Exploring 3D virtual reality technology for spatial ... - Wiley Online Library

Original article doi: 10.1111/jcal.12018

bs_bs_banner

Exploring 3-D virtual reality technology for spatial ability and chemistry achievement Z. Merchant,* E.T. Goetz,† W. Keeney-Kennicutt,‡ L. Cifuentes,§ O. Kwok† & T.J. Davis* *College of Education, Department of Teaching Learning and Culture, Texas A&M University, College Station, Texas, USA †College of Education, Department of Educational Psychology, Texas A&M University, College Station, Texas, USA ‡College of Science, Department of Chemistry, Texas A&M University, College Station, Texas, USA §Department of Distance Learning, Texas A&M University, Corpus Christi, Texas, USA

Abstract

We investigated the potential of Second Life® (SL), a three-dimensional (3-D) virtual world, to enhance undergraduate students’ learning of a vital chemistry concept. A quasiexperimental pre-posttest control group design was used to conduct the study. A total of 387 participants completed three assignment activities either in SL or using two-dimensional (2-D) images. Students were administered an 11-question chemistry achievement test and two measures of spatial ability (Purdue Visualization of Rotations Test, Card Rotations Test). Although analyses of covariance revealed no statistically significant differences between the two groups as a whole for any of the outcome measures, a subgroup analyses was conducted to decompose the relative impact of 3-D virtual reality instruction within SL. We found that students classified as having poor spatial ability showed significantly greater improvement in understanding the 3-D nature of molecules if they did relevant activities in a 3-D virtual world than those students who only worked with 2-D images.

Keywords

3-D virtual worlds, chemistry achievement, Second Life, spatial instruction, virtual environment spatial ability, virtual reality.

A major goal of introductory chemistry classes is to enhance students’ abilities to build cognitive representations of molecular structures and to manipulate them mentally. Instructors of high school and introductory college chemistry courses regularly use twodimensional (2-D) images and three-dimensional (3-D) models (e.g., spheres connected by sticks) as tools to visually represent the components and geometry of molecules. Then, students must mentally translate those visual representations of molecular structures to interpret complex chemical processes and spatial relationships. Using their spatial abilities, students trans-

Accepted: 3 April 2013 Correspondence: Zahira Merchant, College of Education, Department of Teaching Learning and Culture, Texas A&M University, College Station, Texas 77843-4232, USA. Email: [email protected]

© 2013 John Wiley & Sons Ltd

late chemical formulae into molecular structures, visualize possible 3-D configurations, and compare these configurations across different molecular structures. Therefore, the ability to comprehend and mentally manipulate molecular structures is critical for students to understand fundamental concepts and conduct advanced scientific research (Wu & Shah, 2004). Mathewson (1999, p. 36) stated, ‘A spatial image preserves relationships among a complex set of ideas as a single chunk in working memory, increasing the amount of information that can be maintained in consciousness at a given moment’. It is this integration of information that is critical in understanding complex molecular structures and bond angles. To facilitate students’ understanding of many chemistryrelated concepts, it is important to enhance their spatial abilities.

Journal of Computer Assisted Learning (2013), 29, 579–590

579

580

One of the most vital and promising affordances of the virtual reality technologies is to provide spatial instruction. According to Moore (1995) ‘. . . by teaching the students to think in 3-D using visualization techniques, their spatial cognition can be enhanced’ (p. 5). Similarly, Hedberg and Alexander (1994) stated that, ‘As ideas are represented in a 3-D world, 3-D thinking can be enhanced, and the mental transformation of information from two to three dimensions can be facilitated’ (p. 216). Dalgarno, Hedberg, and Harper (2002) propose that ‘If 3-D environment is a metaphorical representation of abstract ideas, it may be that by developing an integrated database of 2-D views of a three 3-D model of the concepts, we are better able to make sense of the concepts than through other instructional approaches’ (p. 8). As espoused by these scholars, one of the critical features of 3-D virtual reality environments is the ability to visually depict and interact with spatial representations of abstract concepts. Therefore, this feature of 3-D virtual environments can be useful in providing instruction for developing spatial ability. Studies reporting the outcome of implementing this approach of integrating 3-D visualization technology for imparting spatial training are discussed in the next section. Significance of spatial ability in chemistry achievement

Researchers have understood spatial ability as a complex and multifaceted skill, although Lohman (1988) differentiated spatial ability into five components: visualization, speeded rotation, closure speed, closure flexibility and perceptual speed. However, the definitions and descriptions of these categories overlap. Therefore, most researchers identify two or three major components of spatial ability. Three-component analyses of spatial ability stipulate spatial visualization, spatial orientation and spatial relation (Ekstrom, French, Harman, & Dermen, 1976; Pellegrino & Kail, 1982; Pellegrino & Hunt, 1991; Robichaux & Guarino, 2000). However, there is a considerable overlap in the definitions of spatial orientation and spatial visualization, leading most researchers to consider only two major components of spatial ability: spatial relation and spatial orientation (Coleman & Gotch, 1998; Harle & Towns, 2011; Mohler, 2008; Piburn et al., 2002). Therefore, we adopted a two-component analysis for the present study.

Z. Merchant et al.

Spatial relation is the ability to mentally rotate an object on its axes (Shepard & Cooper, 1982). Spatial orientation is the ability to mentally manipulate or transform an object into another arrangement (Ekstrom et al., 1976). Studies in the literature have found a positive correlation between the components of spatial ability measures and academic performance. For example, Carter, LaRussa, and Bodner (1987) found that undergraduate students who scored high on spatial ability tests also scored high on the chemistry performance test. Bodner and Guay (1997) and Tuckey, Selvaratnam, and Bradley (1991) also found a strong positive relationship between spatial relation and scores on chemistry tests. Therefore, it can be concluded that both components of spatial ability, spatial relation and spatial orientation, play a significant role in chemistry performance. Upon recognizing the importance of spatial ability in chemistry, researchers have designed studies to examine whether instruction can enhance it. Barnea and Dori (1999) used computer molecular modeling (CMM) virtual reality software to embed spatial instruction into teaching the structure and bonding of molecules to a group of 10th-grade students while another group of students was given traditional instruction using plastic ball-and-stick models. They found that the group given with CMM-based instruction outperformed the control group on the spatial ability and the chemistry performance tests. Copolo and Hounshell (1995) used Molecule Editor to provide spatial training for 11th-grade high school students in an organic chemistry class. They compared the students who received spatial training to three control groups who were taught using 2-D textbook representations, 3-D ball-and-stick models, and a combination of 3-D ball-and-stick and computer models. The results of their study indicated that students receiving a combination of the instructional approaches scored higher on an isomeric identification test, but on a 2-D version of the same test, the group receiving training with 2-D textbook representations performed best. Ferk Savec, Vrtacˇnik, and Gilbert (2005) also investigated the impact of spatial training on secondary school students and found that training with 3-D representations was superior in improving students’ performances on the molecular visualization test. Moreover, Ozmen (2008) reported a quasi-experimental study with 11th-grade students in which statistically significant differences © 2013 John Wiley & Sons Ltd

Exploring 3-D virtual reality technology

were found on a chemical bonding test favouring the experimental group, who had additional computerassisted learning, over the control group, who received traditional instruction in form of lecture and PowerPoint presentation. More recently, Urhahne, Nick, and Schanze (2009) used CHEMnet to embed spatial training into the curriculum while teaching a module on modification of carbon to a freshman class. They compared the effects of 3-D simulations against 2-D images and found no difference in the knowledge gains of both the groups. Limniou, Roberts, and Papadopoulos (2008) used 3-D molecular representations with college students to teach the reactive properties of solutions and compounds. However, they provided two kinds of instruction to the same group of 14 students. This instruction included 2-D images and 3-D interactive representations of molecular structures. During a 3-D interactive virtual reality training session, students used many peripheral devices, such as glass cubicles, shutter glasses and joysticks. Their study found that students comprehended reaction processes better after receiving the training in a 3-D virtual reality environment. Current research literature on the impact of 3-D virtual reality environments when used for spatial training in chemistry classes seems inconclusive. The studies discussed above were mostly conducted with high school students and very few with undergraduate students. In addition, these studies demonstrated that the 3-D environment is superior to the traditional lecture approach to teaching topics involving spatial understanding, but the instructional advantage of working in a 3-D environment over looking at 2-D images is still ambiguous. Purpose

The purpose of this study was to explore the effectiveness of a Second Life® (SL) virtual environment to enhance undergraduate chemistry students’ spatial abilities and chemistry achievements. We selected a foundational concept in the field of chemistry that gives an explanation of the 3-D nature of molecules, which is critical for understanding many of the physical and chemical properties of chemicals. It was hypothesized that the 3-D virtual learning environment would: H1. Increase students’ understanding of three key characteristics of understanding spatial relationships in © 2013 John Wiley & Sons Ltd

581

molecular structures, H1a. molecular angles, H1b. molecular geometry, H1c. species identification, which refer to choosing examples of chemical molecules and ions with a given 3-D structure. H2. Improve students’ spatial abilities as indexed by: H2a. the Purdue Visualization of Rotations Test (PVRT) (Bodner & Guay, 1997) and H2b. the Card Rotations Test (CRT) (Ekstrom et al., 1976). Method

The data collected from the SL group, along with other data that were only available for that group, were used in an article by Merchant et al. (2012) in a study that focused exclusively on the SL group. A theoretical model of learning in 3-D virtual learning environments that are not appropriate to the control group in this study was proposed and tested using structural equation modeling. Design and participants

This study used a pretest/posttest control group quasiexperimental design in which two sections of the course were randomly assigned to be either an experimental group, whose instruction included the use of 3-D virtual environment-based instruction using SL, or the control group who had 2-D image-based instruction The same instructor presented the class lectures to both groups. The activities of the two groups differed on three assignments which were completed using either the SL environment or 2-D images. The study’s participants were 403 undergraduates enrolled in two sections of the Chemistry 101 course at a large south-western university. Of the 403 students enrolled in the course, 2 chose not to participate in the study and another 6 dropped the class. Furthermore, 11 students were dropped from the study because they completed the set of tasks out of order. The final sample consisted of 384 participants of whom 64% were female and 36% were male. Most of the participants’ (91%) ages ranged between 18 and 21 years. They were mostly Caucasian (73%) or Hispanic (15%). A total of 23% identified themselves as proficient gamers, and 3% of the students had some prior experience with SL. More descriptive statistics can be found in Table 1. Chi-square analyses revealed that students who were not included in the study did not differ from students who were included on the demographic variables.

582

Z. Merchant et al.

Table 1. Demographic Statistics for 3-D Virtual Reality and 2-D Images Groups

Variable

Groups

Gender

Male Female 30 Caucasian Hispanic Asia/Pacific Islander African American American Indian/Native Alaskan Others I have never heard of Second Life®. I have heard about it, but had never entered that virtual world. I am familiar with the Second Life® Environment and had created an avatar, but I consider myself a beginner in Second Life. I have spent a lot of time exploring the Second Life® environment. Yes No 0–1 year 2–5 years More than 5 years

Age

Race/Ethnicity

Second Life® experience

Gaming experience Years of gaming experience

3-D virtual reality n

2-D images n

Total n

68 136 4 188 9 3 0 146 34 15 3 2 4 151 42 7

70 108 4 163 9 3 1 136 24 12 6 0 2 102 73 5

138 244 8 351 18 6 1 282 58 27 9 2 6 253 115 12

2 43 161 90 29 67

0 44 136 98 17 59

2 87 297 188 46 126

Note. 2-D = two-dimensional; 3-D = three-dimensional.

Measures

The chemistry achievement and spatial ability measures were administered twice: before and after the intervention. These measures were the valence shell electron pair repulsion (VSEPR) theory test, the CRT, and the PVRT. Chemistry achievement The researcher developed a measure of chemistry achievement that tested students’ understanding of how to apply the principles of VSEPR theory in order to determine the 3-D shapes of molecules and ions. Knowledge of molecular shapes is critical for predicting and explaining how chemicals behave in reactions. The test consisted of 11 knowledge-based multiplechoice questions. Answering the questions required the students to mentally rotate 3-D molecular structures using 2-D perspective drawings (see Appendix I). This test covered three parts of VSEPR theory. The molecule angles section consisted of three questions on interpreting typical 2-D line/wedge drawings of 3-D

molecules. In the molecular geometry section, students had to answer four questions on recognizing the geometry of ball-and-stick representations of molecules and ions drawn in perspective. To answer these questions, the students had to translate 2-D pictures into 3-D molecules. The species identification section entailed four questions where students were given a 2-D balland-stick molecular representation drawn in perspective and they had to choose from a list which species had that geometry. Participants obtained one point for every question answered correctly and a negative point for every incorrect answer. The instructor developed the VSEPR test and three chemistry professors reviewed this test to ensure its content validity. A pilot study was conducted with 53 students who had previously taken Chem102 from the same. The Cronbach’s a reliability coefficient for the reliability of the current study’s score was 0.61, which is an acceptable level recommended for learning achievement tests (Reynolds, Livingston, & Wilson, 2009). The coefficient a for molecular angles was 0.77, molecular geometry was 0.54, and species identification was 0.44. © 2013 John Wiley & Sons Ltd


583

Figure 1 An Example of a Test Question from the Purdue Visualization of Rotations Test

Spatial ability There were two measures of spatial ability: the CRT and the PVRT. Each of these tests, described below, attempted to measure a specific aspect of the spatial ability. Card Rotations Test The CRT is a 2-D mental rotation test in which participants see a 2-D target object with eight other objects and respond whether each of those eight were either rotated or a mirror image (Ekstrom et al., 1976). This is a paper-based test with 20 items that must be completed in 6 min. Its coefficient a in the present study was 0.80. This test was administered before and after the intervention. The coefficient a for the reliability of the current study’s score was 0.61, which is an acceptable level recommended for learning achievement tests (Reynolds et al., 2009). Purdue Visualization of Rotations Test The PVRT, a 20-question test developed by Bodner and Guay (1997), is a widely used measure of spatial ability in chemical education. PVRT items are analogy problems in which students are given an example of a rotation on a 3-D object, and then asked to perform the same rotation on a different object and choose the result from five options. Students are allotted 10 min to complete the 20 questions. Thus, in the problem shown in Figure 1, option D is the correct answer. The test has consistently demonstrated a good reliability (KR20) index in many contexts ranging from 0.78 to 0.80 (Bodner & Guay, 1997). The Cronbach’s a reliability coefficient for the present study was 0.77. This study’s © 2013 John Wiley & Sons Ltd

participants completed the test before and after the intervention. Instructional software

SL, an innovative 3-D technology, launched by Linden Labs in 2003, was used to provide spatial instruction to this study’s participants in the experimental group. This Internet-based immersive virtual environment allows its users, who are called residents, to interact within this environment via their avatars. Within SL, avatars have the ability to build 3-D virtual objects (molecules in this instance). Other interactive features include the ability to interact with an object by zooming in and out, rotating the object, and programming the objects to behave in a certain manner. Chemistry corner Chemistry Corner (http://maps.secondlife.com/ secondlife/12th%20Man/214/238/26) is a SL environment in which undergraduates learn about fundamental chemistry concepts by applying SL affordances to 3-D representations of molecular structures. It was developed by one of the authors and is the only one of its type at this time. Within this environment, three instructional simulations were developed for the purposes of this study: the Molecule Game, Chemist as an Artist, and Tower of VSEPR Theory. Students were familiarized with the environment of SL and its features using seven introductory videos specifically

584

Z. Merchant et al.

a

tiple perspectives. Figure 2a shows the layout of the molecule game in SL. Students had to ‘rez’ (i.e., make an object appear in the SL environment) a molecule and answer a question at five different stations to complete this activity. After rezzing the molecules, students were posed with questions about the molecule they rezzed in SL. For example, when students rezzed the sulphuric acid molecule, a question appeared asking, ‘What is the ratio of oxygen atoms to hydrogen atoms expressed as a whole number?’ The students could view the sulphuric acid molecule, count the atoms, and rotate the molecule to view from different perspectives in order to answer the questions. On selecting their response, students received feedback and other supportive information to proceed further. When finished, the first author would receive a notification e-mail from within SL. In addition, each student e-mailed a snapshot of their avatar taken at one of the five molecule stations in SL. The control group completed students were given 2-D images of the five molecules and were asked to complete the same questions that the experimental group completed in SL. The Chemist as an Artist simulation was designed to further develop students’ SL skills and ability to see molecules in a 3-D perspective by copying, linking and rotating the molecules (Figure 2b). This simulation was designed to prepare the freshman chemistry students to reflect like a chemist, who are expected to think back and forth between 3-D and 2-D perspectives of a molecule. The students were given three molecules to manipulate in SL. After manipulating these three molecules, students had to translate the 3-D perspective of the molecules into 2-D perspective drawings and submit a report as their assignment. In their report, students were required to provide a photograph of themselves with two orientations of their molecules and a 2-D drawing of each orientation using solid lines, wedges and dashed lines. The control group were e-mailed with pictures of three molecules with two orientations and this group also completed the assignment like the experimental group. The Tower of VSEPR simulation was the final and most extensive assignment, designed to deepen students’ understanding of VSEPR theory. During this activity, students in the experimental group rezzed 11 different molecules by clicking on the rotating boxes (Figure 2c) and measured bond angles, determined molecular geometries and solved for Lewis dot struc-

b

c

Figure 2 Intervention 1: Molecule Game (a); Intervention 2: Chemist as an Artist (b); Intervention 3: Tower of VSEPR Theory (c)

developed for this study. Later, students completed three assignments in SL using simulations called the Molecule Game, the Chemist as an Artist, and the Tower of VSEPR Theory. Figure 2 provides a visual representation of the activities conducted in SL. The Molecule Game was designed for students to develop SL skills and see molecules in 3-D from mul-



585

Table 2. Pretest and Unadjusted Posttest Scores for Measures of Learning Outcomes Variables Molecule angles Pretest Posttesta Molecular geometry Pretest Posttesta Species identification Pretest Posttesta PVRT Pretest Posttesta CRT Pretest Posttesta

3-D virtual reality

2-D images

Mean

Mean

SD

SD

F

df

382 382 382 382 382 382 382 382 382 382 382 382 382 382

1.07 0.92

0.47 1.12

1.01 0.78

0.57 1.08

1.44ns 2.41ns

2.34 3.53

1.00 1.27

2.26 3.59

1.10 1.23

4.88ns 0.19ns

-0.61 0.33

2.27 3.01

-0.28 0.06

1.85 3.06

1.08ns 1.01ns

11.86 11.56

3.70 4.12

12.36 11.71

3.48 4.00

0.42ns 0.00ns

100.32 124.84

27.91 24.024

107.92 127.28

30.40 29.18

1.80ns 0.03ns

Note. 2-D = two-dimensional; 3-D = three-dimensional; CRT = Card Rotations Test; ns = non-significant; PVRT = Purdue Visualization of Rotations Test. a ANCOVA with pretest scores as the covariate (N = 384).

tures. The 2-D images group students were e-mailed with images of 11 molecules. At the end of the activity, both the students group completed a VSEPR report on these 11 molecules, discussing electronic and molecular geometries and drawing Lewis dot structures. Procedure

The study began in the fifth week of the semester. During the first week, participants completed the Purdue Visualization of Rotations and the Card Rotations pretests. In the following 2 weeks, participants completed the instructional activity of the Molecule Game. By the seventh week, participants completed the VSEPR theory pretest. In the ninth week of the class, participants completed the activity of the Chemist as an Artist. In the tenth week, the participants began the instructional activity of the Tower of VSEPR Theory, which continued for 3 weeks. In the 13th week, participants took the two spatial ability posttests, the PVRT and the CRT, and completed the survey and the long report on VSEPR theory. In the 14th and the final week of this study, students completed the VSEPR theory posttest. To help the experimental group with the technical issues of SL, several orientation videos and handouts were developed to guide students. In addition, regular office hours were held, both in a computer lab and in the study area. © 2013 John Wiley & Sons Ltd

Three doctoral students observed the instructor to judge the instructor’s consistency on content delivery and usage of instructional strategies in teaching the two sections. The observers were civil engineering, entomology and computer science major students with 1–3 years of teaching experience. They observed the instructor for four consecutive classes on the VSEPR theory. The inter-rater reliability of the observers’ ratings ranged between 75 and 100%. Results

Means and standard deviations for the pretest assessment scores for all measures are shown in Table 2. Independent-sample t-tests were conducted to examine pre-existing differences between the 3-D virtual reality group and the 2-D images group on the three parts of the VSEPR learning test (i.e., molecule angles, molecular geometry, species identification). The results depicted in Table 2 show that there were no statistically significant differences in the pretest scores on any of the measures of students’ knowledge of VSEPR theory. Table 2 also shows that analyses of covariance (ANCOVAs) using pretest scores as the covariate failed to support the hypothesis that the SL environment would enhance students’ understanding of any VSEPR theory on any of the measures of knowledge of VSEPR theory (H1a–H1c).

0.62ns 124.01 124.10 Note. ANCOVA = analysis of covariance; CRT = Card Rotations Test; PVRT = Purdue Visualization of Rotations Test; ns = non-significant.

0.02ns 3.92 4.16 11.22 11.30 0.22ns 0.97 1.03 1.16 1.21 0.17ns 1.27 1.34 1.10 1.04 0.89 0.73

2.24ns

3.51 3.44

1.10 0.93 3.82 3.88 1.20 1.22

87 43 44 297 161 136

0.93 1.02

0.42ns

3.58 3.54 1.62ns 1.04 1.09 0.73 0.88

31.49 24.47

3.07ns 126.86 137.25 1.11ns 3.89 3.84 13.23 12.51 2.70ns 0.01ns

1.16 1.51

1.01 1.03

0.0ns 1.01 0.98 1.22 1.24

0.10ns

1.20 1.20

1.08 0.99 1.43 1.11 0.63ns 1.34 1.26 3.53 3.59 1.91 1.15

138 68 70 244 136 108

0.96 0.86

0.74ns

22.09 16.64

0.00ns 124.69 121.11 0.01ns 4.06 3.97

12.13 11.93 3.44ns

11.55 11.27

4.37 3.88

SD

Mean F SD

Mean F SD

F SD

Mean n

Molecular angles

30.16 25.68

0.23ns 18.16 27.03 131.92 131.21

SD

Mean F

CRT PVRT Species identification Molecular geometry

Mean

Gender Male Experimental Control Female Experimental Control Gaming experience Gamers Experimental Control Non-gamers Experimental Control

The current study tested the premise that working in a 3-D virtual environment created in SL would enhance general chemistry students’ spatial ability and chemistry-related achievements. Many chemistry con-

Measures/Subgroup analyses variables

Discussion

Table 3. Subgroup Analyses of Measures of Learning Outcomes Using ANCOVA Results of Pretest as a Covariate

Likewise, there were no statistically significant pretest differences between the 3-D virtual reality group and the 2-D images group on either the PVRT or the CRT scores, the two measures of spatial ability employed in this study. Further, ANCOVAs of PVRT and CRT failed to support the hypothesis that the 3-D virtual learning environment would improve students’ spatial abilities (H2a–H2b). We further conducted post hoc analyses to explore differences in subgroup performance on the VSEPR theory and the spatial ability tests. We used the variables of gender, gaming experience and pretest spatial ability scores as measured separately by PVRT and CRT. The pretest CRT and PVRT scores were used individually to calculate a median score because PVRT measures students’ abilities to view and rotate objects in 3-D space and CRT is a measure of the ability to view and rotate things in 2-D space. Students who scored below the median score were treated as having low spatial ability levels, and students who scored above the median score were identified as high on spatial ability for the respective measures. We conducted ANCOVAs to find the differences between groups based on the above listed variables and results are presented in Tables 3 and 4. We used pretest as a covariate when we conducted a subgroup analysis with variables of gender and gaming experience. When we conducted a subgroup analysis using the median split on PVRT and CRT pretest scores, we also included gaming experience as a covariate along with the VSEPR theory pretest scores because students’ gaming experiences can also impact their spatial ability levels and hence may confound the results. The subgroup analysis of part 1 of the VSEPR test (molecular angles) in which groups were created by a median split on CRT pretest scores revealed that students with low CRT scores in the 3-D virtual reality environment group showed better performance on the molecule angles items than those with low CRT score in the 2-D images group. None of the other subgroup comparisons yielded statistically significant differences.

Z. Merchant et al.

F

586


1.71ns 113.04 108.00


Note. ANCOVA = analysis of covariance; CRT = Card Rotations Test; PVRT = Purdue Visualization of Rotations Test; ns = non-significant. *Coefficient is significant at the 0.05 level (two tailed).

1.12ns 4.03 3.98 10.96 10.36 0.39ns 1.05 1.02 1.29 1.19 0.05ns 1.27 1.31 1.17 0.99 1.00 0.64

4.29*

3.49 3.49

3.57 3.65 0.02ns 1.09 1.14 0.85 0.88

211 103 108 167 97 70

1.03 0.94 0.61 0.55

22.02 34.29

2.04ns 1.92 1.18

0.13ns

1.28 1.15

1.04 0.95

1.29ns

12.12 12.61

4.17 3.77

0.75ns

136.16 139.77

20.17 15.82

0.43ns 34.51 23.37 1.08 1.14 3.26 3.29

1.14 1.16

PVRT High Experimental Control Low Experimental Control CRT High Experimental Control Low Experimental Control

206 106 100 178 98 80

1.20 0.96

0.23ns

3.79 3.83

1.33 1.30

1.20 1.10

0.13ns

1.40 1.23

0.93 1.02

0.20ns

10.09 9.78

3.83 3.78

0.06ns

132.50 136.06 3.72 3.65 13.21 13.01 1.04 1.01

1.58ns 0.02ns 3.08ns

587

116.49 116.40

21.88 19.91

0.31ns 0.04ns

Mean SD

Mean F SD

Mean F SD SD

F

Mean

PVRT Species identification Molecular geometry Molecular angles

Mean n Measures/Subgroup analyses variables

Table 4. Subgroup Analyses of Measures of Learning Outcomes Using ANCOVA with Pretest Scores and Gaming Experience as a Covariate

F

CRT

SD

F


cepts such as VSEPR theory are abstract and can be difficult to understand. Virtual environments like SL have the affordances to represent molecule structures in 3-D space. Students can then visualize the molecules and interact with these structures to deepen the understanding of the VSEPR model. Although this study did not find a support for the hypothesis that chemistry instruction presented in a 3-D virtual environment can enhance spatial ability and chemistry learning when analysing the class as a whole, we found an interesting subgroup difference between the 3-D virtual environment group and 2-D images group. Students identified as having low spatial abilities as measured by the CRT pretest benefited from the 3-D virtual reality-based instruction and showed better performance compared to the 2-D images group. CRT, as opposed to PVRT, is a measure of a person’s ability to view, rotate and manipulate objects in a 2-D space. This result highlights the effectiveness of the instruction delivered in the 3-D virtual reality environment because students who had the most difficulty conducting 2-D mental rotation were now able to think in a 3-D space and translate the same into 2-D perspective after receiving spatial instruction in SL. Our study’s result also contributed in shedding some light on the perpetual debate that exists in the field about gender differences in performance on spatial ability measures. Our study found no statistical difference between the men and women on their performance on the parts of VSEPR theory test, which could be viewed in a positive light. Many studies in the literature have reported gender differences in the learning outcome of spatial instruction. Researchers have observed that traditionally men have performed better than the women on the spatial ability measures (e.g., Estes & Felker, 2012; Pietsch & Jansen, 2012). Concurrently, there are studies where gender differences on spatial ability were mitigated by spatial instruction (Contreras, Martínez-Molina, & Santacreu, 2011). The results of our study support the latter view of the field, that is, imparting spatial instruction can alleviate the gender differences on performance related to spatial ability measures. This study did not take into account the teaching quality of the instructor. The literature attests to the positive effects of instructional effectiveness on students’ learning achievements (Trigwell & Prosser, 1991). Because the same instructor taught both classes, and this instructor has been recognized as an

588

excellent teacher, it is likely that there was very little room left for further improvement in students’ performances on the various tests. In addition, the three assignments that students completed in SL or with 2-D images were review activities designed over and above the regular classroom teaching sessions. Future studies must be conducted to examine the unique impact of 3-D virtual environment-based instruction compared to other forms of instruction with more extensive treatment time. Furthermore, in the future, results can be analysed controlling such factors as amount of classroom instruction time and teacher quality, which may diminish the impact of 3-D virtual environment-based instruction. Additionally, offering spatial instruction to students at other universities, to see if this outcome is consistent, could further validate the results of this study. Currently, the generalizability of the results is limited to only those students included in the study. Finally, although the goal of imparting the instruction in 3-D virtual environment was to allow students to think in a 3-D space, the assessment method used to evaluate their learning used a 2-D format. The students in the virtual reality environment may have developed the ability to think in a 3-D space but not the ability to articulate this understanding when assessed using 2-D assessments. This highlights the importance of designing assessment strategies that are consistent with the format in which the instruction was imparted. Importance of this study

Improving science academic achievement, particularly for lower achieving students, has been a great concern for educators. For a critical chemistry topic, the 3-D nature of molecules, our study showed that working in 3-D environments like SL can significantly improve the performance of students who need help the most – the students with lower spatial abilities. This is imperative since student gains in spatial ability lead to student gains in science academic performance according to the literature. This study also highlighted the importance of considering students’prior experiences with the instructional technology since sophisticated technology such as SL can pose an extraneous cognitive load on students. References Barnea, N., & Dori, Y. J. (1999). High-School chemistry students’ performance and gender differences in a

Z. Merchant et al.

computerized molecular modeling learning environment. Journal of Science Education and Technology, 8, 257–271. Bodner, G. M., & Guay, R. B. (1997). The Purdue visualization of rotations test. Chemical Educator, 2(4), 1–17. doi: 10.1007/s00897970138a Carter, C., LaRussa, M., & Bodner, G. (1987). A study of two measures of spatial ability as predictors of success in different levels of general chemistry. Journal of Research in Science Teaching, 24, 645–657. Coleman, S., & Gotch, A. (1998). Spatial perception skills of chemistry students. Journal of Chemical Education, 75, 206–209. Contreras, M. J., Martínez-Molina, A., & Santacreu, J. (2011). Do the sex differences play such an important role in explaining performance in spatial tasks? Personality and Individual Differences, 52, 659–663. Copolo, C. E., & Hounshell, P. B. (1995). Using threedimensional models to teach molecular structures in high school chemistry. Journal of Science Education and Technology, 4, 295–305. doi: 10.1007/BF02211261 Dalgarno, B., Hedberg, J., & Harper, B. (2002). The Contribution of 3-D Environments to Conceptual Understanding. In A. Williamson, C. Gunn, A. Young, & T. Clear (Eds.), Winds of change in the sea of learning: Charting the course of digital education, proceedings of the 19th annual conference of the Australasian Society for Computers in Learning in Tertiary Education (pp. 149–158). Auckland, NZ: UNITEC Institute of Technology. Retrieved from http://www.ascilite.org.au/conferences/auckland02/ proceedings/papers/051.pdf Ekstrom, R. B., French, J. W., Harman, H. H., & Dermen, D. (1976). Manual for kit of factor referenced cognitive tests. Princeton, NJ: Educational Testing Service. Estes, Z., & Felker, S. (2012). Confidence mediates the sex difference in mental rotation performance. Archives of Sexual Behavior, 3, 557–570. Ferk Savec, V., Vrtacˇnik, M., & Gilbert, J. K. (2005). Evaluating the educational value of molecular structure representations. In J. K. Gilbert (Ed.), Visualization in science education (Models and modeling in science education, Vol. 1, pp. 269–300). Dordrecht: Springer. Harle, M., & Towns, M. (2011). A review of spatial ability literature, its connection to chemistry, and implications for instruction. Journal of Chemical Education, 88, 351–360. doi: 10.1021/ed900003n Hedberg, J., & Alexander, S. (1994). Virtual reality in education: Defining researchable issues. Educational Media International, 31, 214–220. doi: 10.1080/ 0952398940310402 Limniou, M., Roberts, D., & Papadopoulos, N. (2008). Full immersive virtual environment CAVETM in chemistry



education. Computers and Education, 5, 584–593. doi: 10.1016/j.compedu.2007.06.014 Lohman, D. F. (1988). Spatial abilities as traits, processes, and knowledge. In R. J. Sternberg (Ed.), Advances in the psychology of human intelligence (Vol. 4, pp. 181–248). Hillsdale, NJ: Lawrence Erlbaum Associates. Mathewson, J. H. (1999). Visual-Spatial thinking: An aspect of science overlooked by educators. Science Educator, 83, 33–54. Merchant, Z., Goetz, E., Keeney-Kennicutt, W., Kwok, O., Cifuentes, L., & Davis, T. J. (2012). The learner characteristics, features of desktop 3-D virtual reality environments, and college chemistry instruction: A structural equation modeling analysis. Computers and Education, 59, 551– 568. Mohler, J. L. (2008). A review of spatial ability research. Engineering Design Graphics Journal, 72, 19–30. Moore, P. (1995). Learning and teaching in virtual worlds: Implications of virtual reality for education. Australian Journal of Educational Technology, 11, 91–102. Ozmen, H. (2008). The influence of computer-assisted instruction on students’ conceptual understanding of chemical bonding and attitude toward chemistry: A case of Turkey. Computers and Education, 51, 423–438. Pellegrino, J., & Hunt, E. (1991). Cognitive models for understanding and assessing spatial abilities. In H. Rowe (Ed.), Intelligence: Reconceptualisation and measurement (pp. 203–225). Melbourne: ACER & Hillsdale, NJ: Lawrence Erlbaum Associates. Pellegrino, J. W., & Kail, R. (1982). Process analyses of spatial aptitude. In R. Sternberg (Ed.), Advances in the psychology of human intelligence (Vol. 1, pp. 311–366). Hillsdale, NJ: Erlbaum. Piburn, M., Reynolds, S., Leedy, D., McAuliffe, C., Birk, J., & Johnson, J. (2002). The hidden earth: Visualization of geologic features and their subsurface geometry. Paper


589

presented at the National Association for Research in Science Teaching (NARST) annual meetings, April 7–10, 2002, New Orleans, Louisiana. Pietsch, S., & Jansen, P. (2012). Different mental rotation performance in students of music, sport and education. Learning and Individual Differences, 22, 159–163. Reynolds, C. R., Livingston, R. B., & Wilson, V. (2009). Measurement and assessment in education (2nd ed.). Boston: Allyn & Bacon. Robichaux, R. R., & Guarino, A. J. (2000, November). Predictors of visualization: A structural equation model. Paper presented at the Annual Meeting of the Mid-South Educational Research Association, Bowling Green, KY. Second Life.com. (2011, October). What is Second Life®. Retrieved on January 8, 2012 from http://secondlife.com/ whatis/ Shepard, R. N., & Cooper, L. A. (1982). Mental images and their transformation. Cambridge, MA: MIT Press/ Bradford Books. Trigwell, K., & Prosser, M. (1991). Improving the quality of student learning: The influence of learning context and student approaches to learning on learning outcomes. Higher Education, 22, 251–266. Tuckey, H., Selvaratnam, M., & Bradley, J. (1991). Identification and rectification of student difficulties concerning three-dimensional structures, rotation, and reflection. Journal of Chemical Education, 68, 460–464. doi: 10.1021/ed068p460 Urhahne, D., Nick, S., & Schanze, S. (2009). The effect of three-dimensional simulations on the understanding of chemical structures and their properties. Research in Science Education, 39, 495–513. doi: 10.1007/s11165008-9091-z Wu, H., & Shah, P. (2004). Exploring visuospatial thinking in chemistry learning. Science Education, 88, 465– 492.

590

Z. Merchant et al.

Appendix I Itemized description of each instrument Variables Chemistry Learning Test

Items

Source

VSEPR – molecule angles

1. A typical 3-D representation of a molecule in 2-D space uses wedges for bonds coming towards the viewer, dotted lines for bonds going away from the viewer, and lines for bonds in the plane of the paper. What is the bond angle (rounded to the nearest whole number) expressed by the red dotted line in this molecule? Appendix Figure 1a a) 30° b) 45° c) 60° d) 90° e) 109° f) 120° g) 150° h) 180° Correct response: 120° 2. A typical 3-D representation of a molecule in 2-D space uses wedges for bonds coming towards the viewer, dotted lines for bonds going away from the viewer, and lines for bonds in the plane of the paper. What is the bond angle (rounded to the nearest whole number 0) expressed by the red dotted line in this molecule? Appendix Figure 1b a) 30° b) 45° c) 60° d) 90° e) 109° f) 120° g) 150° h) 180° Correct response: 90° 3. A typical 3-D representation of a molecule in 2-D space uses wedges for bonds coming towards the viewer, dotted lines for bonds going away from the viewer, and lines for bonds in the plane of the paper. What is the bond angle (rounded to the nearest whole number) expressed by the red dotted line in this molecule? Appendix Figure 3 a) 30° b) 45° c) 60° d) 90° e) 109° f) 120° g) 150° h) 180° Correct response: 109°

Self-developed

VSEPR – molecular geometry

4. You are given two 3-D views of the same species. Pick the correct molecular geometry. Appendix Figures 4a and 4b a) bent or angular b) quadrangular c) triangular d) hexagonal e) see-saw f) trigonal bipyramidal h) square planar i) trigonal planar j) octahedral k) square pyramidal l) trigonal pyramidal m) pentagonal n) tetrahedral o) T-shaped Correct response: tetrahedral 5. You are given two 3-D views of the same species. Pick the correct molecular geometry. Appendix Figures 5a and 5b a) bent or angular b) quadrangular c) triangular d) hexagonal e) see-saw f) trigonal bipyramidal h) square planar i) trigonal planar j) octahedral k) square pyramidal l) trigonal pyramidal m) pentagonal n) tetrahedral o) T-shaped Correct response: octahedral 6. You are given two 3-D views of the same species. Pick the correct molecular geometry. Appendix Figures 6a and 6b a) bent or angular b) quadrangular c) triangular d) hexagonal e) see-saw f) trigonal bipyramidal h) square planar i) trigonal planar j) octahedral k) square pyramidal l) trigonal pyramidal m) pentagonal n) tetrahedral o) T-shaped Correct response: bent or angular 7. You are given two 3-D views of the same species. Pick the correct molecular geometry. Appendix Figures 7a and 7b a) bent or angular b) quadrangular c) triangular d) hexagonal e) see-saw f) trigonal bipyramidal h) square planar i) trigonal planar j) octahedral k) square pyramidal l) trigonal pyramidal m) pentagonal n) tetrahedral o) T-shaped Correct response: linear 8. You are given two 3-D views of the same species. Pick the correct molecular geometry. Appendix Figures 8a and 8b a) bent or angular b) quadrangular c) triangular d) hexagonal e) see-saw f) trigonal bipyramidal h) square planar i) trigonal planar j) octahedral k) square pyramidal l) trigonal pyramidal m) pentagonal n) tetrahedral o) T-shaped Correct response: see-saw

VSEPR – species identification questions

g) linear

g) linear

g) linear

g) linear

g) linear

9. You are given two 3-D views of the same species. Ignore the atom colours. Pick all the species with that shape. There may be more than one. Appendix Figures 9a and 9b a) H2S b) SO2 c) BeF2 d) CO2 e) BrF2– f) H2O g) CaCl2 Correct response: BeF2, CO2, BrF210. You are given two 3-D views of the same species. Ignore the atom colours. Pick all the species with that shape. There may be more than one. Appendix Figures 10a and 10b a) BF3 b) PBr3 c) CO32– d) BrF3 e) NH3 f) FeCl3 g) H3O+ Correct response: PBr3, NH3, H3O+ 11. You are given two 3-D views of the same species. Ignore the atom colours. Pick all the species with that shape. There may be more than one. Appendix Figures 11a and 11b a) CF4 b) SCl4 c) NH4+ d) SnCl4 e) AsF4– f) SiH4 g) BrF4+ Correct response: SF4, BrF4+
