The Play Experience Scale: Development and Validation of a Measure of Play

Running Head: THE PLAY EXPERIENCE SCALE 1

The Play Experience Scale: Development and Validation of a Measure of Play

Davin Pavlas1,3

, Florian Jentsch1,3

, Eduardo Salas1,3

, Stephen M. Fiore2,3

, & Valerie Sims1

1Department of Psychology

2Department of Philosophy

3Institute for Simulation & Training

University of Central Florida

Word count:

4,026 (excludes abstract, précis, key points, and references)

Word limit:

4,500 (research report)

THE PLAY EXPERIENCE SCALE 2

Abstract

Objective: A measure of play experience in video games was developed through

literature review and two empirical validation studies.

Background: Despite the considerable attention given to games in the behavioral

sciences, play experience remains empirically under-examined. One reason for this gap is the

absence of a scale that measures play experience.

Methods: In Study 1, the initial Play Experience Scale (PES) was tested through an

online validation that featured three different games (N = 203). In Study 2, a revised PES was

assessed with a serious game in the laboratory (N = 77).

Results: Through principal component analysis of the Study 1 data, the initial 20-item

PES was revised, resulting in the 16-item PES. Study 2 showed the PES-16 to be a robust

instrument with the same patterns of correlations as in Study 1 via: (a) internal consistency

estimates, (b) correlations with established scales of motivation, (c) distributions of PES-16

scores in different game conditions, and (d) examination of the average variance extracted

(AVE) of the PES and the intrinsic motivation scale (IMS).

Conclusion: We suggest that the PES is appropriate for use in further validation studies.

Additional examinations of the scale are required to determine its applicability to other contexts

and its relationship with other constructs.

Application: The PES is potentially relevant to human factors undertakings involving

video games, including: basic research into play, games, and learning; prototype testing; and

exploratory learning studies.

Keywords: serious games, play, measure, motivation, game, PES

THE PLAY EXPERIENCE SCALE 3

Précis: Though games and play are of growing interest to the human factors community, no

psychometric scale of play exists. A play experience scale was created and validated through two

empirical studies. Analysis suggests the Play Experience Scale (PES) is a valid instrument for

investigating play in video games.

THE PLAY EXPERIENCE SCALE 4

The Play Experience Scale:

Development and Validation of a Measure of Play

Play is an essential component of the human experience. Humans seek play not only as

children but also as adults (Anderson, 1998). Play can even be found outside of the human

species, as animals have been shown to engage in spontaneous play (Van Leeuwen & Westwood,

2008). Despite the pervasiveness of play activities, there is a surprising paucity of play

assessment instruments in the literature. Although there are measures of playfulness (i.e., the

propensity of an individual to engage in play; Webster & Martocchio, 1992) and observational

ratings scales for child play exist throughout the developmental literature (e.g., Knox, 1997), a

self-report scale of play behavior is markedly absent. This gap is even more surprising given the

recent surge in investigations into the science of games (e.g., Chertoff, Jerome, Martin, & Knerr,

2008; Cotton, Mayes, Jentsch, & Sims, 2001; Pavlas, Heyne, Bedwell, Lazzara, & Salas, 2010).

Although the human factors community has embraced games for training (e.g., Gopher, Weil, &

Bareket, 1994; Lampton, Bliss, Orvis, Kring, & Glenn, 2009) and examined the basic science of

games, the field has not more closely examined the psychological correlates of game play.

The present research expands upon these recent undertakings through the development

and initial validation of the Play Experience Scale (PES). First, a brief description of the

construct of play is provided. Then, the contents of the initial prototype are described, as are two

empirical studies to test its validity and establish its psychometric properties. Finally, results and

implications for future research are presented.

THE PLAY EXPERIENCE SCALE 5

Play

Intuitively, play is simple to recognize. Defining play in a scientifically useful manner is

more difficult. On the most basic level, play is an intrinsically motivated (autotelic) activity free

of extrinsic goals or consequences (O'Connor & LaPoint, 1980), even though extrinsic

motivators are often present in sports, games for learning, and performances, i.e., in activities

that also include play as a component. These activities can further be characterized based on the

player’s degree of attention—that is, whether it is focused on their play behavior rather than their

external needs or goals (Calleja, 2007; Cowley, Charles, Black, & Hickey, 2008).

Beyond attention and motivation, the notion of freedom within rule frames is key to

understanding play (Gordon, 2008; Salen & Zimmerman, 2004). Zimmerman describes play as

"the free space of movement within a more rigid structure," noting that, "play exists because of

and also despite the more rigid structures of a system" (Zimmerman, in Esposito, 2005, p. 3).

Finally, play may be construed as a modifier for other behaviors, rather than as its own behavior

(Millar, 1968). This distinction is important, as it establishes that play is not a behavior unto

itself but rather something that can contain a wide range of behaviors. Even work activities can

be play; given, however, that work is often considered the opposite of play (Gitlin-Weiner,

1998), it is reasonable to assume that work-play is ―less‖ play-like due to its explicit external

contingencies.

Scale Constructs and Content

Based on the play literature described in the preceding section, we defined play as

behaviors that: (a) are predominantly intrinsically motivated; (b) are performed freely by the

player; (c) require and capture the player’s attention; and (d) are not contingent by external

rewards or consequences. Using these four definitional elements (see Table 1), a candidate Play

THE PLAY EXPERIENCE SCALE 6

Experience Scale (PES) was developed. Responses to the 20 items of the PES-20 were indicated

via a six-point Likert scale ranging from ―strongly disagree‖ (= 1) to ―strongly agree‖ (= 6).

Though the items were created for the context of video games, care was taken to ensure that the

wordings could be changed to refer to other play contexts. In the following sections, the

theoretical underpinnings of these categories are discussed.

Insert Table 1 about here

Autotelic Experience

The first set of items in the PES-20 assessed intrinsic motivation through the lens of

autotelic experience. When an experience is autotelic, an individual engages in it solely for its

own rewards (Cowley, Charles, Black, & Hickey, 2008). In other words, the experience is

intrinsically motivating (Malone, 1981). Consequently, the wording of the items in this category

reflected an internal source of drive (e.g., ―I wanted to‖ and ―I felt like‖).

Freedom

The second set of items targeted freedom. When an individual is free in a play context,

they are able to perform the actions they wish to perform. The notion of self-determination

within the artificial rule set of a play experience is inherent to modern definitions of play (Salen

& Zimmerman, 2004), and thus motivated items assessing freedom as a concept. Items in this

category reflected this direction, referring to the ―wants‖ of the individual within play, as well as

their ability to act on those wants (e.g., self-determination in leisure motivation; Weissinger &

Bandalos, 1995).

THE PLAY EXPERIENCE SCALE 7

Focus

The focus category targeted the states of immersion, concentration, and attentional

narrowing in definitions of play. Immersion is a popular topic in the games literature, describing

a state of intense concentration that is related to engagement (McMahan, 2003) and flow

(Csikszentmihalyi, 1990). When an individual is immersed in their play experience, the salience

of non-play concerns diminishes (Cowley, Charles, Black, & Hickey, 2008). In other words,

individuals who are immersed are focused on the task at hand; their attention and concentration

is allocated to the play task. Items in this category reflected this loss of concern and focused

concentration.

Absence of Extrinsic Motivation

The fourth set of items was based on the construct of extrinsic motivation. Unlike

intrinsic motivation, which emerges from the individual, extrinsic motivation is created through

external impositions (Ryan & Deci, 2000). Most commonly, these impositions take the form of

rewards or punishments (Benabou & Tirole, 2003). Items in this category addressed whether the

respondents felt there were consequences to their play, whether via evaluative judgment by

another entity or through real-world consequences for task performance.

Direct Play Assessment

The last set of questions in the draft PES assessed whether the individual believed they

engaged in play. Unlike the others, this category did not target any particular aspect of play, but

simply play itself. Because of the relative scarcity of play research, there was little to guide the

creation of these questions other than the inherent opposition between work and play (e.g.,

O’Connor & LaPoint, 1980; Yee, 2005). Additionally, these items directly referenced play,

whereas the other four categories did not directly employ the word ―play.‖

THE PLAY EXPERIENCE SCALE 8

Study 1

The 20-item Play Experience Scale (PES-20) was tested via a game-based experiment to

establish its psychometric properties and theoretical alignment.

Method

Participants. Two-hundred-twenty-six undergraduates (80 male, 146 female) enrolled at

the University of Central Florida participated for course credit. Participants’ ages ranged from 18

to 58 years, with a mean age of 21.02 years (SD = 4.95).

Materials. Three games were used as the conditions for the study: (a) a Flash version of

the classic computer game Tetris, (b) the adaptive game flOw, and (c) a custom-designed letter

search task. The search task was purposely designed to be ―boring,‖ comprising only the

minimum elements required to make the task a game (Garris, Ahlers, & Driskell, 2002).

Measures. To determine the psychometric properties and theoretical alignment of the

Play Experience Scale, the PES-20 was joined by measures assessing playfulness, intrinsic

motivation, and extrinsic motivation.

Computer Playfulness Scale. As a measure of trait playfulness (Webster & Martocchio,

1992), the Computer Playfulness Scale (CPS) served as a potential indicator of the PES’s

suitability. The seven-item Likert-scale was anchored from ―strongly disagree‖ (= 1) to ―strongly

agree‖ (= 7).

Intrinsic Motivation Scale. Given the importance of intrinsic motivation to the definition

of play, a measure thereof was key to establishing the PES’s validity. An outcome Intrinsic

Motivation Scale (IMS) measuring respondent reaction to items assessing interest, tension, effort,

and competence (McAuley, Duncan, & Tammen, 1989) was used. The 18-item scale had six

Likert-style anchors ranging from ―strongly disagree‖ to ―strongly agree‖.

THE PLAY EXPERIENCE SCALE 9

Situational Motivation Scale. The external regulation and amotivation components of the

Situational Motivation Scale (SIMS) served as the study’s measure of extrinsic motivation

(Guay, Vallerand, & Blanchard, 2000). The 16-item scale had six Likert-style anchors ranging

from ―strongly disagree‖ to ―strongly agree‖.

Design. The experiment used a 3 x 2 mixed-model design. The between-subjects variable

was ―game type‖ with three levels (i.e., Tetris vs. flOw vs. letter search task). The two-level

within-subjects variable was trial block (first vs. second).

Procedure. Participants were assigned to one of the three game conditions. After

obtaining answers to demographics and the CPS, participants were tasked with playing their

assigned game for five minutes and completing three measures: the PES-20, IMS, and SIMS.

Participants then played a second five-minute round, and again completed the PES-20, IMS, and

SIMS.

Results

Of the 226 participants, 203 provided complete data sets (i.e., completed both game

rounds and surveys). There were no significant differences between participants who completed

and those who dropped from the experiment in terms of age, gender, or condition. Table 2 shows

the means, standard deviations, and correlations among the primary study variables. For the

following analyses, the second-round responses were used (unless otherwise noted).

Insert Table 2 about here

Data Screening. First, the data were assessed for their suitability for factor analyses.

Nineteen of the 20 items were significantly correlated (p < .05) at least r (201) = .25 with another

THE PLAY EXPERIENCE SCALE 10

item, suggesting their suitability for factor analysis. Similarly, the Kaiser-Meyer-Olkin measure

of sampling adequacy was .88, which was well above the recommended value of .60. Finally,

Bartlett’s test of sphericity was significant (2(190) = 2256.83, p < .001). Based on these

statistics, we deemed factor analysis appropriate.

Candidate Scale Structure and Pruning. The underlying structure of the scale was

examined to determine whether it reflected the previously defined dimensions. A principal

component analysis (PCA) extraction with manual five-component truncation was followed by

oblique rotation via Promax (due to expected inter-correlations). Results suggested that the

original five-factor theory was not represented in the data. Instead, the first component consisted

of a merging of the focus and autotelic experience subscales. The fifth component of the scale

consisted of the reverse coded items of the scale.

Next, the scale was pruned to remove four flawed or ineffective items: Item 16 failed to

load onto any of the components and was dropped alongside component 1’s three lowest-loading

items. These changes resulted in a symmetrical 16-item scale (PES-16), which was examined

once more using PCA with Promax rotation. In this iteration, only the components with

Eigenvalue > 1 were retained, which removed Item 4 from the scale due to its ambiguous loading

pattern. A test of the diagnosticity of each element of the scale for the game condition variable

using multiple linear regression (see Table 3) further removed Item 1.

Insert Table 3 about here

A final PCA using Promax was then conducted to examine the properties of the resultant

14-item scale (PES-14). Though the scale was not symmetrical (see Table 4), its items loaded

THE PLAY EXPERIENCE SCALE 11

cleanly onto theory-consistent components. A confirmatory factor analysis of the four-factor

model suggested the PES-14 showed good fit (2(71)=158, p < .001; GFI = .90; SRMR = .06).

This version of the scale was also significantly and positively correlated with playfulness (r(201)

= .20, p < .001). Given these indicators, the PES-14 was used for further analysis.

Insert Table 4 about here

Reliability. Responses to the scale were very slightly skewed (skew = -0.05) and mildly

platykurtic (kurtosis = -0.57). Cronbach’s alpha was calculated for both administrations of the

PES-14. The reliability estimates for the both administrations of the PES-14 were high; alpha

equaled .83 in the first, and .86 in the second administration of the PES-14. Test-retest reliability

was established by examining correlations between the first and second administration of the

PES-14 (r (201) = .73, p < .001).

Convergent Validity. As the construct of play is conceptually similar to that of state-

based intrinsic motivation, correlations between the PES and the IMS were calculated (see Table

5). The correlation between the scales was high (r(201) = .79, p < .001), supporting the

convergent validity of the scale. In addition, the PES-14 and the state-based amotivation subscale

of the SIMS (Guay, Vallerand, & Blanchard, 2000) were negatively correlated (r (201) = -.35, p

< .001). Finally, the PES-14’s relationship to the state-based external regulation (e.g., extrinsic

motivation) subscale of the SIMS was negative (r(201) = -.20, p < .01). Since theory suggests

that play is positively related to intrinsic motivation but negatively to external regulation

(O’Connor & LaPoint, 1980), convergent validity was supported.

THE PLAY EXPERIENCE SCALE 12

Discriminant Validity. The discriminant validity of the PES-14 was assessed by

comparing the average variance extracted (AVE; Fornell & Larcker, 1981; Farrell, 2009) to the

squared correlation of the PES-14 and the IMS. AVE values were calculated using Construct

Validity Calculator 2.0 (Md-Basir et al., 2010). The 14-item PES had an AVE of 0.68; the IMS’s

AVE was 0.77 (see Table 5 for scale correlations). Both of these were higher than the squared

correlation between PES and IMS, r2 = 0.62. Thus, although the two scales were highly

correlated, the PES and intrinsic motivation scales showed discriminant validity.

Insert Table 5 about here

Construct Validity. Because another measure of play, as such, did not exist, construct

validity was assessed by comparing PES aggregated means in the two game conditions (designed

to involve play) to the search task condition (as ―play-free‖ as possible) via ANOVA. Mean

scores for the combined game conditions (M Games = 4.56, SD = 0.73) were significantly higher

than for the search task condition (M Search Task = 3.57, SD = 0.57, F(1,201) = 96.94, 2 = .32, p <

.001). Construct validity was also assessed by the correlation between playfulness as measured

by the Computer Playfulness Scale (CPS) and the PES-14 (r(201) = .20, p < .001).

Discussion

The purpose of Study 1 was to examine the psychometric properties of the initial,

theoretically derived PES. After pruning to 14 items, the scale showed high reliability and

correlated with existing measures of theoretically related constructs. Further, Study 1 provided

evidence of good discriminant, convergent, and predictive validity from the relationships of the

PES-14 with established scales of motivation and playfulness, and the study manipulation. As

THE PLAY EXPERIENCE SCALE 13

expected, the PES-14 was positively correlated with intrinsic motivation (IMS) and playfulness

(CPS), but negatively correlated with the amotivation and external regulation components of the

Situational Motivation Scale (SIMS). These results align with theory.

Study 2

Study 1 suggested that the 14-item version of the PES could serve as an adequate and

meaningful measure of play experience in video games. However, because the PES-14 was non-

symmetrical, the impact of the Freedom dimension (with only 2 items) might be underestimated

in overall PES-14 scores unless practitioners averaged within each scale first. Consequently, we

developed a symmetrical, 16-item version (see Table 6) of the scale by adding two items to the

freedom dimension. To avoid introducing new concepts to the scale, the items added were

similar to the two freedom items already present in the scale (i.e., focused on the player being

able to make the game do what they wanted to do). In Study 2, we tested whether adding the

items had resulted in any change to the scale’s reliability or validity. We also increased the

generalizability of the Study 2 findings by using a previously developed serious game, InnerCell

(Pavlas, Heyne, Bedwell, Lazzara, & Salas, 2010).

Method

Participants. A sample of 77 undergraduate volunteers (36 male, 41 female) enrolled at

the University of Central Florida participated in Study 2 for course credit. They ranged in age

from 18 to 27 years (M = 18.84 yrs, SD = 1.71).

Materials. InnerCell is a strategy game that requires players to learn about the immune

system while defending a patient's body from infection (Pavlas et al., 2010). Players direct the

body's immune cells to defeat pathogens, with each pathogen being vulnerable to a particular

THE PLAY EXPERIENCE SCALE 14

combination of cells and treatments. Each successive round of gameplay increases in difficulty

and complexity, providing new pathogens and options.

Measures. As Study 2 was not only a validation effort for the PES but also an

investigation of a theoretical model of learning and behavior in serious games, several additional

measures were employed. However, for the purposes of this second validation effort, we report

only on those measures meaningful to the validation. These were the 16-item revision of the PES

(PES-16), the CPS (Webster & Martocchio, 1992), and the IMS (McAuley, Duncan, & Tammen,

1989).

Design. This validation effort was part of a non-experimental study examining participant

behaviors and states across a variety of dimensions (e.g., play, motivation, emotion,

performance) within the InnerCell serious game. Given this, there was no manipulated

independent variable. Participants engaged in gameplay across three rounds, with a variety of

measures, including the PES, collected after each round.

Procedure. Participants completed a short set of demographic and pre-measures and

viewed a simple training video explaining the InnerCell game. They then played the game for

three full rounds. Between each round, participants responded to the PES. Finally, participants

completed the intrinsic motivation measure and learning assessments.

Insert Table 6 about here

Results

As previously noted, two items had been added to the PES-14 to make it symmetrical.

The revised 16-item PES-16 (see Table 6) was very highly correlated with the PES-14 (r = .99, p

THE PLAY EXPERIENCE SCALE 15

< .001). Further, mean scores of the PES-16 (M = 4.61, SD = .69) were significantly and

positively correlated with intrinsic motivation (IMS) (M = 4.07, SD = .76; r = .66, p < .001).

Based on item-total correlation, the two new items (items 1 and 4) were meaningful contributors

to the overall scale (item 1: r(74) = .74, p < .001; item 2: r(74) = .53, p < .001). Given the

correlation between the 14-item and the 16-item versions of the scale and the item-total

correlations, it was reasonable to suggest that the 16-item version of the scale did not reduce the

validity of the PES.

Cronbach’s alphas for each of the three administrations of the 16-item version of the

scale were calculated, ranging from 0.86 to 0.90 (Round 1 = .90; Round 2 = .89; Round 3

= .86). For comparison, alphas for the 14-item version of the scale were also calculated (Round 1

= .89; Round 2 = .88; Round 3 = .86). These scores also indicated good reliability.

The utility of the extrinsic motivation items of the PES was also examined. These items

were identical in both the PES-14 and the PES-16. In Study 1, the extrinsic motivation items

were not strongly related to the PES. Consequently, in Study 2, item-total correlations for the

scale were examined to ascertain whether the change in game context resulted in greater meaning

for these items. Items 5 through 8 showed more reasonable item-total correlations (ranging from

.38 to .60) in this outcome-oriented context, suggesting they be retained in the final version of

the scale.

Finally, the link between playfulness and play was examined. The correlation between

mean scores on the CPS and the PES-16 was the same as in Study 1 (r = .20), but not statistically

significant in Study 2 (r(72) = .20, p = .09), due to its smaller sample size (dfStudy1 = 201 vs.

dfStudy2 = 72).

THE PLAY EXPERIENCE SCALE 16

Discussion

Study 2 served to provide additional validation of the Play Experience Scale. Although

factor analysis was not possible due to sample size restrictions, evidence based on correlations

suggested that the two items added to the scale did not alter its meaning or psychometric

properties.

Study 2 also added a learning evaluation which provided participants with a degree of

external motivation and addressed concerns over Study 1’s lack of context for the external

motivation items. This also increased the generalizability of the results from Study 2 to other

game contexts.

The lack of a significant correlation between playfulness and play in Study 2 appeared to

present a problematic result at first. The actual correlations between play and playfulness,

however, were approximately the same in both Study 1 and Study 2 (r = .20); only the

probability values were different due to different sample sizes. Although this was encouraging,

correlations of .20 only explain 4% of the variance in play by playfulness. This rather low value

may be due to several reasons: First, the link between actual behavior and an individual’s traits

or attitudes can be weak (Ajzen & Fishbein, 1977) or complicated by external factors (Kraus,

1995). Second, the low correlations may suggest limited construct validity of the Computer

Playfulness Scale, a concern previously mentioned in the literature (e.g., Kruger, 1995). Further

research will be required to delve into the relationship between playfulness and play.

Overall Discussion

The research reported here was conducted to fill a particular need: until now, no scale

measuring the subjective experience of play existed. Though it was possible to measure

playfulness (i.e., propensity to play), no measure was available assessing an individual’s

THE PLAY EXPERIENCE SCALE 17

manifested play behavior. The Play Experience Scale (PES) was created based on a number of

definitions of play, with an emphasis placed on using concepts that were tied to established

psychological theory.

In two validation studies, the PES’s relationships to existing measures of intrinsic

motivation, external regulation, and amotivation emerged as expected. As suggested by theory,

play behavior was positively correlated with intrinsic motivation, negatively correlated with

external regulation, and negative correlated with amotivation. The scale was also sensitive to

variations in task characteristics in that the restricted, tedious, and decidedly ―non-play‖ word

search task in Study 1 was associated with lower PES scores.

Limitations and Future Research

Although the validation efforts reported in this article were fruitful, additional

investigations into the PES are warranted. First, even though the scale was validated in four

different video game contexts, it is still in need of validation in other settings. In particular, this

effort only examined the utility of the PES within the context of video games. However, play is

observed across the full range of human experience. Future research efforts should examine the

PES in other contexts (e.g., music, performance, sport, work).

Second, the scale was only examined with a relatively limited population (i.e., young

college students). The scale may have different psychometric properties when employed with

different populations. Future research should investigate how the PES functions for different

kinds of respondents.

Third, play as measured by the PES and the propensity for ―Playfulness‖ (here measured

by the CPS) showed only relatively small correlations between the two measures (r ≈ .20) in

both studies. Further research is required to determine whether this reflects the actual magnitude

THE PLAY EXPERIENCE SCALE 18

of the relationship between the two constructs, is due to external factors influencing trait-

behavior correlations, or is a sign of validity issues for the CPS (e.g., Kruger, 1995) or the PES-

16.

Finally, although the PES-16 consists of four subscales, we do not suggest that these

subscales be used diagnostically, as they were not validated for such use. Validation against non-

subjective measures of emotional experience may provide further evidence to determine the

overall appropriateness of the PES. For example, modeling emotional experience through

physiological indicators such as galvanic skin response, electromyography, and heart rate can

provide additional insight into how individuals respond to play experiences (Mandryk & Atkins,

2007). Tying these objective measures of experience to the newly developed subjective measure

of play would provide an excellent addition to the present validation efforts.

Conclusion

The Play Experience Scale (PES) fills a conspicuous gap in the game literature. Because

play is one of the defining characteristics of games (Ang, 2006; Baranauskas, Neto, & Borges,

1999), the toolset available to researchers investigating game-based learning is incomplete

without a method by which to investigate play. With the creation and validation of this new

scale, it is now possible for researchers to directly examine play experiences in games using a

simple post-task measure. Further, the general lack of play studies is indicative of a large

theoretical gap in the literature. The introduction of a scale that assesses play experience will

hopefully spark a surge of interest into investigating how play influences a wide range of

behaviors, states, and experiences.

THE PLAY EXPERIENCE SCALE 19

Key Points:

Though games are a subject of growing interest to the field, the study of games is

constrained by the lack of a measure of play as a subjective experience.

The 16-item version of the Play Experience Scale (PES-16) is a short, yet valid and

reliable measure of play in conventional video games.

The Computer Playfulness Scale (CPS) may require further examination, as extant

concerns over its validity were corroborated by this study.

Further validation and modification of the PES is required for use in other non-video

game contexts.

References

Anderson, M. (1998). The meaning of play as a human experience. In D. P. Fromberg, & D.

Bergen (Eds.) Play from birth to twelve and beyond: Contexts, (pp. 103–108). New York:

Harvard University Press.

Ang, C. S. (2006). Rules, gameplay, and narratives in video games. Simulation & Gaming, 37(3),

306–325. doi:10.1177/1046878105285604.

Azjen, I. & Fishbein, M. (1977). Attitude-behavior relations: A theoretical analysis and review of

empirical research. Psychological Bulletin, 84(5), 888-918.

Baranauskas, M. C. C., Neto, N. G. G., & Borges, M. A. F. (1999). Learning at work through a

multi-user synchronous simulation game. In Proceedings of the PEG 99 Conference, (pp.

137–144). Exeter, UK.

Benabou, R., & Tirole, J. (2003). Intrinsic and extrinsic motivation. Review of Economic Studies,

70(3), 489–520. doi: 10.1111/1467-937X.00253.

THE PLAY EXPERIENCE SCALE 20

Calleja, G. (2007). Revising immersion: A conceptual model for the analysis of digital game

involvement. In Proceedings of the DiGRA 2007 Conference.

Chertoff, D. B., Jerome, C., Martin, G. A., & Knerr, B. W. (2008). GamePAB: A game-based

performance assessment battery application. In Proceedings of the Human Factors and

Ergonomics Society 52nd

Annual Meeting (pp. 1570-1573).

Cotton, J., Mayes, D., Jentsch, F., & Sims, V. (2001). The relationship between video game

characteristics and player ability. In Proceedings of the Human Factors and Ergonomic

Society 47th Annual Meeting (pp. 945-947).

Cowley, B., Charles, D., Black, M., & Hickey, R. (2008). Toward an understanding of flow in

video games. Computers in Entertainment, 6(2), 20:1–27. doi:

10.1145/1371216.1371223.

Csikszentmihalyi, M. (1990). Flow: The Psychology of Optimal Experience. New York: Harper

Perennial.

Esposito, N. (2005). A short and simple definition of what a videogame is. In Proceedings of the

DiGRA 2005 Conference: Changing Views – Worlds at Play.

Farrell, A. M. (2010). Insufficient discriminant validity: A comment on Bove, Pervan, Beatty,

and Shiu (2009). Journal of Business Research, 63, 324-327. doi:

10.1016/j.jbusres.2009.05.003

Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobserv- able

variables and measurement error. Journal of Market Research, 18(1), 39–50.

doi:10.2307/3151312.

Garris, R., Ahlers, R., & Driskell, J. E. (2002). Games, motivation, and learning: A research and

practice model. Simulation & Gaming, 33(4), 441–467. doi:10.1177/1046878102238607.

THE PLAY EXPERIENCE SCALE 21

Gitlin-Weiner, K. (1998). Clinical perspectives on play. In D. P. Fromberg, & D. Bergen (Eds.)

Play from birth to twelve and beyond: Contexts, perspectives, and meanings, (pp. 77–93).

Routledge.

Gopher, D., Weil, M., Bareket, T. (1994). Transfer of skill from a computer game trainer to

flight. Human Factors, 36(3), 387-405.

Gordon, G. (2008). What is play? In search of a universal definition. Play and Culture Studies, 8.

Guay, F., Vallerand, R. J., & Blanchard, C. (2000). On the assessment of situational intrinsic and

extrinsic motivation: The situational motivation scale (SIMS). Motivation and Emotion,

24(3), 175–213.

Knox, S. (1997). Development and current use of the Knox Preschool Play Scale. In L. D.

Parham & L. S. Fazio (Eds.), Play in occupational therapy for children (pp. 35-51). St.

Louis, MO: Mosby.

Kraus, S. J. (1995). Attitudes and the prediction of behavior: A meta-analysis of the empirical

literature. Personality and Social Psychology Bulletin, 21(1), 58-75.

doi:10.1177/0146167295211007

Kruger, A. (1995). The adult playfulness scale: A review. Psychology: A Journal of Human

Behavior, 32, 36–38.

Lampton, D., Bliss, J., Orvis, K., Kring, J., & Martin, G. A. (2009). A distributed game-based

simulation training research testbed. In Proceedings of the Human Factors and

Ergonomic Society 53rd Annual Meeting (pp. 1989-1993).

Malone, T. (1981). Toward a theory of intrinsically motivating instruction. Cognitive Science: A

Multidisciplinary Journal, 54, 333–369. doi: 10.1016/S0364-0213(81)80017-1.

THE PLAY EXPERIENCE SCALE 22

Mandryk, R. L., & Atkins, M. S. (2007). A fuzzy physiological approach for continuously

modeling emotion during interaction with play technologies. International Journal of

Human-Computer Studies, 65, 329–347. doi:10.1016/j.ijhcs.2006.11.011.

McAuley, E., Duncan, T., & Tammen, V. (1989). Causal attributions and affective reactions to

disconfirming outcomes in motor performance. Journal of Sport and Exercise

Psychology, 11(2), 187–200.

McMahan, A. (2003). Immersion, engagement and presence: A new method for analyzing 3-D

video games. In M. J. P. Wolf, & B. Perron (Eds.) The video game theory reader, (pp.

67–87). New York: Routledge.

Md-Basir, H., Mahzan, S., Mokhtar, N., Abdl-Aziz, N., Rahman, M. A., & Daud, D. (2010). The

online construct validity calculator version 2.0.

http://www.hishammb.net/cvc2/index.asp.

Millar, S. (1968). The psychology of play. Oxford, England: Penguin Books.

O’Connor, J., & LaPoint, J. (1980). The phenomenon of play: A multi-dimensional definition.

Recreation Sports Journal, 5(1), 6–11.

Pavlas, D., Heyne, K., Bedwell, W., Lazzara, E., & Salas, E. (2010). Game-based learning: The

impact of flow state and videogame self-efficacy. In Proceedings of the Human Factors

and Ergonomic Society 54th Annual Meeting (pp. 2398-2402).

Ryan, R. M., & Deci, E. L. (2000). Intrinsic and extrinsic motivations: Classic definitions and

new directions. Contemporary Educational Psychology, 25(1), 54-67.

doi:10.1006/ceps.1999.1020.

Salen, K. & Zimmerman, E. (2004). Rules of play: Game design fundamentals. Cambridge, MA:

MIT Press.

THE PLAY EXPERIENCE SCALE 23

Van Leeuwen, L. & Westwood, D. (2008). Adult play, psychology, and design. Digital

Creativity, 19(3), 153-161.

Webster, J., & Martocchio, J. (1992). Microcomputer playfulness: Development of a measure

with workplace implications. MIS Quarterly, 16(2), 201–226. doi:10.2307/249576.

Weissinger, E., & Bandalos, D. (1995). Development, reliability and validity of a scale to

measure intrinsic motivation in leisure. Journal of Leisure Research, 27(4).

Yee, N. (2005). The labor of fun: How video games blur the boundaries of work and play.

Games and Culture, 1(1), 68-71.

THE PLAY EXPERIENCE SCALE 24

Appendix A

Table 1: Initial play experience scale

# Item Text Dimension/Coding 1 I felt that I was free to use whatever strategy I wanted to while I was

using the game Freedom

2 I was able to make the game do what I wanted it to Freedom 3 The game gave me the freedom to act how I wanted to Freedom 4 The game made it difficult to do what I wanted to do Freedom (R) 5 I was not worried about someone judging how I performed in the

game No extrinsic

6 Regardless of how I performed in the game, I knew there wouldn’t be

a real-world consequence No extrinsic

7 My performance in the game was not going to matter outside of the

game No extrinsic

8 I felt like I had to do well, or the experimenter would judge me No extrinsic (R) 9 When I was using the game, it felt like I was playing rather than

working Play-direct

10 I would characterize my experience with the game as ―playing‖ Play-direct 11 I was playing a game rather than working Play-direct 12 Using the game felt like work Play-direct (R) 13 When I was using the game, I didn’t worry about anything in the real

world Focus

14 I was able to concentrate on the game without thinking about other

things Focus

15 When I was using the game, I was focused on the task at hand Focus 16 I had a hard time concentrating on the game Focus (R) 17 I wanted to do well in the game, "just because" Autotelic 18 When I was using the game, I wanted to do as well as possible Autotelic 19 I tried to succeed in the game because I felt like it Autotelic 20 During the game, my performance didn’t matter to me Autotelic (R)

THE PLAY EXPERIENCE SCALE 25

Table 2: Study 1 means, standard deviations, and correlations

Mean SD Age Gender Gamer Playful PES R1 PES R2 Motiv R1 Motiv R2 Ext R1 Ext R2 Amot R1

Age 21.01 4.95

Gender N/A N/A -0.24*** Gamer N/A N/A 0.17* -0.48***

Playful 5.59 0.78 0.05 0.05 0.07

PES R1 4.36 0.78 -0.09 -0.07 0.12 0.09 PES R2 4.22 0.83 -0.04 0.01 0.11 0.20** 0.73***

Motiv R1 3.82 0.76 0.00 0.00 0.12 0.05 0.67*** 0.61***

Motiv R2 3.76 0.84 0.06 -0.03 0.17* 0.10 0.64*** 0.79*** 0.80*** Ext R1 3.55 1.18 -0.23** 0.09 -0.16* -0.07 -0.23*** -0.33*** -0.27*** -0.37***

Ext R2 3.51 1.15 -0.26*** 0.12 -0.10 -0.09 -0.11 -0.20** -0.19** -0.27*** 0.76***

Amot R1 3.29 1.03 -0.03 0.00 -0.06 -0.06 -0.31*** -0.41*** -0.43*** -0.46*** 0.60*** 0.48*** Amot R2 3.36 1.08 -0.07 -0.03 -0.10 -0.04 -0.25*** -0.35*** -0.30*** -0.40*** 0.52*** 0.58*** 0.72***

* p < .05, ** p < .01, *** p < .001

Gender shows correlations for female participants (0 = male, 1 = female).

R1 = Round 1, R2 = Round 2, Playful = Computer Playfulness Scale, Gamer = Self-reported gamer status from demographics,

PES = Play Experience Scale,

Motiv = Intrinsic Motivation, Ext = SIMS External Regulation, Amot = SIMS Amotivation

THE PLAY EXPERIENCE SCALE 26

Table 3: PES individual item – game condition regression

Item B SE t Sig.

Item 1 0.01 0.03 0.02 0.29 0.77

Item 2 0.15 0.02 0.42 6.64 0.00***

Item 3 0.10 0.03 0.27 3.99 0.00***

Item 5 0.06 0.02 0.17 2.47 0.01*

Item 6 0.03 0.03 0.07 1.00 0.32

Item 7 0.02 0.03 0.04 0.52 0.61

Item 8 0.05 0.02 0.13 1.84 0.07

Item 9 0.17 0.02 0.58 0.00 0.00***

Item 10 0.18 0.02 0.59 0.00 0.00***

Item 11 0.18 0.02 0.58 0.00 0.00***

Item 12 0.17 0.02 0.57 0.00 0.00***

Item 15 0.13 0.02 0.36 0.00 0.00***

Item 17 0.11 0.03 0.29 0.00 0.00***

Item 18 0.11 0.03 0.29 0.00 0.00***

Item 19 0.09 0.02 0.25 0.00 0.00***

* p < .05, *** p < .001

Note: All regressions run individually to determine univariate impact.

For all cases, df = 1,201

THE PLAY EXPERIENCE SCALE 27

Table 4: Principal component analysis: 14-item PES

Item C1 C2 C3 C4 Communality

Item 2 -.07 .14 -.03 .87 0.81

Item 3 -.10 -.01 .01 .94 0.80

Item 5 .08 -.03 .76 .16 0.67

Item 6 .01 .07 .67 -.12 0.43

Item 7 .36 -.36 .57 .15 0.55

Item 8 -.30 .27 .79 -.12 0.67

Item 9 .20 .79 -.02 -.01 0.81

Item 10 .17 .76 -.01 .10 0.82

Item 11 .18 .75 -.04 .11 0.82

Item 12 -.12 .90 .11 -.03 0.73

Item 15 .72 .03 -.05 .20 0.71

Item 17 .83 .12 .08 -.12 0.74

Item 18 .91 .04 -.05 -.06 0.81

Item 19 .90 .06 .01 -.14 0.76 Factor loadings >.40 shaded. Oblique rotation conducted using the Promax function. The four components

accounted for 72.41% of the total variance in the 14 items.

THE PLAY EXPERIENCE SCALE 28

Table 5: Scale correlations, round 2

PES-14 PES-4 Playfulness IMS SIMS External

PES-14

PES-4 0.76***

Playfulness 0.20** 0.12

IMS 0.79*** 0.63*** 0.10

SIMS External -0.20** -0.12 -0.09 -0.27***

SIMS Amotivation -0.35*** -0.33*** -0.04 -0.40*** 0.58*** ** p <.01, *** p <.001

N = 203, two-tailed significance values reported.

PES-14 = Play Experience Scale 14 item, PES-4 = Play Experience Scale 4 item, IMS = Intrinsic Motivation Scale,

SIMS = Situational Motivation Scale

THE PLAY EXPERIENCE SCALE 29

Table 6: Play scale 16-item version

# Item Text Dimension/Coding

1 If I wanted to do something in the game, I was able to do it Freedom 2 I was able to make the game do what I wanted it to Freedom 3 The game gave me the freedom to act how I wanted to Freedom 4 The game made it difficult to perform the actions that I wanted

to Freedom (R)

5 I was not worried about someone judging how I performed in

the game No extrinsic

6 Regardless of how I performed in the game, I knew there

wouldn’t be a real-world consequence No extrinsic

7 My performance in the game was not going to matter outside

of the game No extrinsic

8 I felt like I had to do well, or the experimenter would judge me No extrinsic (R) 9 When I was using the game, it felt like I was playing rather

than working Play-direct

10 I would characterize my experience with the game as playing Play-direct 11 I was playing a game rather than working Play-direct 12 Using the game felt like work Play-direct (R) 13 When I was using the game, I was focused on the task at hand Autotelic-Focus 14 I wanted to do well in the game, "just because" Autotelic-Focus 15 When I was using the game, I wanted to do as well as possible Autotelic-Focus 16 I tried to succeed in the game because I felt like it Autotelic-Focus Note: Item numbers do not correspond to original item numbers, due to removal of original items.

THE PLAY EXPERIENCE SCALE 30

Biographies

Davin Pavlas is a recent graduate of the Applied Experimental & Human Factors Psychology

doctoral program at the University of Central Florida. Davin obtained his Ph.D. in 2010 through

a dissertation that established a model of flow and play in games for learning. As a researcher at

the Institute for Simulation and Training, Davin works on a variety of projects that examine

teams using simulations. To date, his research has examined the relationship between aesthetics

and usability in user interfaces, the nature of flow state in games, the attributes of serious games

that contribute to learning, and the role of play experience in learning.

Florian Jentsch is director of the Team Performance Laboratory, and an Associate Professor of

Psychology at the University of Central Florida, with a joint appointment in the Department of

Psychology and at the Institute for Simulation & Training. He received his Ph.D. in human

factors psychology at the University of Central Florida in 1997. He also holds master's degrees in

aeronautical engineering from the Technical University of Berlin and in aeronautical science

from Embry-Riddle Aeronautical University. His dissertation for junior flight crewmembers won

the APA 1998 George E. Briggs Award for the best dissertation in applied/experimental

psychology. Florian also won the APA Earl Allusi award in 2002 for early career achievement.

He has co-authored over 150 publications, presentation, and technical reports, with research

ranging from pilot training and aviation research, to transportation security x-ray baggage

screening and human robot teaming.

Eduardo Salas is Trustee Chair and Professor of Psychology at the University of Central Florida

where he also holds an appointment as Program Director for the Human Systems Integration

Research Department at the Institute for Simulation and Training. Previously, he was the

Director of UCF's Applied Experimental & Human Factors Ph.D. Program. Before joining IST,

he was a senior research psychologist and Head of the Training Technology Development

Branch of NAWC-TSD for 15 years. Dr. Salas has co-authored over 300 journal articles & book

chapters and has co-edited 19 books. His expertise includes assisting organizations in how to

foster teamwork, design and implement team training strategies, facilitate training effectiveness,

manage decision making under stress, and develop performance measurement tools. Dr. Salas

received his Ph.D. in Industrial/Organizational Psychology at Old Dominion University in 1984.

Stephen M. Fiore is on the faculty of the University of Central Florida’s Cognitive Sciences

Program in the Department of Philosophy and is the director of the Cognitive Sciences

Laboratory at the University of Central Florida’s Institute for Simulation and Training. He

earned his Ph.D. degree (2000) in cognitive psychology from the University of Pittsburgh,

Learning Research and Development Center.

Valerie Sims is the director of the Applied Cognition and Technology (ACAT) Laboratory and

an Associate Professor of Psychology at the University of Central Florida. Her research centers

primarily around individual differences in spatial ability and human-robot interaction. She

received her Ph.D. in applied cognitive psychology from the University of California, Santa

Barbara in 1996.

THE PLAY EXPERIENCE SCALE 31

Acknowledgments

The InnerCell game used in this study was created as a result of earlier research that was

partially supported by subcontract No. 0450-1382 from Aptima under prime contract from the

Naval Air Warfare Center, contract No. N68335-07-C-0444. All opinions expressed in this paper

are those of the authors and do not necessarily reflect the official position of the University of

Central Florida, Aptima, or the Department of Defense.