STEP 2: SPECIFY MORE PRECISE RESEARCH QUESTIONS
As a research question, “Are video games good for science learning?” is vague and underspecified. Several authors have also noted this flaw in the existing research: such a broad research question is unanswerable (Gee, 2011). To illustrate a more nuanced version of a research question, consider the following: Under what conditions, and for whom, can video games facilitate or improve science achievement relative to achievement attained in a comparison group or groups?
Now we can unpack the constituent elements of this question.
Under what conditions: A variety of educationally relevant conditions could be enumerated including, for example, whether games are used for instruction or assessment, or whether the goal is to teach science concepts or science process skills. It may be that some games are better suited for content within particular science domains, or for individual or collaborative learning. We need to determine which types of conditions afford learning via a gaming context. For whom: Participant groups must be specified to avoid making assumptions about “one size fits all.” Is an intervention effective for at-risk students, advanced placement students, or for particular age groups (i.e., preschool, elementary, middle, or high school)? It is also important to specify participant characteristics to answer questions about aptitude-treatment interactions. Video games: “Video games” (and simulations) are very broad categories (Tobias & Fletcher, 2011). Games can and should be defined along dimensions of quality, complexity, genre, time scales, extent of collaboration, competitiveness, and so on. It is important to note whether a game was designed for entertainment but has instructional potential or whether it was designed as an instructional game.
Science achievement: As will be discussed in more detail below, K-12 science education is quite broad, and includes concepts, processes, and crosscutting ideas, all of which need to be taught and assessed in developmentally appropriate ways. Comparison group(s): Prensky (2011), for example, has noted that it is time to go beyond research implementing the prototypical “treatment vs. standard practice” research design.
Previous reviews that support the common conclusion that research to date is “inconclusive” may, in part, be due to overly broad research questions. Dede (2011) discusses the need for more precision in research questions, and provides samples for research on educational games in general. For the sake of illustration, a sample of his questions (p. 240) adapted for science content might include:
- • To what extent can educational games replicate authentic science inquiry activities that learners can master?
- • Are there particular individual difference characteristics of learners (e.g., the tendency to approach inquiry with either an “experimenter” or a “theorist” mindset/strategy; Dunbar & Klahr, 1989; Zimmerman, Raghavan, & Sartoris, 2003) that predict which students will (or will not) find educational games and simulations motivating and effective?
- • In hands-on inquiry tasks, students show a different pattern of performance when they are given the goals of a “scientist” or “engineer” (Schauble, Klopfer, & Raghavan, 1991). Does this same pattern generalize to role-playing in video games?