recognizes creative solutions. Consequently, acreative “event” cannot be guaranteed to occur within thedesign process. Thus, the study on creative design seemsproblematic (Dorst and Cross, 2001).

According to Demirkan (2010), “In architectural designprocess the interaction between person, creative process andcreative product inside a creative environment should beconsidered as a total act in assessing creativity”. Hasirci andDemirkan (2003)considered the four elements of creativity(i.e., person, process, product, and environment) while select-ing two sixth-grade art rooms as the setting. The authorsconcluded that three creativity elements (person, process, andproduct) significantly influence the design process differently. Ina later study, the effects of these three creativity elementshave been analyzed by focusing on cognition phases in the crea-tive decision making of design studio students (Hasirci andDemirkan, 2007).

While criticizing terms such as “creativity test” and“measure of the creative process”, Piffer (2012) indicatedthat three specific dimensions of creativity (novelty, appro-priateness, and impact) constitute a framework that helpsdefine and measure creativity by answering if creativity canbe measured.

Squalli and Wilson (2014) argued that the terms intelli-gence, creativity, and innovation are initially assumed to bewell understood generally, but defining, assessing, andmeasuring the inter-relationships among these terms arecontroversial. The authors conducted the “first test of theintelligence–innovation hypothesis” that contributed to thecreativity–intelligence debate in the psychology literature.

Two different theories exist in terms of the relationshipbetween IQ and creativity in the history of psychologicalresearch. In the first theory, IQ and creativity belong to thesame mental processes (conjoint theory). In the secondtheory, IQ and creativity represent two separate mentalprocesses (disjoint theory). Various researchers have recom-mended some evidence since the 1950s to prove thecorrelation between creativity and intelligence. In theseprevious studies, the correlation between these two con-cepts is extremely low that distinguishing the two conceptscan be justified (Batey and Furnham, 2006). Severalresearchers contend that creativity and intelligence origi-nate from the same intellectual cognitive process, and canonly be interpreted as creativity because of the outcomes ofboth concepts. For example, a complete new object isproduced through the cognitive process. This approach iscalled the “nothing special” hypothesis (O’Hara andSternberg, 1999).

The model usually adopted in this type of research is knownas the “threshold hypothesis”. This hypothesis indicates thatcreativity requires a high IQ level. However, only having a highIQ is insufficient (Guilford, 1967). Therefore, although a positivecorrelation exists between creativity and intelligence, thiscorrelation would either disappear or lose meaning if an

individual’s IQ score is higher than 120, which is beyond thethreshold. This model is acceptable for many researchers, but isalso challenging in different cases (Heilman et al., 2003).

2.7. Architecture education in Iran

The architecture undergraduate program in Iran universitiesis no less than a four-year program. The program com-prises140 course credits, including 6 credits for the finalthesis. The courses are divided into general (20 credits),basic (29 credits), major (60 credits), and professional (27credits). Thirteen optional courses are also offered, such asethics in architecture, research methods, design processand methods, new structures, and software applications inarchitecture. Students should pass two optional courses inthe program. ADS courses comprise 5 courses (ADS-1, ADS-2,ADS-3, ADS-4, and ADS-5) with 5 credits each. ADS coursesbegin from the fourth semester (one ADS course in everysemester) (Supreme Council for Planning, 2007).

Each design studio course begins with a tutor proposing adesign project according to the current syllabus of theMinistry of Science. The students start designing by studyingsome environmental factors and architectural standards andby considering other similarly designed projects. Semestercritique sessions are held, during which students obtain theopinions of a tutor and of fellow students regarding theimprovements of the students’ design. ADS assessment isbased on the delivered project at the end of the semester(including a 3D model, floor plans, elevations and sections,and internal and external perspectives), and students’activities during critique sessions.

3. Main hypotheses

This study started from the following hypotheses:

1)

Does any relationship exist between IQ and mean scoresfor architecture design studio courses?

2)

Does any relationship exist between IQ and scores forarchitecture design studio courses 1 to 5(ADS-1 to ADS-5)?

4. Test subjects

The statistical population in this study consisted of allDeylaman Institute of Higher Education (Lahijan-IRAN)architecture students enrolled in 2011. The Department ofEducation provided a complete list of students, from whichthe sample was selected. Moreover, 184 students wereselected as the study sample. Finite population samplingwas adopted to estimate the sample size (n) as follows:

n¼NZ2

α2pq

ε2ðN�1Þ�Z2α2pq

;

where

�

P=0.5 is the estimating ratio for trait in this study (genderratio is considered the trait ratio in this population);

�

Zα/2=1.96 is the corresponding value with 95% confi-dence level in a standard normal distribution;

323Effect of intelligence quotient on students’ design skills

�

ε=0.1 is the value of allowable error; � N=184 is the statistical population size; and �

Figure 2 Illustrative progressive matrices item. The respon-dents are asked to recognize the piece required to completethe design based on the corresponding options.

n is the least sample size, where

n¼ 184� 1:96½ �2 � 0:5� 0:5

0:1ð Þ2 � 184�1ð Þþ 1:96ð Þ2 � 0:5� 0:5ffi64

This formula shows that the minimum sample size obtainedwas 64. A larger sample size of 69 was considered becausethe confidence level was increased and access to thestudent population was provided. Simple random samplingwas adopted because the sampling framework and popula-tion members were predetermined. Participation in thisresearch was voluntary. Individuals who participated in thisresearch were assured that the data would only be used forresearch purposes, and the data were analyzed collectively.

Forty-six female students (66.7%) and 23 male students(33.3%) comprised the study sample. The average age of thestudents was 23.39. The youngest student was 21 years old,whereas the oldest was 31 years old.

5. Research methodology

The approach adopted in this research was statisticalinference associated with testing statistical hypotheses.The students’ scores were prepared, collected, recorded,and subsequently classified, controlled, and analyzed usingSPSS. The students’ IQ was based on Raven’s ProgressiveMatrices test, whereas the design talent index was based onthe design course scores. The reliability and validity of theRaven test in Iran as well as the evaluation and reliability ofthe design scores are described in the subsequent sections.

5.1. Raven’s IQ test

Raven’s Progressive Matrices, either the simple form orRaven’s Matrices themselves, are classified as non-verbal IQtests employed for educational purposes. These tests areamong the most extensively used and comprehensive teststhat can be used for five-year-old children to elderly people(Kaplan and Saccuzzo, 2008). Raven test was created byJohn C. Raven in England in 1936 (Raven, 1936).

Raven’s Advanced Progressive Matrices (Figure 2) arespecific forms of Raven’s Matrices test, which are particu-larly designed to distinguish individuals whose intelligence isbeyond normal. This form of the test is designed as twodifferent sets of questions in separate booklets. The firstbooklet contains 12 questions that are solely designed todistinguish among individuals with varying levels of intelli-gence, whereas the second booklet contains 36 questionsthat clarify and more precisely distinguish individuals. Allof the questions in the second booklet are designed asrectangular matrices with three columns and rows thatcontain organized figures and images. The final cell in thismatrix is always blank. The figure contents in the eightother cells are specified based on optional and abstractrules. An individual studied should discover these rulesthrough trial and error, and subsequently guess the contentof the ninth cell based on these rules. Six to eight optionalanswers are designed for every question. In this form oftest, an individual’s ability for abstract reasoning is

evaluated, specifically the individual’s ability to solve/guessthe relationship among the components of each question,identify the fundamental rules by which the cells areconstructed, and use these rules to recognize the correctanswer (Mackintosh and Bennett, 2005).

5.2. Reliability and validity of Raven test in Iran

Raven’s Progressive Matrices are among the IQ tests whosereliability and validity are accepted to measure and evalu-ate the overall intelligence, which is the “g factor”. Anadvanced form of this test is employed as a tool formensuring the intelligence of individuals regarded as brilliantand distinguished people, and as top and gifted students.Rahmani (2006) investigated the reliability and validity ofthis test in research in which students’ intelligence wasmeasured. Rahmani obtained raw data associated withindividuals’ IQ. Moreover, 707 individuals were initiallyselected as samples from a statistical population of studentsstudying in the Khorasgan branch of Azad University (Iran)from school years 2005 to 2008. The samples were testedusing Raven’s Progressive Matrices. The results indicatedthat Raven’s Progressive Matrices were significantly reliableand valid (Po0.01). The students’ overall intelligence wasmeasured through IQ equivalents on the Wechsler intelli-gence scale, in which a mean of 100 and a standarddeviation of 15 were obtained using the standard methodto calculate scores (z scores). No significant difference(Po0.01) was observed in the raw mean scores of thefemale and male individuals. A comparison of the mean rawscores of the subjects studied, whose age range differed,showed no difference in the mean scores among individualsover 18 years old.

5.3. Students’ scores in architecture designstudio courses

The architecture design studio scores in this study wereused as indicators of students’ design competence.

S. Nazidizaji et al.324

However, debates about these issues persist, that is,whether students’ final term scores obtained from designcourse projects and design critiques represent students’actual design skills, and whether students with high scoresin design courses are likely to be better designers in thefuture. Design projects, whether professional or academic,are examined to answer issues on whether a systematicmechanism can be developed to evaluate architecturedesigns. Additionally, how the referees’ judgements areaffected by the referees’ presumptions (Carmona and Sieh,2004; Prasad, 2004) is examined. These works are designedto identify better assessment methods for design projects.Studies about the reliability of the scores of college designcourses are limited.

Table 1 presents the design course titles, purposes, andsubjects considered in the Deylaman Institute of HigherEducation in Iran for the architecture undergraduateprogram.

Considering every ADS syllabus(Supreme Council forPlanning, 2007) and some related studies, such as “expertsand novices” subject studies (Björklund, 2013; Ozkan andDogan, 2013), the following differences among ADS-1 toADS-5 can be recognized:

1)

Normally in design research, first year students areregarded as novices, whereas final year students areconsidered experts. From ADS-1 to ADS-5, studentstransition from novice to experts.

Table 1 Architecture Design Studio (ADS) course titles, subjecin Iran.

Coursetitle

Course purpose

ADS-1 Learning simple and tangible functions Paying attentioneffective factors in design

ADS-2 Learning housing concept and factors affecting it

ADS-3 Meeting various cultural, artistic, dialect and semanticdimensions with simplicity in functional system

ADS-4 Learning specific and complicated functional systems apaying attention to installations and structural system

ADS-5 Micro and macro-scale residential complex designed tocultural, climatic and economic conditions

Table 2 Results of the correlation between the ADS scores an

Hypothesis Pearson correlation

IQ scores and mean ADS scores 0.13IQ scores and ADS-1 scores 0.028IQ scores and ADS-2 scores 0.07IQ scores and ADS-3 scores �0.06IQ scores and ADS-4 scores 0.21IQ scores and ADS-5 scores 0.26

2)

ts,

to

nd

sui

d t

Based on the attention to form and function in thesyllabus, the first design projects (ADS-1 and ADS-2) areless functional, whereas the final design projects (ADS-4and ADS-5) are highly functional. The ADS-3 project ismore conceptual and artistic.

3)

The complexity of a project in terms of the number ofspaces and variations, land area, and required technol-ogy increases from ADS-1 to ADS-5.

4)

Based on the syllabus, required design constraints, suchas environmental and economic factors, and numericalstandards, increase from ADS-1 to ADS-5.

5)

The degree of being ill-defined decreases in the finaldesign projects (ADS-4 and ADS-5). A difference betweenADS-4 and ADS-5 is that ADS-5 students should deal withan urban network design for designed buildings asidefrom designing buildings.

6. Testing the main hypotheses

The existence of a significant relationship between IQ andADS courses 1 to 5 scores, and the means of these scores isquestioned.

Inferential statistical methods, including a correlationtest, were used in testing the main hypotheses. First, theexistence or absence of a significant relationship amongthese variables was studied through these methods; a 5%confidence level was considered in this study.

and purposes in the architecture undergraduate program

Design subject Landarea(m2)

Fruit market, simple fair exhibit site, smallpassenger terminal

1500

Residential units/buildings in urban contextfor an extended family.

1000

Museum, monument, cultural center,special exhibition sites

2000–3000

Small hospitals, small airports, portfacilities, nursing home for disabled people

6002

t Forty-floor residential complex based onmedium or high population density

8000

he mean of these scores with the IQ scores.

P-value correlation Type of correlation

0.26 No correlation –

0.82 No correlation –

0.56 No correlation –

0.61 No correlation –

0.07 No correlation –

0.02 Correlated Positive

Figure 3 Matrix showing the distribution between the IQ scores and the mean of scores from architecture design studio (ADS-1 to ADS-5).

Figure 4 Distributions of the scores of ADS-1 to ADS-5.

325Effect of intelligence quotient on students’ design skills

Given that the P-value of the testing correlation betweenIQ and ADS-1 to ADS-4 is more than 0.05, the null hypothesisis accepted and the correlation is insignificant for the 95%confidence level (see Table 2).

However, the P-value of the testing correlation betweenIQ and ADS-5 is 0.02, which is lower than 0.05. Therefore,the null hypothesis is rejected and these two variables arecorrelated. The correlation value is 0.26.

Figure 3 shows the mean distribution in every coursecompared with the IQ scores. Evidently, the relationshipamong the means is extremely low that this relationship canbe disregarded. Figure 4 demonstrates the relationshipsbetween IQ and the five design courses. Obviously, no correla-tions exist between IQ and ADS-1, ADS-2, ADS-3, and ADS-4.

7. Gender difference hypotheses

1)

Does a significant difference exist between femaleand male architecture students in terms of intelligencequotient scores?

2)

Does a significant difference exist between female andmale architecture students regarding the mean ADSscores of both genders?

7.1. Testing the hypotheses

H0 (null hypothesis) μ1=μ2: No significant difference existsbetween the IQ scores of female and male students.

H1 (alternative hypothesis) μ1aμ2: A significant differ-ence exists between the IQ scores of female and malestudents.

Descriptive statistics and the t-statistical test for twoindependent groups (independent t-sample test) are used inthis section to recognize the present condition of thepopulation under study. These statistics investigate thedifference between the mean IQ scores and the meanarchitecture design scores in the two student samplesexamined (i.e., female and male individuals).

Table 3 shows the data related to both IQ scores and themean scores for architecture design studio courses offemale and male students. Descriptive statistics are indivi-dually calculated for each group.

Out of the 69 students studied (46 females and 23 males),the mean IQ is 111.02 in the female student group, whereasthat in the male student group is 111.91. Moreover, thestandard deviation shown in Table 3 indicates that the IQdistribution is wider in the female student group than thatin the male student group. The mean of the architecturedesign average scores is 16.09 in the female student groupand 15.35 in the male student group. The standard devia-tion obtained implies that the distribution of the ADSaverage scores is slightly wider in the male student groupthan in the female student group. A mean comparison testshould be performed to reject or accept this hypothesis.

Two columns of the 95% confidence interval and the P-value are used to conclude whether the means differ fromor are similar to each other (significance of the difference inthe mean scores). The P-value shows whether the nullhypothesis (H0) should be rejected or accepted. This valueis compared with 0.05. First, a test for equality of variancesis performed, and a test for equality of means is

Table 3 Group statistics: IQ and ADS scores based on gender.

Gender N Mean Std. deviation Std. error mean

Intelligence quotient Female 46 111.02 13.84 2.041Male 23 111.91 10.71 2.233

Architecture design studio mean scores Female 46 16.09 1.25 0.184Male 23 16.35 1.53 0.318

S. Nazidizaji et al.326

subsequently conducted based on the test for equality ofvari-ances. If the variances are proven to be equal to each other,then the first row of means is tested for equality of means;otherwise, the second row is tested.

The P-value is 0.21 according to the test for equality ofvariances for the mean IQ variant scores; moreover, the P-value is 0.21 according to the test for equality of variancesfor the mean scores of the architecture design courses. A0.05% confidence level is taken. Given that the P-value forboth tests is lower than 0.05, the null hypothesis forequality of variances is accepted. Thus, the variances oftwo populations are considered equal for both variables.Therefore, the first row of means is considered to test everyvariable because the variances are equal (see Table 4).

According to the P-value obtained from the test forequality of means for the mean IQ scores, which is asfollows:

P-value=0/7840/05, the null hypothesis, which is theequality of means, is not rejected with the 95% confidencelevel. Therefore, no significant difference exists betweenthe mean IQ scores associated with female and malestudents.

According to the P-value obtained from the test forequality of means for the mean scores for ADS courses,which is as follows:

Sig=0/7840/05, the null hypothesis, which is equality ofmeans, is not rejected with the 95% confidence level.Therefore, no significant difference exists between themean scores for the architecture design courses associatedwith female and male students.

Two numbers shown at the 95% confidence interval of thedifference are 0, which indicates that the null hypothesis isaccepted in both comparison tests.

The box charts in Figure 5 show numerous descriptivestatistics, including the maximum and minimum data,median, quartiles, and range, along with data distributionassociated with the scores of architecture designing coursesand IQ scores in the two groups of male and femalestudents.

8. Discussion

This research demonstrates that no significant relationshipsexist between students’ IQ and variables that include (1)ADS-1 to ADS-4, and (2) mean scores for design courses.Moreover, IQ and ADS-5 scores are correlated.

According to the explained difference among ADS sylla-busses in Section 5.3, the effect of IQ on expert design skillsis better than that on novices (the P-value for IQ and ADS-4is 0.07 and is close tobeingcorrelated). Moreover, when the

complexity of a project and design constraints increase, andthe degree of being ill-defined in a design problem decr-eases, the influence of IQ on design skills becomes evident.However, the item (designer experience, complexity andfunctionality of design project, and a design problem beingless ill-defined) that has a larger effect on this correlationcannot be indentified because no continual decrease in P-value occurs from ADS-1 to ADS-5 (because of both ADS-2and an IQ P-value of 0.56).

Testing the second hypothesis about the effect of genderdifference on IQ and design scores shows that no significantdifference is observed in the IQ and design scores of malesand females in this study compared with other predictors(i.e., spatial ability) that indicate that males are betterthan females (Newcombe et al., 1983). The implication isthat if spatial ability is regarded as an indicator of designabilities, then spatial ability contradicts the results in termsof the absence of a significant difference between malesand females in the aspect of design scores.

The correlation between IQ scores and the total GPA isalso measured in this study (P-value=0.217, Pearsoncorrelation=�0.15, no correlation). The results indicatethe lack of a significant difference between the average ofdesign scores and students’ GPA.

Furthermore, the threshold theory of creativity–intelli-gence about intelligence design for an IQ above 120 ismeasured. For students with an IQ above 120, the correla-tion between IQ and the mean ADS is the Pearsoncorrelation=�0.073 and the P-value=0.79, which indicatesthe lack of a significant relationship, and threshold theoryabout creativity–intelligence is not confirmed in intelligencedesign.

9. Conclusion

This study examined the correlation between architecturestudents’ IQs and (1) the students’ ADS-1 to ADS-5scores,and (2) the mean ADS scores. The results indicated that asthe degree of complexity of a project increases, a designer’sexperience may boost the effect of IQ in the design process.As the factors that influence designers’ success are identi-fied, more productive and effective design programs may bedevised in the future. Furthermore, the guidance for thefuture occupation of students in this field can be assured byidentifying mental talents that empower design ability,particularly talents that can be quantitatively measured.

The study reported in this paper should be repeated inother architectural schools to confirm if a correlationonly exists in the final year course. Further research onthe correlation between urban design courses and IQ wouldhelp clarify this topic.

Table 4 Independent sample test results.

Levene’stest forequalityofvariances

t-Test for equality of means

F Sig. t df Sig. (2-tailed)

Meandifference

Std. errordifference

95%Confidenceinterval ofthedifference

Lower Upper

Intelligence quotient Equal variancesassumed

1.58 0.21 �0.27 67 0.78 �0.89 3.29 �7.46 5.68

Equal variancesnot assumed

�0.29 55.24 0.76 �0.89 3.02 �6.95 5.17

Mean architecture designstudio scores

Equal variancesassumed

1.27 0.26 �0.77 67 0.44 �0.26 0.343 �0.95 0.41

Equal variancesnot assumed

�0.72 37.13 0.47 �0.26 0.367 �1.01 0.47

Figure 5 Box charts of IQ and mean ADS with respect to gender.

327Effect of intelligence quotient on students’ design skills

S. Nazidizaji et al.328

The major limitation of the research approach was therelevance or accuracy of the evaluation methods in prac-tical architecture courses, specifically design courses. Cer-tain doubts that emerged in students’ scores in designcourses accurately represented the students’ actual designability. Therefore, this study encourages further research onthis issue. New studies are currently being developedthrough different assessment methods.

Other concerns that emerged were about creativityversus intelligence tests. Meanwhile, researchers haveemphasized the role of creativity in design, in which thebroad concept of creativity induced difficulties in under-standing the exact role of creativity in design. Moreover, themeasurability of creativity and creativity tests is underdebate. Certain intelligence innovation tests (Squalli andWilson, 2014) that can be used for future studies areavailable.

The present study is recommended to be repeated onlarger statistical populations and in different countries orcities. Repeating this research in a broader context andusing the new results may help design a new questionnaireor cognitive tests to identify future high potential designers.

Based on different effective design factors and the out-comes (mental, cognitive, social interaction, collaboration,personality, problem-solving skills) of these factors, everydesigned predictive test should consider all aspects or mightbe combinations of cognitive and personality tests. In termsof success in designing a reliable test or questionnaire,Cross’ (1999) theory can validate that design is a special andseparate type of intelligence.

Acknowledgement

This research has been done by financial support of PortugalCalouste Gulbenkian Foundation (Grant number 129645)(Fundação Calouste Gulbenkian). We most sincerely thankMr Dariuosh Poordadashi for his cooperation in the analysisof the statistical results and the staff of Deylaman Instituteof Higher Education for their cooperation in preparing theresearch data.

References

Akin, Ö., 1990. Necessary conditions for design expertise andcreativity. Des. Stud. 11 (2), 107–113.

Alexiou, K., Zamenopoulos, T., Johnson, J.H., Gilbert, S.J., 2009.Exploring the neurological basis of design cognition using brainimaging: some preliminary results. Des. Stud. 30 (6), 623–647.

Allison, L.N., 2008. Designerly Ways of Knowing. MIT Press,Cambridge.

Batey, M., Furnham, A., 2006. Creativity, intelligence, and person-ality: a critical review of the scattered literature. Genet. Soc.Gen. Psychol. Monogr. 132 (4), 355–429.

Benavides, F., Dumont, H., Istance, D., 2010. The Nature ofLearning. Using Research to Inspire Practice. OECD Publishing,Paris.

Björklund, T.A., 2013. Initial mental representations of designproblems: differences between experts and novices. Des. Stud.34 (2), 135–160.

Brouwers, S.A., Van de Vijver, F.J.R., Van Hemert, D.A., 2009.Variation in Raven’s progressive matrices scores across time andplace. Learn. Individ. Differ. 19 (3), 330–338.

Bucciarelli, L.L., 1984. Reflective practice in engineering design.Des. Stud. 5 (3), 185–190.

Bühner, M., Kröner, S., Ziegler, M., 2008. Working memory, visual–spatial-intelligence and their relationship to problem-solving.Intelligence 36 (6), 672–680.

Carmona, M., L. Sieh, 2004. Measuring Quality in Planning: Mana-ging the Performance Process. Routledge.

Clements, D.H., 1998. Geometric and Spatial Thinking in YoungChildren.

Cross, N., 1999. Natural intelligence in design. Des. Stud. 20 (1),25–39.

Cross, N., 2001. Design cognition: Results from protocol and otherempirical studies of design activity. Des. Knowing Learn.: Cogn.Des. Educ. 79–103.

Cross, N., Cross, A.C., 1995. Observations of teamwork and socialprocesses in design. Des. Stud. 16 (2), 143–170.

Demirbaş, O.O., Demirkan, H., 2007. Learning styles of designstudents and the relationship of academic performance andgender in design education. Learn. Instr. 17 (3), 345–359.

Demirbaş, O.O., Demirkan, H., 2003. Focus on architectural designprocess through learning styles. Des. Stud. 24 (5), 437–456.

Demirkan, H., 1998. Integration of reasoning systems in architec-tural modeling activities. Autom. Constr. 7 (2), 229–236.

Demirkan, H., 2010. From Theory to Practice – 39 Opinions.Creativity, Design and Education. Theories Positions and Chal-lenges. pp. 56–59.

Demirkan, H., Afacan, Y., 2012. Assessing creativity in designeducation: analysis of creativity factors in the first-year designstudio. Des. Stud. 33 (3), 262–278.

Dorst, K., 2011. The core of ‘design thinking’ and its application.Des. Stud. 32 (6), 521–532.

Dorst, K., Cross, N., 2001. Creativity in the design process: co-evolution of problem–solution. Des. Stud. 22 (5), 425–437.

Eastman, C., Newstetter, W., McCracken, M., 2001. Design Knowingand Learning: Cognition in Design Education. Elsevier.

Eastman, C.M., 1969. Cognitive Processes and Ill-defined Problems:A Case Study from Design.

Ernst, G.W., A. Newell, 1969. GPS: A Case Study in Generality andProblem Solving. Academic Press.

Goel, V., 1994. A comparison of design and nondesign problemspaces. Artif. Intell. Eng. 9 (1), 53–72.

Gregory, R.L., Zangwill, O.L., 1987. The Oxford Companion to theMind. Oxford University Press.

Guilford, J.P., 1967. The Nature of Human Intelligence.Guttman, R., 1974. Genetic analysis of analytical spatial ability:

Raven’s progressive matrices. Behav. Genet. 4, 273–284.Hasirci, D., Demirkan, H., 2003. Creativity in learning environ-

ments: the case of two sixth grade art rooms. J. Creative Behav.37 (1), 17–41.

Hasirci, D., Demirkan, H., 2007. Understanding the effects ofcognition in creative decision making: a creativity model forenhancing the design studio process. Creativity Res. J. 19 (2–3),259–271.

Heilman, K.M., Nadeau, S.E., Beversdorf, D.O., 2003. Creativeinnovation: possible brain mechanisms. Neurocase. 9 (5),369–379.

Iivari, J., Hirschheim, R., 1996. Analyzing information systemsdevelopment: a comparison and analysis of eight IS developmentapproaches. Inf. Syst. 21 (7), 551–575.

Kamphaus, R.W., M.D. Petoskey, A.W. Morgan, 1997. A History ofIntelligence Test Interpretation.

Kaplan, R.M., Saccuzzo, D.P., 2008. Psychological Testing: Princi-ples, Applications, and Issues. Wadsworth Publishing Company.

Kim, M.J., Maher, M.L., 2008. The impact of tangible user inter-faces on spatial cognition during collaborative design. Des. Stud.29 (3), 222–253.

Kröner, S., Plass, J.L., Leutner, D., 2005. Intelligence assessmentwith computer simulations. Intelligence 33 (4), 347–368.

329Effect of intelligence quotient on students’ design skills

Kunda, M., McGreggor, K., Goel, A.K., 2013. A computational modelfor solving problems from the Raven’s progressive matricesintelligence test using iconic visual representations. Cogn. Syst.Res. 22–23 (0), 47–66.

Kvan, T., Jia, Y., 2005. Students’ learning styles and their correla-tion with performance in architectural design studio. Des. Stud.26 (1), 19–34.

Leutner, D., 2002. The fuzzy relationship of intelligence andproblem solving in computer simulations. Comput. Hum. Behav.18 (6), 685–697.

Linn, M.C., Petersen, A.C., 1985. Emergence and characterizationof sex differences in spatial ability: a meta-analysis. Child Dev.1479–1498.

Lohman, D.F., 1996. Spatial Ability and G. Human Abilities: TheirNature and Measurement. pp. 97–116.

Lu, C.-C., 2015. The relationship between student design cognitiontypes and creative design outcomes. Des. Stud. 36, 59–76.

Mackintosh, N.J., Bennett, E.S., 2005. What do Raven’s matricesmeasure? An analysis in terms of sex differences. Intelligence 33(6), 663–674.

Nazidizaji, S., Tomé, A., Regateiro, F., 2014. Search for designintelligence: a field study on the role of emotional intelligencein architectural design studios. Front. Archit. Res. 3 (4),413–423.

Newcombe, N., Bandura, M.M., Taylor, D.G., 1983. Sex differencesin spatial ability and spatial activities. Sex Roles 9 (3), 377–386.

Newell, A., Simon, H.A., 1972. Human Problem Solving. Prentice-Hall, Englewood Cliffs, NJ.

Nguyen, T.A., Y. Zeng, 2012. Clustering designers’ mental activitiesbased on EEG power. Tools and Methods of Competitive Engi-neering, Karlsruhe.

O’Hara, L.A., Sternberg, R.J., 1999. Learning styles. Encycl.Creativity 2, 147–153.

Ochsner, J.K., 2000. Behind the mask: a psychoanalytic perspectiveon interaction in the design studio. J. Archit. Educ. 53 (4),194–206.

Oh, Y., Ishizaki, S., Gross, M.D., Do, E.Y.-L., 2013. A theoreticalframework of design critiquing in architecture studios. Des.Stud. 34 (3), 302–325.

Oxman, R., 1995. Viewpoint observing the observers: researchissues in analysing design activity. Des. Stud. 16 (2), 275–283.

Oxman, R., 1996. Cognition and design. Des. Stud. 17 (4), 337–340.Oxman, R., 2001. The mind in design: a conceptual framework for

cognition in design education. In: Eastman, C., McCracken, M.,Newstetter, W. (Eds.). Design Knowing and Learning: Cognitionin Design Education. Elsevier Science Ltd., Oxford, 269–295.

Oxman, R., 2004. Think-maps: teaching design thinking in designeducation. Des. Stud. 25 (1), 63–91.

Ozkan, O., Dogan, F., 2013. Cognitive strategies of analogicalreasoning in design: differences between expert and novicedesigners. Des. Stud. 34 (2), 161–192.

Papamichael, K., J.P. Protzen, 1993. The limits of intelligence indesign.

Piffer, D., 2012. Can creativity be measured? An attempt to clarifythe notion of creativity and general directions for futureresearch. Think. Skills Creativity 7 (3), 258–264.

Potter, C., E. van der Merwe, 2001. Spatial Ability, Visual Imageryand Academic Performance in Engineering Graphics.

Prasad, S., 2004. Clarifying intentions: the design quality indicator.Build. Res. Inf. 32 (6), 548–551.

Rahmani, J., 2006. Reliability, validity and standardization ofadvanced Raven’s progressive matrices test between studentsof Azad University Students in Iran. Sci. Res. Psychol. 34, 61–74.

Raven, J., 1936. Mental Tests Used in Genetic Studies: ThePerformance of Related Individuals on Tests Mainly Educativeand Mainly Reproductive (Unpublished Master’s thesis). Univer-sity of London.

Raven, J., 2000. The Raven’s progressive matrices: change andstability over culture and time. Cogn. Psychol. 41 (1), 1–48.

Resnick, L.B., R. Glaser, 1975. Problem Solving and Intelligence.Rigas, G., Carling, E., Brehmer, B., 2002. Reliability and validity of

performance measures in microworlds. Intelligence 30 (5),463–480.

Rowe, P.G., 1987. Design Thinking. The MIT Press, Cambridge, MA.Rüedi, K., 1996. Architectural Education and the Culture of

Simulation: history against the Grain. In: Hardy, A., Teymur,N. (Eds.), Architectural Education and the Culture of Simulation:history against the Grain, pp. 109–125.

Salthouse, T.A., 1987. Adult age differences in integrative spatialability. Psychol. Aging 2 (3), 254.

Sarkar, P., Chakrabarti, A., 2011. Assessing design creativity. Des.Stud. 32 (4), 348–383.

Schön, D.A., 1983. The Reflective Practitioner: How ProfessionalsThink in Action, Basic Books.

Schön, D.A., 1985. The Design Studio: An Exploration of itsTraditions and Potentials. RIBA Publications for RIBA BuildingIndustry Trust, London.

Schweizer, K., Goldhammer, F., Rauch, W., Moosbrugger, H., 2007.On the validity of Raven’s matrices test: does spatial abilitycontribute to performance? Personal. Individ. Differ. 43 (8),1998–2010.

Shih, S.-G., Hu, T.-P., Chen, C.-N., 2006. A game theory-basedapproach to the analysis of cooperative learning in designstudios. Des. Stud. 27 (6), 711–722.

Snow, R.E., Kyllonen, P.C., Marshalek, B., 1984. The topography ofability and learning correlations. Adv. Psychol. Hum. Intell. 2(S47), 103.

Sorby, S.A., 2005. Assessment of a “new and improved” course for thedevelopment of 3-d spatial skills. Eng. Des. Graph. J. 69 (3), 6.

Squalli, J., Wilson, K., 2014. Intelligence, creativity, and innova-tion. Intelligence 46, 250–257.

Supreme Council for Planning, S., 2007. General characteristicsProgramme and syllabus of courses – B.Sc. of Architecture.Ministry of Science Research and Technology, Tehran.

Süß, H.-M., 1996. Intelligenz, Wissen und Problemlösen. Göttingen:Hogrefe.

Sutton, K., A. Williams, 2010a. Implications of Spatial Abilities onDesign Thinking. Design & Complexity. Design Research Society,Montreal (Quebec), Canada.

Sutton, K., A. Williams, 2010b. Implications of Spatial Abilities onDesign Thinking.

Sutton, K.J., A.P. Williams, 2007. Spatial cognition and its implica-tions for design. International Association of Societies of DesignResearch, Hong Kong, China.

Tate, A., 1987. The Making of Interiors: An Introduction.HarperCollins.

Tversky, B., 2005. Functional Significance of Visuospatial Representa-tions. Handbook of Higher-level Visuospatial Thinking, pp. 1–34.

Verma, N., 1997. Design theory education: how useful is previousdesign experience? Des. Stud. 18 (1), 89–99.

Vyas, D., Van der Veer, G., Nijholt, A., 2013. Creative practices inthe design studio culture: collaboration and communication.Cogn. Technol. Work. 15 (4), 415–443.

Yang, M.C., 2010. Consensus and single leader decision-making inteams using structured design methods. Des. Stud. 31 (4),345–362.