LEARNING AND GENDER DIFFERENCES IN A

NARRATIVE-CENTERED LEARNING ENVIRONMENT

John L. Nietfeld, Lucy R. Shores and Kristin F. Hoffmann

Department of Curriculum and Instruction, North Carolina State University, Raleigh, NC, U.S.A.

Keywords: Narrative-centered learning environments, Gender differences.

Abstract: This study examined the performance of 8

grade students within a narrative-centered learning environment

(NLE) focused on science content. Students showed significant increases in science content knowledge after

gameplay. Performance in the NLE was related to higher science self-efficacy and also led to increased

presence, interest, and transfer of science concepts. No differences were found between genders for content

knowledge, however different variables predicted success by gender. In addition, males entered the study

with higher levels of reported gaming experience and showed early advantages for game score that

dissipated by the end of play. Further findings and implications for learning are discussed.

1 INTRODUCTION

Situating game-based learning exercises within a

narrative-centered approach is one strategy for

enhancing student learning and engagement. Stories

draw audiences into plots and settings, thereby

opening perceptual, emotional, and motivational

opportunities for learning. Fantasy contexts in

educational games have been shown to provide

motivational benefits for learning (Parker & Lepper,

1992). Narrative features such as pacing and tension

can introduce additional challenge to learning tasks

and contribute to student motivation. Establishing

concrete connections between narrative context and

pedagogical subject matter has also been said to

support the assimilation of new ideas in young

learners (Wells, 1986).

Narrative learning environments (NLEs)

leverage a range of techniques for providing

effective, engaging learning experiences. Multi-user

virtual environments such as Quest Atlantis (Barab

et al., 2007) and River City (Ketelhut et al., 2007)

use rich narrative settings to contextualize inquiry-

based science learning scenarios with strong social

and ethical dimensions. Several empirical studies

from this work have yielded promising learning

results that support the promise of NLEs in the

classroom.

Gender is also a variable of interest within

NLE’s. Many factors influence gender’s effects on

student performance, engagement, and interest in

game-based learning environments. These include

but are not limited to ability, beliefs, virtual gaming

preferences, and characteristics of the environment

itself. Characteristics that make video games

appealing to boys and girls have been noted as

consistently different (Cassell & Jenkins, 1998).

Girls tend to prefer story development, relationships,

and collaboration, whereas boys tend to prefer

competition and aggression (Cassell & Jenkins,

1998). Also, recent research has found that girls

report higher self-efficacy for science and self-

efficacy for self-regulated learning than boys at the

middle school level (Britner & Pajares, 2006).

Moreover, the gender of characters that student

interact with within virtual environments have been

shown to affect student attitudes about math and

science (Kim & Baylor, 2006). Therefore, the extent

to which these elements generalize and impact

performance in NLEs is an important line of

research.

A number of interesting gender differences have

emerged in recent studies using NLEs. For example,

males tend to engage in more off-task behaviors

(Rowe et al., 2009), females report higher levels of

presence (McQuiggan et al., 2008a), and females,

when allowed, record greater amounts of notes

(McQuiggan et al., 2008b). Further investigation of

these effects as well as other factors that influence

male and female performance motivation within this

context is imperative for the production of NLEs

that are effective for both genders.

140

L. Nietfeld J., R. Shores L. and F. Hoffmann K..

LEARNING AND GENDER DIFFERENCES IN A NARRATIVE-CENTERED LEARNING ENVIRONMENT.

DOI: 10.5220/0003349301400144

In Proceedings of the 3rd International Conference on Computer Supported Education (CSEDU-2011), pages 140-144

ISBN: 978-989-8425-49-2

 2011 SCITEPRESS (Science and Technology Publications, Lda.)

1.1 CRYSTAL ISLAND

In our laboratory we have developed an inquiry-

based NLE called CRYSTAL ISLAND - OUTBREAK (see

Figure 1)

in the domain of microbiology for middle

school students. CRYSTAL ISLAND - OUTBREAK

features a science mystery set on a recently

discovered volcanic island where a research station

has been established to study the unique flora and

fauna. The student plays the protagonist attempting

to ultimately discover that the milk on the island is

carrying an unidentified infectious disease by

utilizing resources at the research station. The story

opens by introducing the student to the island and

the members of the research team. As members of

the research team fall ill, it is the student’s task to

discover the cause of the specific source of the

outbreak. She is free to explore the world and

interact with other characters while forming

questions, generating hypotheses, collecting data,

and testing the hypotheses. Throughout the mystery,

the student can walk around the island and visit the

infirmary, the lab, the dining hall, and the living

quarters of each member of the team. The student

can pick up and manipulate objects, talk with

characters to gather clues about the source of the

disease. Facts and clues revealed during the

student’s interaction can be stored in a virtual

factsheet. Some characters within the game serve as

local experts on certain pathogenic diseases. During

the student’s interactions with such expert

characters, in-game quizzes are presented to test the

student’s knowledge of the character’s specific

expertise. In the course of the adventure, the student

must gather enough evidence to correctly identify

that is in fact the milk has been contaminated with

E-Coli. To win the game, the student must have

submit a correctly filled factsheet with information

about the source object, disease, and treatment.

Figure 1: The CRYSTAL ISLAND - OUTBREAK environment.

2 METHOD

An experiment was conducted involving the entire

eighth grade population of a North Carolina middle

school. The research questions included: 1. What

effect does completing the C

RYSTAL ISLAND -

OUTBREAK mystery have upon learning? 2. How

does gender affect gameplay?

2.1 Participants

A total of 130 eighth grade students ranging in age

from 12 to 15 (M = 13.28, SD = 0.48) participated in

our study. 73 were male and 57 were female.

Approximately 3% of the participants were

American Indian or Alaska Native, 2% were Asian,

32% were African American, 13% were Hispanic or

Latino, and 50% were European American. The

study was conducted prior to students’ exposure to

the microbiology curriculum unit of the North

Carolina state standard course of study in their

regular classes.

2.2 Materials

Gaming Demographics Survey. A three-question,

researcher-developed gaming demographics survey

was completed by the participants prior to

interacting with C

RYSTAL ISLAND - OUTBREAK. The

survey asked students to rate how frequently they

play games and their perceived game-playing skill

on a 5-point Likert scale. Also, the students were

asked to estimate how many hours they spend

playing games per week.

Science Learning Self-efficacy. A science

learning self-efficacy inventory was given prior to

the students’ interaction with C

RYSTAL ISLAND -

OUTBREAK. The inventory consisted of eight items

answered on a 5-point Likert scale and has been

shown to be internally reliable (Nietfeld, Cao, &

Osborne, 2006).

Perceived Interest Questionnaire (PIQ).

Situational interest was measured immediately

following the students’ interaction with C

RYSTAL

ISLAND - OUTBREAK. The perceived interest

questionnaire consists of ten items measured on a 5-

point Likert scale and has been shown to be

internally reliable (Schraw, 1997)

Presence Questionnaire (PQ). The Presence

Questionnaire is a 32-question validated measure

containing several subscales, including

involvement/control, naturalism of experience and

quality of interface. The Presence Questionnaire is

intended to measure a user’s perceptions of the

LEARNING AND GENDER DIFFERENCES IN A NARRATIVE-CENTERED LEARNING ENVIRONMENT

141

transparency of the actual technology. In other

words, this measure provides a quantitative

assessment of how immersed the user feels in the

CRYSTAL ISLAND - OUTBREAK environment.

Microbiology domain knowledge was measured

with a researcher-constructed, 16-item multiple-

choice test based upon North Carolina Standard

Course of Study curriculum and intended to measure

domain-related material integrated within the

learning environment. The measure was broken

down into 8 fact-level and 8 application-level

questions and administered before the students

interacted with C

RYSTAL ISLAND - OUTBREAK and

again immediately following gameplay.

Microbiology Transfer Tasks. Two researcher-

developed transfer tasks were created to test

students’ abilities to transfer skills and domain

knowledge utilized in the learning environment to a

similar microbiology-related task. Students

completed the two, short-answer transfer tasks

immediately following the interaction.

2.3 Procedure

Students completed the self-efficacy for science and

gaming demographic inventories two weeks prior to

the intervention. On the day of the experiment,

students received a brief orientation to the C

RYSTAL

ISLAND - OUTBREAK environment as well as

completed the microbiology content test. Students

interacted with the learning environment until they

completed the game or 60 minutes elapsed.

Immediately following the interaction, students

completed the microbiology posttest, PQ, and the

PIQ.

In-game actions were recorded by the system

during gameplay that included virtual book reading,

consulting a microbiology field manual,

conversations with non-player characters,

performance on in-game mini-quizzes, testing

objects for contaminants, diagnosis worksheet usage,

and completion of seven major sub-goals: talking to

the camp nurse, interviewing the two expert

characters, talking to the camp cook, working with

the laboratory scientist, testing the contaminated

object with the lab equipment, and submitting a

correct diagnosis worksheet. Diagnosis worksheet

usage was quantified based upon a summed total for

each entry.

Students received the most points for a correct

entry, half points for an incorrect entry, and no

points for not using the space at all. Also, a running,

progressive score was calculated based upon student

efficiency and performance during the interaction.

3 RESULTS

3.1 What Effect does Completing the

CRYSTAL ISLAND - OUTBREAK

Mystery have on Learning?

A hierarchical regression was performed to predict

performance on the microbiology post-test. As

microbiology prior knowledge was found to be

highly correlated with post-test performance (r =

.42), microbiology pre-test score was entered into

the first block and overall presence, self-efficacy for

science, diagnosis worksheet performance, final in-

game score, and goals completed were entered into

the second block. Both models were found to be

significant (F

(1, 126)

= 30.94, p < .001, R

= .20; F

(6,

121)

= 11.41, p < .001, R

= .36). For the second

model, only prior knowledge and number of goals

completed were found to be significant predictors.

Given these results, comparisons were made

between students who completed the C

RYSTAL

ISLAND - OUTBREAK mystery and those who did not

for variables related to self-regulated learning and

content learning gains. Overall, 58 students

completed the mystery. An ANCOVA controlling

for prior knowledge found students who completed

the C

RYSTAL ISLAND - OUTBREAK mystery scored

significantly better on the microbiology post-test

than those students who did not complete the

scenario (F

(2, 27)

= 26.67, p < .001). Moreover,

students who completed the mystery reported

significantly higher levels of presence (t

(126)

= 3.65,

p < .001), situational interest (t

(128)

= 2.45, p < .05),

diagnosis worksheet total (t

(128)

= 14.14, p < .001),

and performance on the transfer tasks (respectively,

(128)

= 2.49, p < .05; t

(128)

= 2.95, p < .01). Finally,

those who completed the mystery also had higher

initial levels of self-efficacy for science (t

(128)

= 4.41,

p < .001) than those who did not finish.

3.2 How does Gender Affect

Gameplay?

Prior to interacting with the environment, males

reported playing games more frequently (t

(128)

7.91, p < .01) and higher levels of perceived

gameplay skill (t

(128)

= 7.39, p < .01) than females.

During gameplay, males completed significantly

more goals (t

(128)

= 2.64, p < .01) and more

accurately and comprehensively completed the

diagnosis worksheet (t

(128)

= 2.14, p < .05) than

females. Significant gender differences in favour of

males were only found for the mandatory diagnosis

CSEDU 2011 - 3rd International Conference on Computer Supported Education

142

section (t

(128)

= 2.33, p < .05) not the hypothesis,

testing, or symptoms section. Interestingly, when

number of hours playing games was used as a

covariate there were no remaining gender

differences for goals completed or for the diagnosis

section of the diagnosis worksheet. Also, males

tended to receive significantly higher in-game scores

than females at the beginning but not at the end of

the interaction (see Table 1).

Table 1: Score by gender and minutes played.

Time

(minutes)

Males

(M, SD)

Females

(M, SD)

5 34.78 (24.42)** 23.26 (18.86)**

10 90.73 (98.32)* 56.12 (66.56)*

15 115.82 (110.89) 93.05 (100.38)

20 170.93 (180.81)* 115.19 (117.45)*

25 210.38 (239.39)* 135.07 (115.71)*

30 270.10 (320.63)* 155.95 (161.56)*

35 294.04 (395.15)* 181.11 (217.39)*

40 328.33 (385.07) 236.04 (287.46)

45 378.89 (413.54) 278.51 (329.26)

50 376.56 (419.75) 301.04 (354.21)

55 428.47 (434.08) 374.61 (393.05)

60 462.21 (442.19) 355.09 (399.79)

Note: * = p < .05, ** = p < .01

Of the 73 males that interacted with CRYSTAL

ISLAND - OUTBREAK 36 completed the mystery. For

the males, those who completed the mystery had

significantly higher levels of prior knowledge (t

(71)

3.05, p < .01), reported higher levels of overall

presence (t

(71)

= 2.26, p < .05), situational interest

(71)

= 2.26, p < .05), self-efficacy for science (t

(71)

3.30, p < .01), diagnosis worksheet total (t

(71)

= 9.74,

p < .001), performance on the microbiology post-test

(71)

= 4.41, p < .001) and performance on the

transfer tasks (t

(71)

= 2.45, p < .01; t

(71)

= 3.06, p <

.01). Of the 57 females, 22 of them completed the

RYSTAL ISLAND - OUTBREAK mystery. Considering

only the female population, those who completed the

mystery reported significantly higher levels of

overall presence (t

(54)

= 2.61, p < .05), self-efficacy

for science (t

(55)

= 2.84, p < .01), diagnosis

worksheet performance (t

(55)

= 10.25, p < .001), and

performance on the microbiology post-test (t

(55)

3.28, p < .01). No differences were found between

groups with microbiology prior knowledge.

Following the interaction with C

RYSTAL ISLAND -

OUTBREAK, males reported significantly higher

levels of overall presence than females (t (126) =

2.88, p < .01). Specifically, males reported

significantly higher levels of presence for the

involvement/control subscale (t (126) = 3.30, p <

.01) than females, but not for the naturalism or

interface subscales. Finally, comparisons of

microbiology prior knowledge (M = 6.26, SD =

1.99) and microbiology posttest knowledge (M =

8.66, SD = 2.98) revealed a significant increase in

scores overall (t

(129)

= 9.81, p < .001). However, no

differences were found between genders for content

learning gains, prior knowledge, or posttest

performance on microbiology. Both genders showed

significant increases in content knowledge (for boys,

(72)

= -7.76, p < .001; for girls, t

(56)

= -6.02, p <

.001).

4 DISCUSSION

Given the emerging field of on-line and computer-

based learning it is essential that appropriate

mechanisms are in place to encourage effective and

self-regulated learning for all students. The current

study reports on one particular attempt to examine

overall learning and differences associated with

gender within a game-based NLE.

In sum, playing C

RYSTAL ISLAND - OUTBREAK

led to significant content learning gains.

Performance was predicted both by prior knowledge

and goals completed within the game. The

pedagogical approach built into the character

interaction and student exploration was successful as

evidenced by the fact that students who completed

the mystery scored higher on the science posttest.

Moreover, these students reported higher levels of

engagement (presence, situational interest), strategy

use (diagnosis worksheet), and the ability to transfer

their knowledge than their peers who did not

complete the mystery.

Interesting gender differences also emerged

during interaction with C

RYSTAL ISLAND -

OUTBREAK. Boys who solved the mystery had

different predictors of success (e.g., prior

knowledge) than girls that finished. These findings

suggest girls and boys may be motivated to perform

well in these environments for different reasons.

Also, boys entered the study with higher perceived

skill and experience with computer games than girls.

LEARNING AND GENDER DIFFERENCES IN A NARRATIVE-CENTERED LEARNING ENVIRONMENT

143

Accordingly, they showed an advantage over girls in

the early stages of C

RYSTAL ISLAND - OUTBREAK

with respect to performance (goals completed) but

closed this gap to the point of statistically

nonsignificant differences by the end of play. Boys

still reported a higher level of presence after

gameplay but no differences were found between

genders on content knowledge. Boys also showed

advantages for overall game score and performance

on the diagnostic facet of the diagnosis worksheet

yet these advantages disappear when controlling for

prior experience playing games.

It is important to continue to identify such

differences as reported above in order to develop

gender-inclusive learning environments that promote

learning and motivation for all users. From a self-

regulated learning standpoint it will be important to

study means by which to scaffold strategy use,

metacognitive regulation, and motivational control

for all students in game-based environments such as

RYSTAL ISLAND – OUTBREAK. Future research

should take up this challenge of integrating effective

learning scaffolds within such engaging

environments because they offer a unique

opportunity to customize learning for the individual

learner in ways traditional learning environments

cannot.

ACKNOWLEDGEMENTS

The authors wish to thank members of the

IntelliMedia lab for their assistance, Omer

Sturlovich and Pavel Turzo for use of their 3D

model libraries, and Valve Software for access to the

Source

engine and SDK. This research was

supported by the National Science Foundation under

Grants REC-0632450, IIS-0757535, DRL-0822200,

IIS-0812291, and CNS-0540523. Any opinions,

findings, and conclusions or recommendations

expressed in this material are those of the authors

and do not necessarily reflect the views of the

National Science Foundation.

REFERENCES

Barab, S., Dodge, T., Tuzun, H., Job-Sluder, K., Jackson,

C., Arici, A., Job-Sluder, L., Carteaux, R., Gilbertson

Jr., J. & Heiselt, C. (2007). The Quest Atlantis project:

A socially-responsive play space for learning. In B. E.

Shelton & D. Wiley (Eds.) The Educational Design

and Use of Simulation Computer Games, (pp. 159-

186), Rotterdam, The Netherlands: Sense Publishers.

Britner, S. L., and Pajares, F. (2006). Sources of science

self-efficacy beliefs of middle school students. Journal

of Research in Science Teaching, 43, 485–499.

Cassell, J. & Jenkins, H. (1998). From Barbie to Mortal

Kombat: Gender and computer games. Cambirdge,

MA: MIT Press.

Egri, L. (1960). The Art of Dramatic Writing. New York,

NY: Simon and Schuster.

Harp, S. and Mayer, R. (1998). How seductive details do

their damage: A theory of cognitive interest in science

learning. Journal of Educational Psychology, 90(3),

414-434.

Ketelhut, D., Dede, C., Clarke, J., Nelson, B. & Bowman,

C. (2007). Studying situated learning in a multi-user

virtual environment. In E. Baker, J. Dickieson, W.

Wulfeck and H. O'Neil (Eds.) Assessment of Problem

Solving Using Simulations, (pp. 37-58), Lawrence

Erlbaum, Mahwah, NJ.

Kim, Y., Baylor, A. (2006). A social-cognitive framework

for pedagogical agents as learning companions.

Educational Technology Research & Development,

54, 569-590.

McQuiggan, S, Rowe, J., Lee, S., & Lester, J. (2008a).

Story-Based Learning: The Impact of Narrative on

Learning Experiences and Outcomes. In Proceedings

of the Ninth International Conference on Intelligent

Tutoring Systems, Montreal, Canada, pp. 530-539.

McQuiggan, S., Goth, J., Ha, E. Rowe, J. & Lester, J.

(2008b). Student Note-Taking in Narrative-Centered

Learning Environments: Individual Differences and

Learning Outcomes. In Proceedings of the Ninth

International Conference on Intelligent Tutoring

Systems (ITS-2008), Montreal, Canada, 510-519.

Nietfeld, J. L., Cao, L., & Osborne, J. W. (2006). The

effect of distributed monitoring exercises and feedback

on performance and monitoring accuracy.

Metacognition and Learning, 2, 159-179.

Parker, L. E., & Lepper, M. R. (1992). The effects of

fantasy contexts on children's learning and motivation:

Making learning more fun. Journal of Personality and

Social Psychology, 62, 625-633.

Schraw, G. (1997). Situational Interest in Literary Text.

Contemporary Educational Psychology. 22, 436-456.

Wells, C. (1986). The Meaning Makers: Children

Learning Language and Using Language to Learn,

Portsmouth, NH: Heineman

CSEDU 2011 - 3rd International Conference on Computer Supported Education

144