A Virtual Steel Sculpture for Structural The Author(s) …hpcg.purdue.edu/idealab/2016 publications/2. Journal of...nected. Unlike the physical steel sculpture, this interactive virtual

Article

A Virtual SteelSculpture for StructuralEngineering Education:Development andInitial Findings

Hazar Nicholas Dib1 andNicoletta Adamo-Villani2

Abstract

We describe the development and evaluation of a virtual steel sculpture for engin-

eering education. A good connection design requires the engineer to have a solid

understanding of the mechanics and steel behavior. To help students better under-

stand various connection types, many schools have acquired steel sculptures. A steel

sculpture is a physical system that shows 48 types of connection found in standard

construction practices. Unfortunately, because of its size and location, students do

not always have easy access to it. The virtual sculpture described in the article can

provide an effective learning alternative. It allows the students to see the sculpture

from multiple points of view anywhere, anytime and shows the close-up view of each

connection with description of how it may be used, potential failure modes, sample

calculations, and field examples. Findings from a formative study with engineering

students support the pedagogical efficacy of the interactive sculpture.

Keywords

virtual learning environments, steel sculptures, engineering education, steel design,

user studies

Journal of Educational Technology

Systems

2016, Vol. 44(4) 430–449

! The Author(s) 2016

Reprints and permissions:

sagepub.com/journalsPermissions.nav

DOI: 10.1177/0047239515626704

ets.sagepub.com

1School of Construction Management Technology and Computer Graphics Technology, Purdue University,

West Lafayette, IN, USA2Computer Graphics Technology, Purdue University, West Lafayette, IN, USA

Corresponding Author:

Hazar Nicholas Dib, Purdue University, 401 N Grant Street, West Lafayette, IN 47907, USA.

Email: [email protected]

Introduction

The Accreditation Board for Engineering and Technology requires that all civilengineering graduates demonstrate that they can apply knowledge of four tech-nical areas appropriate to civil engineering. Usually structural engineering is oneof these areas. To meet the structural competency requirement, students take adesign course in steel or reinforced concrete after they have completed the struc-tural analysis course. A typical introductory level steel design course includes thefollowing topics: determination of load combinations with appropriate loadfactors; sizing of tension (axial), compression (columns), and flexural (beamsand girders) members; and design of tension connections using mechanicalfasteners as well as welds. In some cases, an instructor may be able to coveradditional topics such as shear–moment connection design. Furthermore, for atypical 15-week long semester course, approximately 1 to 2 weeks are devoted toconnection design. The lack of emphasis (about 10% of the course) on connec-tion design is by no means a reflection of its significance to the integrity of astructure. Instead, it is due to the time constraint and the common belief thatconnections are standardized details that should be left to the fabricators anddetailers. However, the connections are the glue that holds a structure together.Historically, connection failures have contributed to many structural failures,for example, the Hartford Civic Center in 1977 (Smith & Epstein, 1980), theHyatt Regency Hotel in Kansas City in 1980, (Pfrang & Marshall, 1982) andmore recently, the I-35W Bridge in Minneapolis (Holt & Hartmann, 2008).Since the Hyatt Regency failure, many state licensing boards have made theconnection design the responsibility of the engineer-of-record.

Moreover, there are situations for which standard connection types wouldnot be applicable, and engineers are required to design connections unique to astructure. A good connection design requires the engineer to have a good under-standing of mechanics and steel behavior and know the fabricator’s limitationsand experience.

Often, what may appear to be an acceptable design (theoretically), in practice,it may not be feasible to fabricate in the shop or in the field. American Instituteof Steel Construction (AISC) Code of Standard Practice (AISC, 2005) recom-mends the engineers to work closely with the fabricators and detailers whiledesigning the connections. Hence, for these reasons, it is important that stu-dents, preparing for practice in structural engineering, develop a good under-standing of connection design limitations and assembly. Moreover, because oftheir three-dimensional (3D) nature, students have more difficulty understandingthe assembly of connections than the design of major components such as beamsand columns.

The web-based interactive 3D tool (virtual sculpture) described in the articleaims to overcome some of the obstacles described earlier. Currently, to ourknowledge, there are no web-based interactive 3D tools that show in detailhow various components of a steel structure are connected. Our interactive

Dib and Adamo-Villani 431

virtual steel sculpture helps students better visualize how members are con-nected. Unlike the physical steel sculpture, this interactive virtual tool is access-ible to students and educators 24/7 in the United States and abroad. The virtualstructure also benefits other affiliated personnel such as architects, design tech-nicians (drafting people), and construction managers. Moreover, this tool canallow other educators and engineers to contribute to the database of designcalculations and field examples.

Background

Steel Connections

Steel connections have primarily been designed as two-dimensional (2D) elem-ents (x-y and z-y planes) despite the fact that their load bearing behavior is 3D.Combining two 2D designs to yield a connection that supports a 3D loadbearing behavior is a concept that is usually very difficult for students to visu-alize. In general, the connection details are shown as a series of 2D drawingssuch as the typical shop drawings that are provided by the fabricator to theengineers for review and approval. Although the shop drawings depict exactlyhow the connections are to be assembled, they are not easily understood by thestudents.

Taking students to actual construction sites is one way to help them see howsteel members are assembled. Although this is a good approach, it is a majorchallenge to find construction projects that are nearby and during the term inwhich the design course is taught. Often, liability issues also prevent the facultyfrom taking students to construction sites. Because of these issues, many facultymembers have resorted to taking photographs of connection types from con-struction sites and then showing them to their students. Unfortunately, thephotographs still do not show the true 3D nature of connections.

In the summer of 1985, after seeing the inability of his students to visualizeeven simple connections, Professor Duane Ellifritt of Florida State Universitydesigned a steel connection sculpture as a visual aid to teach his students aboutthe many ways steel members could be connected (Florida State University,College of Engineering eNews Archives, 1998). This 13-feet tall sculpture hasnow been duplicated (duplications are only 8-feet tall) and erected at over 135campuses across the United States. In addition to the physical sculpture, theAISC also has prepared a teaching guide for the connection types (Green, Sputo,& Veltri, 2008) which can be downloaded from the AISC website.

Virtual Learning Environments

In this article, the term interactive virtual learning environment is used broadly,and because of this broad usage, it can refer to a website with static web pages.

432 Journal of Educational Technology Systems 44(4)

At the other end of the continuum, an interactive virtual learning environmentcan be a highly integrated system, including the associated infrastructure, that:(a) the information space has been designed; (b) is immersive, meaning thateducational interactions occur within the environment making the environmenta “place”; (c) is a designed information space in which the information is expli-citly represented, and the educational interactions occur so that students are notonly active, but actors, that is, they coconstruct the information space; (d) over-laps and is an extension of the physical environment, thereby possessing thepotential to enrich classroom activities; and (e) can integrate heterogeneoustechnologies and multiple pedagogical approaches (Dillenburg, 2000).

Research on the educational benefits of interactive virtual environments isfairly recent and stems from the fields of computer graphics, cognitive psych-ology, visual cognition, and educational psychology. In general, these earlyresearch findings show that virtual learning tools can be more effective thantraditional teaching tools (Dalgarno & Harper, 2004; Shin, 2003; Winn, 2002).Research also shows that virtual reality (VR) technology is particularly suitableto mathematics and science education. VR technology can present abstract con-cepts in concrete terms, offers the opportunity to manipulate concrete objects,and to bridge manipulatives with other representational forms. Technologies,such as VR, can be used to create interactive learning environments wherelearners can visualize concepts easily and receive feedback to build new know-ledge and understanding (Bransford, Brown, & Cocking, 1999; Hmelo &Williams, 1998).

Computer simulations have been shown to be an effective approach toimprove student learning and have the potential to help students developmore accurate conceptions (Gorsky & Finegold, 1992; Kangassalo, 1994;Zietsman & Hewson, 1986). Research shows that the use of simulation toolsoften reinforces learning and leads to performance improvements in a variety ofdisciplines. Therefore, recently, there has been significant progress in develop-ment of computer-based interactive learning tools in many different areas.

These early findings suggest the development of virtual learning tools haspromise. In response, we are seeing the advancement of virtual learning toolprojects in engineering and science. For example, Del Alamo (Mannix, 2000), aprofessor of electrical engineering at MIT, created a web-based virtual micro-electronics lab for his students in 1998. At Johns Hopkins University, Karweit(2010) has simulated various engineering and science laboratories and instru-ments on the web. At the University of Illinois Urbana-Champaign (UIUC),researchers have developed a virtual laboratory for earthquake engineering(Smart Structures Technology Laboratory [SSTE] at UIUC, 2008). Kuo,Kang, Lu, Hsieh, and Lin (2007) have recently developed a virtual survey instru-ment (SimuSurvey) and studied the feasibility of introducing SimuSurvey inregular surveyor training courses to teach students how to operate virtual sur-veying instruments. At Purdue University Dib et al. have developed a Virtual


Learning Environment for undergraduate learning of surveying principles andprocedures (e.g., VELS; Dib & Adamo-Villani, 2011). Results of several studiesindicated improved student ability to use the instruments and positive attitudetoward including VELS in regular surveying courses (Dib, Adamo-Villani, &Garver, 2014). Still at Purdue, Richardson, and Adamo-Villani have designed avirtual lab for undergraduate instruction in microcontroller technology. A for-mative evaluation with 42 students showed that students perceived the VirtualLearning Environment (VLE) experience comparable to the physical laboratoryexperience (Richardson & Adamo-Villani, 2010).

The Virtual Steel Sculpture

The interactive virtual steel structure is an accurate 3D replica of the physicalsteel structure located next to the Civil Engineering building on the PurdueUniversity campus in West Lafayette, IN. The sculpture was modeled to scaleand can be observed from any point of view. Figure 1 shows the full model of thesculpture. The interface allows for 360� rotation of the view, zooming in and out,

Figure 1. Full model of the steel sculpture—screenshot from the interactive web

deliverable application.


panning, and tracking. In addition, students can click on the individual connec-tions to observe them more closely (Figure 2). Every connection is identified by anumber; by clicking on the number, students can access additional informationsuch as limit states and real world examples (Figure 3).

The interactive steel structure is deliverable via web or CD-ROM on standardpersonal computers (PCs and Macs). The platform for the project is based onthe highest end in 3D interactive animation.

We used Autodesk Maya software to model and texture the structure and toanimate its functionality. Interactivity with the 3D components was pro-grammed in C# using Unity 3D (2013) game development platform. Thechoice of Unity platform was based on the following considerations:

– Unity has an optimized graphics pipeline that supports interactive renderingof complex animated 3D meshes and advanced lighting and textures even oncomputers with limited graphics capabilities.

– Unity interfaces seamlessly with major 3D animation tools (i.e., AutodeskMaya and 3D Studio Max) and file formats and allows for instantaneousimport and update of asset files and animations.

Figure 2. Individual connection.


– It supports a wide range of publishing platforms, including: standalone buildsfor Mac OS and Windows, web delivery through the Unity Web Player Plug-in (3 MB), and Wii and Iphone publishing. The interactive steel sculpturecan be accessed online at: https://engineering.purdue.edu/IVM/InteractiveSteelSculpture/InteractiveSteelSculpture.html. Download of the Unity plug-in is required.

Formative Evaluation

The formative evaluation of the interactive virtual steel structure included quan-titative and qualitative methods. The evaluation instrument was an onlinesurvey with 33 rating questions, 11 open-ended questions, and 8 demographicsquestions. The objectives of the formative study were to: (a) understand stu-dents’ perceived usefulness and usability of the tool; (b) understand students’perceived effectiveness of the virtual steel sculpture in enhancing their under-standing of connection types, assembly, and accessibility; and (c) solicit feedbackfor improvement. In addition, the study aimed to determine whether the stu-dent’s academic classification (e.g., PhD, MS, undergraduate status), number of

Figure 3. Additional details.


quarters or semesters of steel design courses taken, the interaction between thesetwo factors, and the number of hours that the subject spent interacting with thevirtual steel sculpture had a significant effect on the answers to the ratingquestions.

Subjects

A total of 34 civil engineering students participated in the study; two studentsdid not complete the online survey and were removed from the pool of subjects.The 32 students who offered valid responses classified as follows (Tables 1–3):

Stimuli

Stimuli included the online interactive steel sculpture. The virtual steel structurecan be displayed in a web browser, such as Safari, Internet Explorer, and Firefox(download of the unity 3D web plugin is required). Users can interact with thevirtual tool using a standard mouse or track pad.

Procedure

The experiment was conducted in a Structural Steel Design Class (CE 470) in theDepartment of Civil Engineering at Purdue University. One of the class

Table 2. Students’ Areas of Concentration.

Area of concentration Number Percentage

Civil 25 45.45

Structural 16 29.09

Architectural 2 3.64

Structural + Civil 10 18.18

Architectural + Civil 1 1.82

Table 1. Students’ Academic Classification.

Year Number Percentage

M.S. 1 2.86

Year 5 4 11.43

Senior 22 62.86

Junior 5 14.29


assignments requires the students to study the steel sculpture, identify the vari-ous steel connections, and draw cross sections and perspective views of theseconnections. The students were offered the virtual sculpture as an alternative tothe actual full size sculpture that exists at Purdue University; 34 studentsenrolled in this course chose to use the virtual structure in place of the physicalone and participated in the study. Students accessed the virtual sculpture ontheir own computers and spent as much time as they desired interacting with it.After the interaction with the tool, they were asked to fill out the online survey.The survey remained active for 2 weeks.

Findings

Overall findings show that students found the interactive steel structure easy touse and useful. They also perceived it as an effective tool for improving theirunderstanding of steel connection types, assembly, and accessibility.

Answers to rating questions. The mean value and standard deviation for eachrating question are reported in Table 4. The percentage of ratings is reportedin Table 5.

Answers to open-ended questions. Students found the following to be the best attri-butes of the virtual steel structure:

– Actual visualization of connections. Students commented:

The ability to see connections from different angles and connections that are high

up is a great plus

Explicit viewing of high number of connection types is extremely useful; Seeing the

elements gives you an actual look at things

Table 3. Semester the Steel Design Course Was Taken.

Semester the steel design course taken Number Percentage

Yes 30 85.71

Yes—1 quarter 4 12.50

Yes—3 quarters 1 3.13

Yes—0.5 semester 1 3.13

Yes—1 semester 25 78.13

Yes—1 quarter + 1 semester 1 3.13


Table 4. Mean Value and Standard Deviation for Each Rating Question.

Question

Mean

1¼ low rating;

5¼ high rating STD

Webpage helpfulness and maneuverability

The layout and visual design 3.85 0.905

The structure and organization of information 3.97 0.81

The ease of navigation to find information 3.79 0.86

The quality of tutorial videos 3.72 1.02

The usefulness of tutorial videos 3.61 1.04

The ease of downloading files 3.92 0.92

Interactive virtual steel sculpture helpfulness and maneuverability

Rotation capabilities 4.0 1.03

Zooming capabilities 3.87 1.07

Isolating a connection capabilities 3.79 1.05

The ease of navigation to view a connection

from different angles

3.84 1.08

The ease of navigation to view different connection types 3.67 0.96

Helpfulness and maneuverability of the 2D pdf

file for each connection

The ease of navigation to obtain information for a connection 3.93 1.09

Quality of blueprint 4.11 0.76

Usefulness of blueprint 3.96 0.77

Quality of close-up views 4.13 0.77

Usefulness of close-up views 4.20 0.71

Quality of field examples 4.07 0.70

Usefulness of field examples 4.07 0.84

Quality of sample calculations 4.12 0.79

Usefulness of sample calculations 4.0 0.72

Quality of FEA 3.87 0.74

Usefulness of FEA 3.91 0.77

Helpfulness of sample calculations

The clarity of connection description 4.26 0.65

The clarity of references to AISC specifications 4.12 0.78

The usefulness and clarity of schematic drawings 4.12 0.72

The clarity of sample calculation steps 4.17 0.82

(continued)


It is easy to learn and recognize the connections by directly observing them.

– Large number of connection types. Students commented: “the virtual struc-ture provides tons of examples”;

“it covers all types of connections.”

– Ease of navigation. Students commented:

I thought that the tree was very easy to navigate in the sense of zooming, panning,

and rotating

It’s great to have the ability to focus on one connection so you are not confused by

the rest of the sculpture around it

It is really cool the way the connection you are looking at changes colors because

that makes sure you are looking at what you want to

Easy Maneuverability. It was easy to move the image around to see the different

connections and being able to easily click on them.

– Description of connections. Students commented:

I liked that you could click on the connections and it showed the limit states

Table 4. Continued

Question

Mean

1¼ low rating;

5¼ high rating STD

The effectiveness of the virtual steel sculpture in enhancing understanding of

connection types, assembly, and accessibility

Tension connections 4.09 0.72

Shear connections 4.06 0.70

Shear–moment connections 4.06 0.75

Anchorages 3.93 0.73

How connections are assembled (need for coping, etc.) 4.09 0.78

Allowance for mechanical/electrical conduits 3.93 0.91

Note. AISC¼American Institute of Steel Construction; FEA¼ Finite Element Analysis.


Tab

le5.

Perc

enta

geof

Rat

ings

for

Eac

hQ

uest

ion.

Meas

ure

ments

Subm

eas

ure

ments

1¼

Poor

(%)

2¼

Acc

epta

ble

(%)

3¼

Fair

(%)

4¼

Very

good

(%)

5¼

Exce

llent

(%)

NA¼

Not

Applic

able

(%)

Bla

nk

(%)

Val

id

Perc

enta

ge

The

virt

ual

steel

sculp

ture

webpag

e

help

fuln

ess

and

man

euve

rabili

ty

layo

ut

and

visu

aldesi

gn2.8

60.0

028.5

740.0

022.8

62.8

62.8

694.2

9

stru

cture

and

org

aniz

atio

n

of

info

rmat

ion

0.0

02.8

622.8

642.8

625.7

12.8

62.8

694.2

9

eas

eof

nav

igat

ion

tofin

d

info

rmat

ion

0.0

05.7

128.5

740.0

020.0

02.8

62.8

694.2

9

qual

ity

of

tuto

rial

videos

2.8

65.7

114.2

934.2

914.2

925.7

12.8

671.4

3

use

fuln

ess

of

tuto

rial

videos

2.8

65.7

122.8

625.7

114.2

925.7

12.8

671.4

3

eas

eof

dow

nlo

adin

gfil

es

0.0

02.8

625.7

122.8

625.7

120.0

02.8

677.1

4

The

inte

ract

ive

virt

ual

steelsc

ulp

ture

help

fuln

ess

and

man

euve

rabili

ty

Rota

tion

capab

ilities

5.7

10.0

014.2

942.8

634.2

90.0

02.8

697.1

4

Zoom

ing

capab

ilities

2.8

68.5

714.2

937.1

431.4

32.8

62.8

694.2

9

Isola

ting

aco

nnect

ion

capab

ilities

2.8

68.5

720.0

037.1

428.5

70.0

02.8

697.1

4

eas

eof

nav

igat

ion

tovi

ew

aco

nnect

ion

from

diff

ere

nt

angl

es

2.8

68.5

717.1

437.1

428.5

72.8

62.8

694.2

9

eas

eof

nav

igat

ion

tovi

ew

diff

ere

nt

connect

ion

types

2.8

65.7

128.5

742.8

617.1

40.0

02.8

697.1

4

(continued)

441

Tab

le5.

Continued

Meas

ure

ments

Subm

eas

ure

ments

1¼

Poor

(%)

2¼

Acc

epta

ble

(%)

3¼

Fair

(%)

4¼

Very

good

(%)

5¼

Exce

llent

(%)

NA¼

Not

Applic

able

(%)

Bla

nk

(%)

Val

id

Perc

enta

ge

The

help

fuln

ess

and

man

euve

rabili

tyof

the

2D

pdf

file

for

eac

hco

nnect

ion

eas

eof

nav

igat

ion

to

obta

inin

form

atio

nfo

r

aco

nnect

ion

2.8

65.7

120.0

025.7

134.2

98.5

72.8

688.5

7

Qual

ity

of

blu

epri

nt

0.0

00.0

017.1

431.4

325.7

120.0

05.7

174.2

9

Use

fuln

ess

of

blu

epri

nt

0.0

00.0

022.8

631.4

320.0

020.0

05.7

174.2

9

Qual

ity

of

close

-up

view

s0.0

00.0

020.0

034.2

931.4

38.5

75.7

185.7

1

Use

fuln

ess

of

close

-up

view

s

0.0

00.0

014.2

940.0

031.4

38.5

75.7

185.7

1

Qual

ity

of

field

exam

ple

s0.0

00.0

017.1

442.8

622.8

611.4

35.7

182.8

6

Use

fuln

ess

of

field

exam

ple

s

0.0

02.8

617.1

434.2

928.5

711.4

35.7

182.8

6

Qual

ity

of

sam

ple

calc

ula

tions

0.0

00.0

017.1

425.7

125.7

125.7

15.7

168.5

7

Use

fuln

ess

of

sam

ple

calc

ula

tions

0.0

00.0

017.1

434.2

917.1

425.7

15.7

168.5

7

Qual

ity

of

FEA

0.0

00.0

022.8

631.4

314.2

925.7

15.7

168.5

7

Use

fuln

ess

of

FEA

0.0

00.0

022.8

628.5

717.1

425.7

15.7

168.5

7

The

sam

ple

calc

ula

-

tions

help

fuln

ess

clar

ity

of

connect

ion

desc

ription

0.0

00.0

08.5

740.0

028.5

717.1

45.7

177.1

4

clar

ity

of

refe

rence

sto

AIS

CSp

eci

ficat

ions

0.0

00.0

017.1

428.5

725.7

122.8

65.7

171.4

3

(continued)

442

Tab

le5.

Continued

Meas

ure

ments

Subm

eas

ure

ments

1¼

Poor

(%)

2¼

Acc

epta

ble

(%)

3¼

Fair

(%)

4¼

Very

good

(%)

5¼

Exce

llent

(%)

NA¼

Not

Applic

able

(%)

Bla

nk

(%)

Val

id

Perc

enta

ge

use

fuln

ess

and

clar

ity

of

schem

atic

dra

win

gs

0.0

00.0

014.2

934.2

922.8

622.8

65.7

171.4

3

clar

ity

of

sam

ple

calc

ula

-

tion

steps

0.0

00.0

017.1

422.8

628.5

725.7

15.7

168.5

7

The

effect

iveness

of

the

virt

ual

steel

sculp

ture

in

enhan

cing

your

unders

tandin

gof

connect

ion

types,

asse

mbly

,an

d

acce

ssib

ility

Tensi

on

connect

ions

0.0

00.0

020.0

048.5

728.5

70.0

02.8

697.1

4

Shear

connect

ions

0.0

00.0

020.0

051.4

325.7

10.0

02.8

697.1

4

Shear

-Mom

ent

connect

ions

0.0

00.0

022.8

645.7

128.5

70.0

02.8

697.1

4

Anch

ora

ges

0.0

00.0

025.7

145.7

120.0

02.8

65.7

191.4

3

How

connect

ions

are

asse

mble

d

0.0

00.0

022.8

640.0

031.4

30.0

05.7

194.2

9

Allo

wan

cefo

rm

ech

anic

al/

ele

ctri

calco

nduits

0.0

02.8

628.5

728.5

728.5

75.7

15.7

188.5

7

443

Description of connection was very useful (e.g. welded to beam web and bolted to

girder web)

It’s nice to know which limit states would need to be checked in calculations.

– Real world examples. Students commented:

The tool provides real world examples show what could actually happen

The real life example was very useful as well, brought a realistic image to it

The ability to zoom in and get a real life example of what the connection looks like

really helps.

Factor correlation analysis. A factor correlation analysis was conducted to deter-mine whether academic classification, number of semesters of steel designcourses taken, and number of hours spent interacting with the tool had aneffect on the answers to rating questions.

H1: The academic classification will have no significant effect on the respondent’s

Likert score

H2: The number of quarters or semesters that respondent has taken for steel design

courses will have no significant effect on the respondent’s Likert score

H3: The interaction between academic classification and number of quarters or

semesters of steel design courses that respondent has taken will not have significant

effect on the respondent’s Likert score

H4: There is no significant difference in the respondent’s Likert mean score based

on its number of hours exploring the virtual sculpture.

Findings

Results of the statistical analysis failed to reject the null hypotheses. The aca-demic classification of the respondent, number of quarters or semesters of steeldesign courses that respondent has taken, and the interaction between these twofactors have no significant effect on the respondent’s mean Likert score. Thenumber of hours that respondent has spent on exploring the virtual sculpturehas no effect on its mean Likert score.


Description of Statistical Analysis

Respondent’s Mean Score¼ the academic classification+number of quarters orsemesters of steel design courses+ the academic classification� number ofquarters or semesters of steel design courses

Academic classification:

1: PhD; 2: MS; 3: Undergraduate Year 5; 4: Senior; 5: Junior; 6: Sophomore; 7:Freshman; and 8: Other.

Number of quarters or semesters of steel design courses that respondent hastaken:

1: 1 quarter; 2: 2 quarters; 3: 3 quarters; 4: 4 quarters; 5: 0.5 semester; 6: 1semester; 7: 2 semesters: 8: 3 semesters; 9: Other.

Details of the univariate analysis of variance (ANOVA) are reported inTables 6 and 7

One-way ANOVA was used to determine whether there is a difference inthe respondents’ Likert mean score based on the number of hours spentexploring the virtual sculpture (1: less than 1 hour; 2: 1–2 hours; 3: 2–3hours; 4: more than 4 hours). Details of the statistical analysis are reported inTables 8 to 10.

Cronbach alpha reliability analysis. To test the reliability of the data collected fromthe rating questions and measure internal consistency, we used Cronbach’s alphareliability analysis and exploratory factor analysis. The analyses confirmed datareliability and internal consistency. Details of the statistical analyses areincluded in the online Appendix A (available at http://ets.sagepub.com/supplemental).

Table 6. Between-Subjects Factors.

N

Academic classifications 3 1

4 16

5 5

Number of semester/quarter courses 1 2

3 1

5 1

6 18


Discussion and Conclusion

In this article, we have reported the development and initial evaluation of aninteractive virtual steel sculpture designed to help undergraduate and graduatestudents understand the mechanics and steel behavior. Findings of a formative

Table 7. Tests of Between-Subjects Effects.

Dependent Variable: SCORE

Source

Type III sum

of squares df

Mean

square F Sig.

Corrected model 4.494a 6 .749 2.675 .057

Intercept 94.381 1 94.381 337.150 .000

Academic classifications 1.980 2 .990 3.537 .055

Number of semester quarter courses 2.369 3 .790 2.821 .075

Academic classifications*

Number of semester quarter courses

.067 1 .067 .240 .632

Error 4.199 15 .280

Total 372.922 22

Corrected total 8.693 21

aR2¼ .517 (Adjusted R2

¼ .324).

Table 8. Descriptives.

Score

95% CI for MeanBetween-

N Mean SD SE

Lower

bound

Upper

bound Minimum Maximum

component

variance

1 14 3.932 .6290 .1681 3.569 4.295 2.8 5.0

2 7 4.196 .6376 .2410 3.607 4.786 3.3 5.0

3 2 4.203 1.1270 .7969 �5.922 14.328 3.4 5.0

4 1 3.000 . . . . 3.0 3.0

Total 24 3.993 .6682 .1364 3.711 4.275 2.8 5.0

Model

Fixed

effects

.6653 .1358 3.710 4.276

Random

effects

.1458 3.529 4.457 .0065


evaluation with 32 students showed that the virtual steel structure was perceivedeasy to use, useful, and effective for improving students’ understanding of steelconnection types, assembly, and accessibility.

The results of this study are important as they add to the growing body ofevidence that suggests that interactive 3D virtual learning environments couldimprove students learning in STEM disciplines. Future work will focus on con-ducting a summative evaluation with a larger pool of students. If future studiesconfirm the pedagogical efficacy of the virtual steel structure tool, extending theVR approach to other areas of civil engineering seems to be a logical step inwhich to proceed.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research,authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, author-

ship, and/or publication of this article: This research work is supported by NSF- TUES,Division of Undergraduate Education, Award # 1140563.

Table 9. Test of Homogeneity of Variances.

Score

Levene statistic df1 df2 Sig.

.738a 2 20 .490

aGroups with only one case are ignored in computing the test of

homogeneity of variance for SCORE.

Table 10. ANOVA.

Score

Sum of

squares df

Mean

square F Sig.

Between groups 1.416 3 .472 1.067 .386

Within groups 8.852 20 .443

Total 10.269 23


References

American Institute of Steel Construction [AISC]. (2005). Code of standard practices forsteel buildings and bridges. Available at: https://www.aisc.org/uploadedFiles/Steel_Solution_Center/Technical_Resources/Engineering_FAQs/Code_of_Standard_

Practice.pdfBransford, J. D., Brown, A. L., & Cocking, R. R. (1999). How people learn: Brain, mind,

experience, and school. Washington, DC: National Academies Press. Retrieved fromhttp://books.nap.edu/catalog/6160.html

Dalgarno, B., & Harper, B. (2004). User control and task authenticity for spatiallearning in 3D environments. Australasian Journal of Educational Technology,20(1), 1–17.

Dib, H., & Adamo-Villani, N. (2011). An innovative software application for surveyingeducation. Journal of Computer Applications in Engineering Education, 22(3), 551–562.

Dib, H., Adamo-Villani, N., & Garver, S. (2014, Winter). An interactive virtual envir-

onment for learning differential leveling: Development and initial findings. Advancesin Engineering Education (ASEE), 4(1). Available at: http://advances.asee.org/?publi-cation=an-interactive-virtual-environment-for-learning-differential-leveling-develop-

ment-and-initial-findingsDillenburg, P. (2000). Virtual learning environments. Proc. of EUN Conference 2000 –

Workshop on Virtual Learning Environments, Geneva, Switzerland.Florida State University, College of Engineering eNews Archives. (1998). Steel connection

sculpture. Retrieved from http://www.eng.fsu.edu/shared/news/article.html?s¼steel_connections

Gorsky, P., & Finegold, M. (1992). Using computer simulations to restructure students’

conceptions of force. Journal of Computers in Mathematics and Science Teaching, 11,163–178.

Green, P. S., Sputo, T., & Veltri, P. (2008). Connections teaching toolkit – A teaching guide

for structural steel connections, AISC, 2008. Retrieved from http://www.engr.mun.ca/�sbruneau/thingstuff/munstuff/civilroom/guide.pdf

Hmelo, C., & Williams, S. M. (Eds.). (1998). Special issue: Learning through problemsolving. The Journal of the Learning Sciences, 7(3), 4.

Holt, R., & Hartmann, J. (2008, January). Adequacy of the U10 & L11 Gussett PlateDesign of Minnesota Bridge No. 9340, I-35W over Mississippi River. (Interim Report),Federal Highway Administration.

Kangassalo, M. (1994). Children’s independent exploration of a natural phenomenon byusing a pictorial computer-based simulation. Journal of Computing in ChildhoodEducation, 5(3/4), 285–297.

Karweit, M. (2010). A virtual engineering/science laboratory course. Retrieved from http://www.jhu.edu/�virtlab/virtlab.html

Kuo, H. L., Kang, S. C., Lu, C. C., Hsieh, S. H., & Lin, Y. H. (2007). Feasibility

study: Using a virtual surveying instrument in surveying training. Coimbra, Portugal:ICEE.

Mannix, M. (2000). The virtues of virtual labs. ASEE Prism. Retrieved from http://www.prism-magazine.org/sept00/html/toolbox.cfm

Pfrang, E. O., & Marshall, R. (1982, July). Collapse of the Kansas City Hyatt Regencywalkways, civil engineering. ASCE, 52(7), 65–69.


Richardson, J., & Adamo-Villani, N. (2010). A virtual embedded microcontroller labora-tory for undergraduate education: Development and evaluation. Engineering Design

Graphics Journal (ASEE), 74(3), 1–12.Shin, Y. S. (2003). Virtual experiment environments design for science education. Proc. of

IEEE International Conference on Cyberworlds (CW03), Washington DC.

Smart Structures Technology Laboratory (SSTE) at UIUC. (2008). Virtual laboratory forearthquake engineering. Retrieved from http://sstl.cee.illinois.edu/

Smith, E. A., & Epstein, H. I. (1980). Hartford coliseum roof collapse: Structural collapsesequence and lessons learned. Civil Engineering—ASCE, 50(4), 59–62.

Unity 3D. (2013). Retrieved from Unity3D.comWinn, W. (2002). Current trends in educational technology research: The study of learn-

ing environments. Educational Psychology Review, 14(3), 331–351.

Zietsman, A. I., & Hewson, P. W. (1986). Effect of instruction using microcomputersimulations and conceptual change strategies on science learning. Journal ofResearch in Science Teaching, 23, 27–39.

Author Biographies

Hazar Nicholas Dib is associate professor of Building Information Managementin the School of Construction Management Technology at Purdue Polytechnic,Purdue University. His research areas include the integration of 3D technologiesin information modeling techniques, development of knowledge acquisitionmethods and strategies for examining and capturing expert design knowledge,increase efficiency in the construction processes, and design and development ofinnovative teaching technologies for engineering education. Prior to joining thefaculty at Purdue University, Dr. Dib worked on management of commercialconstruction for major construction management companies. He received hisBSc and MSc in Civil Engineering in 2002 from the University of Balamand(Lebanon) and his PhD in Building Construction Management with an empha-sis on information sharing and information management from the University ofFlorida in 2007.

Nicoletta Adamo-Villani is professor of Computer Graphics technology and aPurdue University Faculty Scholar. She is an award-winning animator andgraphic designer and creator of several 2D and 3D animations that have airedon national television. Her area of expertise is in character animation and char-acter design and her research interests focus on the application of 3D animationtechnology to education, HCC (Human Computer Communication), and visua-lization. Nicoletta is co-founder and director of the IDeaLaboratory. She has anMS in Architecture from University of Florence, Italy and she is a certifiedanimator and instructor for Autodesk.


Appendix A

Cronbach Alpha Reliability Analysis

To test the reliability of the data collected from the rating questions and measureinternal consistency, we used Cronbach’s alpha reliability analysis and explora-tory factor analysis. The analyses confirmed data reliability and internalconsistency.

Q1: The Virtual Steel Sculpture Webpage helpfulness andManeuverability (Reliable)

Q1_1: Layout and visual designQ1_2: Structure and organization of informationQ1_3: Ease of navigation to find informationQ1_4: Quality of tutorial videosQ1_5: Usefulness of tutorial videosQ1_6: Ease of downloading files

Case processing summary

Reliability statistics

Item statistics

N %

Cases Valid 24 70.6

Excludeda 10 29.4

Total 34 100.0

aList-wise deletion based on all variables in the procedure.

Cronbach’s alpha Number of items

.863 6

Mean SD N

Q1_1 3.88 .992 24

Q1_2 3.83 .816 24

Q1_3 3.75 .847 24

Q1_4 3.71 1.042 24

Q1_5 3.58 1.060 24

Q1_6 3.88 .900 24

Item-total statistics

Scale statistics

Q2: The Interactive Virtual Steel Sculpture Helpfulness andManeuverability (Reliable)

Q2_1: Rotation capabilitiesQ2_2: Zooming capabilitiesQ2_3: Isolating a connection capabilitiesQ2_4: Ease of navigation to view a connection from different anglesQ2_5: Ease of navigation to view different connection types



Scale mean if

item deleted

Scale variance

if item deleted

Corrected

item-total

correlation

Cronbach’s alpha

if item deleted

Q1_1 18.75 13.239 .690 .834

Q1_2 18.79 14.433 .661 .841

Q1_3 18.88 13.766 .751 .826

Q1_4 18.92 12.167 .819 .808

Q1_5 19.04 11.955 .836 .804

Q1_6 18.75 16.630 .240 .906

Mean Variance SD Number of items

22.63 19.201 4.382 6

N %

Cases Valid 32 94.1

Excludeda 2 5.9

Total 34 100.0



.903 5

Item statistics


Scale statistics

Q3: The Helpfulness and Maneuverability of the 2D Pdf File forEach/Connection (Reliable)

Q3_1: Ease of navigation to obtain information for a connectionQ3_2: Quality of blueprintQ3_3: Usefulness of blueprintQ3_4: Quality of close-up viewsQ3_5: Usefulness of close-up viewsQ3_6: Quality of field examplesQ3_7: Usefulness of field examplesQ3_8: Quality of sample calculationsQ3_9: Usefulness of sample calculationsQ3_10: Quality of FEAQ3_11: Usefulness of FEA

Mean SD N

Q2_1 4.03 1.062 32

Q2_2 3.94 1.076 32

Q2_3 3.84 1.081 32

Q2_4 3.84 1.081 32

Q2_5 3.72 0.958 32

Scale mean if

item deleted

Scale variance

if item deleted

Corrected

item-total

correlation

Cronbach’s

alpha if item

deleted

Q2_1 15.34 13.846 .634 .908

Q2_2 15.44 12.835 .777 .878

Q2_3 15.53 12.773 .782 .877

Q2_4 15.53 12.709 .792 .875

Q2_5 15.66 13.330 .820 .871


19.38 19.984 4.470 5



Item statistics


Mean SD N

Q3_1 3.96 1.147 23

Q3_2 4.22 .736 23

Q3_3 4.00 .798 23

Q3_4 4.17 .778 23

Q3_5 4.22 .736 23

Q3_6 4.04 .706 23

Q3_7 4.00 .853 23

Q3_8 4.13 .815 23

Q3_9 4.00 .739 23

Q3_10 3.87 .757 23

Q3_11 3.91 .793 23

Scale mean if

item deleted

Scale variance

if item deleted

Corrected

item-total

correlation

Cronbach’s

alpha if

item deleted

Q3_1 40.57 45.984 .646 .967

Q3_2 40.30 50.040 .650 .962

(continued)

N %

Cases Valid 23 67.6

Excludeda 11 32.4

Total 34 100.0



.960 11

Scale statistics

Q4: The Sample Calculations Helpfulness (Reliable)

Q4_1: Clarity of connection descriptionQ4_2: Clarity of references to AISC SpecificationsQ4_3: Usefulness and clarity of schematic drawingsQ4_4: Clarity of sample calculation steps


Continued

Scale mean if

item deleted

Scale variance

if item deleted

Corrected

item-total

correlation

Cronbach’s

alpha if

item deleted

Q3_3 40.52 47.170 .871 .955

Q3_4 40.35 47.964 .815 .957

Q3_5 40.30 47.858 .879 .955

Q3_6 40.48 48.625 .836 .956

Q3_7 40.52 46.715 .850 .955

Q3_8 40.39 47.431 .825 .956

Q3_9 40.52 47.079 .960 .952

Q3_10 40.65 47.692 .869 .955

Q3_11 40.61 46.885 .906 .954

N %

Cases Valid 24 70.6

Excludeda 10 29.4

Total 34 100.0



44.52 57.352 7.573 11


Item statistics


Scale statistics

Q5: The Effectiveness of the Virtual Steel Sculpture in EnhancingYour Understanding of Connection Types, Assembly, and Accessibility(Reliable)

Q5_1: Tension connectionsQ5_2: Shear connections

Scale mean if

item deleted

Scale variance

if item deleted

Corrected

item-total

correlation

Cronbach’s

alpha if

item deleted

Q4_1 12.46 4.955 .817 .956

Q4_2 12.54 4.433 .891 .934

Q4_3 12.58 4.428 .927 .923

Q4_4 12.54 4.172 .908 .930

Mean SD N

Q4_1 4.25 .676 24

Q4_2 4.17 .761 24

Q4_3 4.13 .741 24

Q4_4 4.17 .816 24


16.71 7.868 2.805 4


.952 4

Q5_3: Shear-Moment connectionsQ5_4: AnchoragesQ5_5: How connections are assembledQ5_6: Allowance for mechanical/electrical conduits



Item statistics


Mean SD N

Q5_1 4.16 .688 31

Q5_2 4.10 .700 31

Q5_3 4.10 .746 31

Q5_4 3.94 .727 31

Q5_5 4.13 .763 31

Q5_6 3.94 .892 31

Scale mean if

item deleted

Scale variance

if item deleted

Corrected

item-total

correlation

Cronbach’s

alpha if

item deleted

Q5_1 20.19 12.361 .896 .951

Q5_2 20.26 12.065 .949 .946

Q5_3 20.26 11.865 .924 .948

(continued)

N %

Cases Valid 31 91.2

Excludeda 3 8.8

Total 34 100.0



.960 6

Scale statistics

Continued

Scale mean if

item deleted

Scale variance

if item deleted

Corrected

item-total

correlation

Cronbach’s

alpha if

item deleted

Q5_4 20.42 12.118 .893 .951

Q5_5 20.23 12.381 .783 .963

Q5_6 20.42 11.318 .842 .959


24.35 17.170 4.144 6

A Virtual Steel Sculpture for Structural The Author(s) …hpcg.purdue.edu/idealab/2016 publications/2. Journal of...nected. Unlike the physical steel sculpture, this interactive virtual

Documents