Simulations

Method Overview

In contrast with other research methods that seek to study past experiences, simulation focuses on “what if” analysis. Thus, it creates observations by “moving forward”. Most popular simulation methods that are used are computer simulations: discrete event simulation, system dynamics, and agent-based simulation. More qualitative and practical simulation types for education and social research include:

1)  Gaming

Gaming is a formalized play with pre-set rules that players must follow. Actual game applications include board games and computer games.

Advantages: data collection can be automated if delivered via computer network (leads to less chance of computer errors), controllable behaviour laboratory

Disadvantages: hard to validated research findings due to complexity of decisions of game participants and game rules.

Benefits for game participants: no punishment, intrinsic motivation, high participation rate, immediate feedback;

Drawbacks for game participants: absence of critique, no innovation (locking in best strategy), low realism, costly participation;

Computer Simulation (used for games & models)

• Data collection can be automated (less chance for errors and omissions)
• Subject to a computer or online affect on subjects
• Internet simulation via computer is even better as an administrator can completely manipulate, control, collect and research the informal channels of information between simulation participants

Board Game Simulation

• Some games have high detail-complexity and thus implementing a game without a computer is more difficult.
• A person must take notes during the board game simulation

2) Role-playing

Role-playing is a simulation method that requires participants to assume specific parts in defined social situations.

Advantages: allows observing participants when they confront novel and difficult situations, prepares participants for uncertain future

Disadvantages: participants may feel childish, may overact and fail to focus on the goals of exercise; trainers have no control over the immediate feedback and rewards; the number of people involved is limited; time-consuming and expensive to prepare for;

Benefits for simulation participants: role-playing develops skills in sales, supervision, leadership, interviewing, and other human interactions.

3) Simulators

Simulators are systems in which operators and machines interact in ways that approximate real life. They are extensively used for training purposes.

4) Modeling

Modeling is designing some form of physical or symbolic representation of a system, experimenting with it, and analyzing the results. The more developed the model is, the higher fidelity degree will be; thus, more complicated models result in higher-quality output information. However, lower level of models (with low fidelity) allow for simpler modeling, faster computation, and more detailed and complex analysis. Simpler modeling is especially effective when the researcher does not know much about the system.

Advantages

• substantially reduces costs and time;
• allows exploring different solutions to a problem and identifying design flaws;
• high flexibility & sensitivity analysis;
• allows for experimentation without disrupting the real system;
  • results in better understanding of the system represented;

Disadvantage: “garbage-in-garbage out”

Types of modeling: physical or symbolic, static or dynamic, deterministic or stochastic, normative or descriptive;

Procedure

General Procedure

1. Begin with a research question/hypothesis/theory to test
2. Choose a simulation approach that fits with the theory/hypothesis
   • Look for theories, where data is challenging to obtain
3. Create simulation (see below steps for each simulation method)
4. Verify computational/physical representation
5. Experiment
6. Analyze findings
7. Validate by comparing with empirical data

General Guidelines for Specific Simulation Methods:

Gaming:

1. Chose/construct games that have a relevance to your research hypothesis
2. Train evaluators/moderators of the game
3. Conduct a trial game
4. Provide an evaluation/questionnaire for participants and receive feedback from evaluators
5. Implement necessary changes to the game and game rules
6. Conduct actual game
7. Allow participants share how they changed their understanding of the game situation and modified their strategies at the end of the game

Role-Playing:

For general guidelines please refer to “Observational Research” method.

Modeling

I. Model design

Create a model that represents the physical system. A model can be

• a declarative (verbal description of the system and its operations)
• a spatial (a pictorial or graphic representation of a system, usually depicted in two or three dimensions
• a functional (a physical or symbolic representation of a system that portrays functional relationships)
• a multi-model (a model that consists of several different multi-models)

II. Model execution (on a computer)

Process sequence is required (a computer program that will “step” through time to update the variables in the model. Some of the modeling tools are: Extend, iThink, Stella.

III.  Model Analysis

Interpret a model’s graphs by noting how selected variables change over time

Advantages

• Simulation’s primary value occurs in creative experimentation to produce novel theory or test the existing theories
• Extremely useful for sensitivity analysis (“what if”)
• Particularly useful when the theoretical focus is longitudinal, nonlinear or when empirical data are challenging to obtain
• Can be beneficial and interesting for the subjects (thus, inducing participation)

Disadvantages

• Simulations are simply “toy models” of actual phenomena (either replicate the obvious or strip away so much realism that they are simply too inaccurate to yield valid theoretical insights)
• Might be costly and time-consuming to execute
• Sampling: some of the representatives of population participating in any simulation (game or role-playing particularly) may be worse/better in playing games or playing parts
  • Thus, noting each subject’s game/playing ability will be required
  • And more weight should be transferred to game outcomes of the less active participants

Recommended Usage

Usage of simulation methods will probably limited for SERA activities. However, it can greatly complement other research methods through concurrent usage of another research method and one of the simulation methods or/and consecutive usage of both methods.

Some peripheral usage ideas:

• Conduct role-play style surveys instead of typical surveys
• Through games (online and software) identify which type mind-mapping applications/programs (other technology proposed by SLD of Sauder) result in higher teamwork results/educational outcomes;
• Physical modeling can be applied for the Canaccord Learning Commons (combining modeling with other research methods to get a feedback)
• Classroom and study space design; New Learning Commons
• A summer class can be selected with a “moving” classroom: a classroom facilities/layout or location will change every time and observatory research conducted to determine the best layout for classroom
• Conduct case competitions for students on various Sauder topics (curriculum, etc.) (this will allow their voice to be heard, and find solutions for the issues); (in-baskets; similar to role-play)
• Simulate/introduce different teaching ways and evaluate outcomes (especially in Student Directed Seminars, where there is more freedom to course design ) (focus on enhancing student learning experience)

Sources