Total Pageviews

Sunday, October 2, 2016

Scientific Validity behind Success in the Video Game Industry



     An analysis of successful video games and studios reveals that causal research, otherwise known as explanatory research, plays a very significant, scientific role. This paper will begin by explaining causal research and its two largest components: validity and reliability. Next, there are three different types of validity that will be analyzed. Construct validity, external validity, and internal validity are all methods for inferring data or compiling statistics. These tools have several threats in the form of counter-arguments they have to overcome in order to maintain validity. Once there is a solid understanding of causal relationships, that information can by synthesized and applied to success in the video game industry. A clear and present correlation will be shown between these causal research methods and the techniques employed by thriving, industry leaders.  

Causality

     The process of examining cause-and-effect is known as causal research (Trochim, 2006). Everything around us is a result of a never-ending cycle of cause-and-effect. This even includes the business world and industries such as the world of video game production. There are two main methods for deciphering the causal relationships between events: experimentation and research (Trochim, 2006). Empirical research can be conducted on gathered data to get an idea of causative changes and their impact on the industry as a whole. Opposite of statistical research, experiments can allow for specific controls to manipulate the mechanics of the experience. Regardless of which method or combination is being used, a strong understanding of validity and reliability is vital when it comes to inferring data and drawing conclusions.

Reliability

     Reliability is a fairly common word and, in the world of causal events and research, it means essentially the same thing. Being able to build or find statistical patterns adds to the reliability of a conclusion or inference (Trochim, 2006).  Since reliability is really part of a ratio, there is another component known as variance (Trochim, 2006). Variance is a measurement that shows us how much of an offset there can be from the reliable baseline. These two factors come together to complete an estimation of reliability, from 100% to 0% (Trochim, 2006).

Validity

     After reliability has been established, we can move forward to validation. While measurements or samples cannot be validated, the conclusions we infer from the data can have validity as an attribute (Trochim, 2006). The methodology of research needs to remain under constant vigilance in order to continuously guarantee both reliable and validated suppositions. A measure or consensus that is both reliable and validated can then be referred to as a dependable evaluation. Theory and observation are the two territories involved in all forms of causal research (Trochim, 2006). Because this is a cause-and-effect study, there is a cause construct and an effect construct located in the theory hemisphere that are simply theorized contemplations (Trochim, 2006). In the observational hemisphere, there is the program (what you do) and the observations (what you see) that are linked by a program-outcome relationship (Trochim, 2006). There are four different modes of validity that function when studying these causal interactions and they all have different, distinct areas of the research realm from which they pull.

Internal Validity

     Once a relationship has been established between two variables in the research, one must then examine the affiliation to see if it is causal in nature. Once a cause-effect relationship is established, a conclusion can be drawn from that relationship that is then labeled as an internal validity (Trochim, 2006). Internal validity is only relevant to the exact study being conducted. Because everything is contained with an internal validity, there is a chance for threats to arise easily. Various groups or people in a position of leadership that are involved in the study have the capacity to completely invalidate the findings because they have forced their own will upon the results (Trochim, 2006). Isolating the social threats from one another is the best course of action to avoid any potential threats to validity of your inferences.

Construct Validity  

     Construct validity takes things a bit farther than an internal validation offers. A construct validation can only occur when two objectives are met, in addition to the objective of an internal validity (Trochim, 2006). If the program section of the study was implemented correctly and the outcomes were measured as intended; then we can claim a construct validity (Trochim, 2006). Extrapolating upon this, by labeling your conclusion as a construct validity, the argument is then being made that there was a knowledge of how the all parts of the study were constructed. The threats that arise from a construct validation often come from critics (Trochim, 2006). Being unable to adequately explain the constructs of the study, will result in a lack of widespread validation. Common issues arise when there is an overlap between constructs and their responsibilities or the observable size of a study is simply too small. These threats can be avoided by following basic protocols such as thinking through your concepts to entirety, obtaining subject matter experts to critique the conclusions you have drawn, or by utilizing methods that will help enhance your concepts (Trochim, 2006).

External Validity

     Another form of validity is referred to as an external authentication because they are only useful for generalizations. One method of achieving external validity would be through the sampling model; where one takes a sampling of the population and then infers the results and projects them onto the population at large, arguing that a wide enough sample was collected. The proximal similarity model on the other hand, is a much more accurate system because it only allows you to apply generalizations to groups similar to those you sampled. However, you still have no certain guarantee of an outcome from a study fortified with external validity (Trochim, 2006). The biggest threat opposing external validation is the ease with which a critic can question your generalized conclusion. Any stipulation or imposing specifications you must follow when conducting a study can also misrepresent themselves when generalized.

Validity and Video Game Design

     Bruce Shelley, known for his contributions to the commercially successful franchises Civilization and Age of Empires, expressed his guidelines for creating commercially fruitful endeavors in the game industry (Shelley, 2001). Shelley’s parameters can easily be coupled with the scientific validity established earlier to bring about a truly successful venture. One of the guidelines entails building upon the work of previous games. While imitating a successful game will seem lazy and uninspired, advancing and innovating upon a previous idea is a great way to take an old idea and make it fresh (Shelley, 2001). A way that a developer could follow this recommendation would be by employing the research setup that includes the cause construct, effect construct, program, and observation. This setup will allow developers to formulate a theory on consumer wants and needs, conduct the study, and then potentially validate their ideas. By knowing the desired effect, utilization of the effect construct can be used to reverse-engineer the necessary variables of the cause construct.

     Shelley covers a variety of other guidelines, but it is easy enough to boil them down to their roots: making an enjoyable experience for as wide of a fan base as possible, building upon well-known foundations of the gaming industry along the way (Shelley, 2001). The video game industry is dependent upon experimentation and growth and being able to validate a reliable variable will ultimately give developers a better idea of how to appease a widespread audience. Conducting research in this manner will present the same pitfalls, though. Developers risk over-generalizing statistics or not accounting for permutations to offset the results. Developers will have to remain vigilant in their endeavors to ensure they validate any perceived trends in the industry. The gaming world is a constantly causal system as the industry is constantly reacting to tendencies and permutations.

     Being able to approach the video game industry with a scientific outlook will provide game designers with the ability to couple two effective skills together to create a successful project. Understanding the causal system and how to attack it critically will provide you with a much better understanding on how to research and enact validity in the most effective way possible. Approaching problems or obstacles in the game world with a scientific mind will potentially yield a much higher rate of success.





References


Shelley, B. (2014). Gamasutra - Guidelines for Developing Successful Games. Gamasutra.com. Retrieved 3 August 2016, from http://www.gamasutra.com/view/feature/131450/guidelines_for_developing_.php?print=1

Trochim, W. (2006). Home. Socialresearchmethods.net. Retrieved 3 August 2016, from http://www.socialresearchmethods.net/kb/index.php



Disclaimer: This blog post was originally submitted in August of 2016 to Full Sail University under my Master of Science degree in Video Game Design

No comments:

Post a Comment