Evidence suggests that cell phone usage and car accidents are correlated. Currently, twelve states prohibit all drivers from using hand-held cell phones while driving (Governer's Highway Safety Association, www.ghsa.org). These laws are supported by research and hypothesis such as this one: "The NSC model estimates 21 percent of crashes or 1.1 million crashes in 2010 involve talking on hand-held and hands-free cell phones" (National Safety Council, www.nsc.org). However, correlation isn't causation. In order to avoid falacy, researchers cannot rule out rival hypothesis. Virginia Tech conducted research and deducted that talking on a phone isn't dangerous. "Our analyses separated talking and dialing tasks and found that talking on a…show more content… If the hypothesis turns out to be correct, the participants should not be in real cars. Each condition, such as type of simulation, environment (i.e. climate), time of day, and duration of the simulation should be exactly the same for every single participant. These are the dependant variables. Half of the participants will "drive" their simulation with these variables only and in a controlled setting, with no cell phones. This will be the controlled group. The other half of the participants will "drive" their simulation in the same situation, but with the independant variable of a cell phone. As this hypothesis concerns talking on a cell phone, not texting or dialing, participants in the experimental group will answer questions and respond to a researcher speaking to them on a hand-held device. The researcher will repeat the same questions and statements for each participant in the experimental group. Using said research design, if there are any differences in driving performance researchers can conclude that it is due to the independant variable of phone usage and not any other factor.
As previously stated, each person in the pool of participants should have an equal chance to participate. This way, each participant has an equal chance of being assigned to the control group as well as the experimental group. This also ensures an equal