Society essay relate to video game

Media Psychology, 19:101–125, 2016 Copyright © Taylor & Francis ISSN: 1521-3269 print/1532-785X online DOI: 10.1080/15213269.2014.902318

Development and Validation of an Instrument for Measuring Individual Motives

for Playing Digital Games

FREDERIK DE GROVE Department of Communication Sciences, Ghent University, Research Group for Media and

ICT (iMinds-MICT), Ghent, Belgium

VEROLIEN CAUBERGHE Department of Communication Sciences, Ghent University, Center for Persuasive

Communication (CEPEC), Ghent, Belgium

JAN VAN LOOY Department of Communication Sciences, Ghent University, Research Group for Media and

ICT (iMinds-MICT), Ghent, Belgium

Individual motives for playing digital games have been studied from a variety of theoretical perspectives using different measure- ment instruments. However, an instrument that roots the social, rule-based narrative essence of digital games in a theory on hu- man behavior acknowledging that not all behavior is consciously motivated has hitherto been lacking. A framework based on so- cial cognitive theory that integrates these dimensions is proposed. After comparing the advantages of this framework to existing ap- proaches, the development of a measurement instrument is dis- cussed. This development concerns the generation and evaluation of an item pool and also testing the instrument for reliability and validity on different samples and different populations. Re- sults suggest psychometric as well as theoretical soundness of the instrument.

Playing digital games has become a widespread phenomenon in everyday leisure (Entertainment Software Association, 2013; Interactive Software Fed- eration of Europe, 2012). Understanding why millions of people engage in this activity is important for several reasons. In the first place, it allows

Address correspondence to Frederik De Grove, Department of Communication Sciences, Ghent University, Research Group for Media and ICT (iMinds-MICT), Korte Meer 7-9-11, Ghent 9000, Belgium. E-mail: [email protected]

101

102 F. De Grove et al.

for explaining and understanding the growing popularity of the medium in question while it additionally fosters further inquiry into motivational processes regarding mediated human action. Furthermore, understanding motives for play provides a necessary starting point for related research questions such as those concerning positive or negative effects of playing digital games (Ferguson & Olson, 2013). Academic research on the topic has been approached from different perspectives. Drawing on self-determination theory (Deci & Ryan, 1985), scholars have regarded motives for playing digital games as an intrinsically enjoyable experience explained by the need for competence, relatedness, and autonomy (Przybylski, Rigby, & Ryan, 2010; Tamborini, Bowman, Eden, Grizzard, & Organ, 2010). From a uses and gratifications perspective, digital play has been explained by constructs such as arousal, challenge, competition, diversion, fantasy, and social interaction ( Jansz & Tanis, 2007; Sherry, Lucas, Greenberg, & Lachlan, 2006). In a similar vein, Lee and LaRose (2007) used social cognitive theory (SCT) and flow to explain play behavior in terms of concentration, merging of action and awareness, enjoyment, optimal balance, habit, self-reactive outcomes, and self-regulation. On a theoretical level, these theories have proven to present distinct yet fruitful starting points to approach motives for playing digital games. On a conceptual level, they all result in motivational dimensions that are more often similar than not. It has been remarked, however, that these approaches do not fully account for the specificity of the medium under scrutiny. In order to overcome this obstacle, De Grove, Cauberghe, and Van Looy (2014) have conceptualized games as social, rule-based narrative systems and have linked this to social cognitive theory, a broad theory on human behavior.

A SOCIAL COGNITIVE THEORY OF DIGITAL GAMES

According to SCT, people are motivated to perform certain actions by cogni- tively processing consequences of behavior (Bandura, 1986). Indeed, action is partly determined by anticipating consequences of behavior through the capability of forethought. These consequences can be based on experiential or vicarious learning. In the latter case, people form beliefs about conse- quences observed from others. Outcomes can be classified based on their main source of production. First, self-produced outcomes result from evaluat- ing the outcome of actions according to personal or environmental standards, which results in self-produced tangible or affective self-reactions. A second outcomes type stems from factors external to the person, such as material, sensory, token, and social outcomes. In practice, behavior emerges from combining these different types of outcome expectations. Moreover, both types are interconnected. Indeed, external outcomes also depend on self- regulatory influences when it comes to their impact (Bandura, 1986). Guiding

Development and Validation of the DGMS 103

the functioning of these outcome expectations is the belief of self-efficacy, which concerns the belief in one’s capabilities to perform certain behavior. The belief in one’s capabilities will influence which goals to set and how subsequent behavior is observed and evaluated. While research on media use has shown that outcome expectations serve as predictors of behavior, it has also been pointed out that not all behavior is consciously motivated (LaRose & Eastin, 2004). Indeed, repeated behavior in stable contexts leads to habit formation in order to reduce the cognitive load related to decision making (LaRose, 2010). Hence, over time, outcome expectations are transformed into habits, which guide behavior that was previously consciously motivated.

From a digital games perspective, three relevant types of expected outcomes have been defined: game-internal, game-external, and normative outcomes (De Grove et al., 2014). Game-internal outcomes are outcomes that stem from digital games being conceptualized as social rule-based narrative systems and are formed based on direct consequences when playing digital games, be they firsthand or vicarious. They are internal in that the main source of the outcomes lies in the characteristics of the activity. In fact, these outcomes can be mainly considered as those that make gaming intrinsically enjoyable. It should be noted that these game-internal outcome expectations are initially produced by the game, yet they also tend to become self- produced due to their susceptibility to the self-regulation process (De Grove et al., 2014). Similarly, game-external outcomes can also be considered as direct outcomes. The underlying goal is not enjoyment however. Instead, play serves as a mediator between individual and context. In this sense, playing digital games becomes a means and not an end in itself. Normative outcomes, finally, are self-produced outcomes based on moral standards. In contrast to game-internal and game-external outcomes, normative outcomes do not directly stem from the behavior in question but are related to the status of digital games as cultural artifacts in contemporary society. In fact, normative outcomes refer to the idea that the activity of playing digital games is not a neutral one. As a consequence, the morality of the behavior can play a role in the judgmental process of self-regulation, which can, in turn, lead to affective self-reactive outcomes. Indeed, even if one expects the outcome of an activity to be pleasurable, this does not ensure the behavior will be executed. When the behavior in question is negatively evaluated in a normative way, it might prevent performing the relevant activity.

Using SCT as a framework for understanding individual motives for play is advantageous for at least two reasons (De Grove et al., 2014). First, it has proven its value in providing a flexible framework in which to place and de- fine different determinants of digital play (Lee & LaRose, 2007). Second, out- come expectations can be based on first-hand experiences and on vicarious learning. As a consequence, people can hold expectations about outcomes of actions they did not perform themselves. Hence, outcome expectations allow for understanding why people do not play certain game genres or

104 F. De Grove et al.

why people do not play digital games altogether. This reasoning is also useful in accounting for the variation in the rule-based, social, and narrative dimensions. Indeed, if one never plays a digital game with a story, it does not mean that expectations concerning the narrative dimension are absent. These advantages become apparent when comparing the SCT framework with that of self-determination theory (SDT). Indeed, the needs proposed by SDT do not fully account for gaming as a social rule-based narrative practice. This is especially true for the content dimension which is absent in SDT. Furthermore, since SDT is limited to three universal needs, there seems to be little flexibility in accounting for other relevant or culture- specific motives such as normative evaluations of the behavior in question. In addition, research on digital games and SDT typically restricts itself to intrinsic motivation (see Przybylski et al., 2010; Tamborini et al., 2010). While it is true that digital games are generally played for pleasure, this is seldom exclusively the case (Kallio, Mäyrä, & Kaipainen, 2011). What is more, the conceptualization of SDT does not account for habitual behavior which has proven to significantly mediate outcome expectations and predict game attendance (Lee & LaRose, 2007). In other words, SCT allows for taking into account the interplay between consciously and less consciously motivated behavior. Finally, SDT is concerned with understanding past behavior. In contrast to SCT, it does not allow for understanding why people would start playing games in the first place, nor why people would refrain from playing digital games or certain game genres.

Drawing on SCT, literature on motivations for gaming, and in-depth interviews, De Grove and colleagues (2014) have proposed nine expected outcomes for play: performance, agency, believability, involvement, socia- bility, status, moral self-reaction, escapism, and pastime. In addition to these outcomes, the concept of habit has been proposed to account for less con- scious behavior. Pastime and escapism can be considered as game-external outcomes, moral self-reaction as a normative outcome, and all other out- comes as game-internal outcomes. A short discussion of those constructs follows. For a more elaborate overview, we refer to De Grove et al. (2014), and for a schematic overview, we refer to Table 1.

Performance refers to the player’s expectation to perform well when playing digital games. Performance happens when a goal is reached. Goals, however, can have different origins. First, they can be set by the game, such as beating a boss or completing a level. Second, players can set their own goals. Third, some goals are set by other people involved in the game, for instance, when trying to beat each other’s scores in a racing game. These goals should not be seen as necessarily independent from each other. Agency refers to the expectations of the player regarding his or her ability to play the game according to his or her own preferences. It concerns having the feeling of playing a game instead of being played by the game. This can refer to narrative as well as ludological aspects of

Development and Validation of the DGMS 105

TABLE 1 Overview of Motives for Digital Play

Construct Short description

Performance The expectation to perform well when playing digital games. Agency The expectation to play the game according to the gamers’ own

preferences. Status The expectation of being respected by other players. Sociability The expectation to enact non-competitive social behavior when

playing. Believability The expectation about coherence and believability of the game

environment. Involvement The expectation about involvement with aspects of the game world. Escapism The expectation to leave the daily routine behind and experience

things that would not be possible outside of the game. Moral self-reaction Expectations resulting from comparing playing digital games with own,

social, or moral norms. Pastime The expectation to kill time when playing. Habit Refers to media use that is not active. It concerns starting to play

games without really thinking about it.

playing games. Believability concerns expectations about coherence and realism of the game environment in terms of audiovisual aspects, the story, the setting, the actions, the topic, and the characters in the game. It is not so much about actual resemblance to the real world but more about being in a believable, coherent game environment. It has to be noted, however, that although focused on the narrative aspect, believability also relates to actions in the game. As a consequence, the believability component incorporates both narrative and rule-based dimensions of play. Involvement refers to how involved the player expects to be with different aspects of the game world. Similar to believability, it concerns audiovisual aspects, the story, the setting, or the characters. Unlike believability, however, action is not part of involvement. Sociability concerns expectations of the player to enact noncompetitive social behavior when playing digital games. It is about being together, playing together, making new friends, as well as sustaining existing friendships. Status concerns expectations of being respected by other players. In contrast to sociability, status implies inequality in the relation with other players. This inequality has its origin in the recognition of the player’s performance by others. Escapism refers to the player’s expectation to leave the daily routine behind and experience things that would not be possible outside of the game. It takes into account what players escape from and where they go. Pastime concerns the expectation to kill time when playing. It is an expectation that is not so much about digital games or other players as about filling free time. Playing games is then considered as a casual activity rather than a purposeful one (Kallio et al., 2011). Moral self-reaction refers to self- produced normative evaluations regarding the activity of playing digital

106 F. De Grove et al.

games. Habit, finally, takes into account that repeated behavior becomes more automatic and less self-directed over time in order to make functioning more efficient and to lessen the cognitive load of active decision making (LaRose, 2010).

This theoretical and conceptual framework is the foundation on which the Digital Games Motivation Scale (DGMS) is operationalized. When con- structing a measurement instrument, an important step is to assess its criterion- related validity. If the measurement instrument behaves as could be expected based on its theoretical assumptions, it is an indication that it is measuring what it was intended to measure. In this case, the behavior of interest is playing digital games. Such behavior can be conceptualized in different ways. A first possibility is to look at the frequency of digital play. Several studies have operationalized this as the time someone plays during a typical week, measured in hours and minutes (LaRose & Eastin, 2004; Sherry et al., 2006). Hence, it is expected that differences in frequency will be related to differences in motives for digital play. Research has shown, however, that measuring behavior in hours and minutes raises problems regarding reliability (Blake & Klimmt, 2012). To avoid such problems, we argue that using categorical instead of ratio variables is a solution. As such, we expect a positive relation between the motives for play and the frequency of play. Similar to frequency, duration of an average gaming session is a measure that has been used as a dependent variable in relation with motives for play (Hou, 2011). Therefore, we expect differences in the average play session duration to be related positively with all motives for play except for pastime, where we expect a negative relation. The rationale for expecting this negative relation is that playing just to pass time suggests less investment in the gaming situation.

Behavior related to playing digital games, however, is more than fre- quency or duration alone. People differ on account of the games and genres that they play. Furthermore, most people play a variety of digital games and multiple games interchangeably (Williams, Yee, & Caplan, 2008). To conceptualize this kind of behavior, we draw on the field of media studies in which the concept of media repertoires is used to refer to the exposure to a variety of different media (Hasebrink & Popp, 2006). Similar to the idea of media repertoires, we advance the concept of game repertoires. In this case, it refers to the exposure of a player to different game genres. Previous studies on gaming motives have either focused on single games (Ryan, Rigby, & Przybylski, 2006), single game genres ( Jansz & Tanis, 2007), or digital games in general (Lee & LaRose, 2007) when assessing motives for play. By focusing on single games or genres, it is not taken into account that most people play more than only one game or genre. Looking at digital games in general, however, makes an abstraction of the differences in content a player is exposed to. Hence, the idea of game repertoires allows for taking into account the diversity of exposures between people.