Rear End Conversations: A Content Analysis
...d that a nonrandom sampling method can be both effective and an ineffective way of gathering data. Merrigan and Huston (2004) demonstrated that “nonrandom selection method will weaken your ability to claim that your sampled results apply to some larger population, or that your sampled results adequately represent phenomena in a specific context” (p. 42). Focusing on a target group is considered a purposive sample. One limitation of purposive sampling is the lack of representatives; therefore, it is not possible to generalize the results to a population. Categorizing the Collection As the next step in our process, we placed the ten bumper stickers into content-based categories. Categorizing the data helped to code the bumper stickers. We then placed the bumper stickers into both broad and specific categories. Broadly categorized, four of these bumper stickers were humorous and the other six were non-humorous. We then further categorized the non-humorous bumper stickers into more specific categories. These three specific categories included feminist, political, and religious. Of these six bumper stickers, two relayed a feminist message, three represented political views, and one conveyed a religious opinion. Displayed by both male and female drivers, the humorous bumper stickers had a variety of messages. For instance, a female displayed a sticker “Sticks and stones may break my bones but chains and whips excite me” and a male driver showed the message “Mama ran off with the drive by truckers.” Both of these had humorous content. The array of these bumper stickers clearly indicated the driver’s personality with different levels of humor. The non-humorous stickers were focused on profound messages that expressed particular views. We found that stickers with a feminist point of view were more likely to be on vehicles with female owners. In one case we saw a male get out of a vehicle that displayed a sticker which stated “Well behaved women never made history.” We were confused by the message and then asked him if he was the owner of the vehicle. After he assured us that he wasn’t, we were somewhat comically relieved. All of the stickers with a political message held negative views of the government, such as “Don’t steal the government hates competition.” Overall, the non-humorous bumper stickers centered on personal self expression. Future Research Question The process of collecting and categorizing data left a great deal of room for expanding on our findings for future a research question. One study has found that “more than one-third of men and women ages 25-29 have a tattoo” (Gayagoy & Hammers, 2004 p. 32). Tattoos seem to be an intrinsic part of today’s society. There is an increase among individuals who want to express themselves or their ideas. One question that interested our group was is there a correlation between the types of tattoos people have to the types of bumper stickers displayed on their vehicle? Riffe (198) claimed that a “correlation between two variables is one that is associated with the other” (p. 38). Keep in mind that a correlation is not necessarily a causation. The first step in answering our question would be to conduct two separate content analyses, one of the bumper sticker content and another of the actual tattoos. In order to collect data needed for the analysis, a nonrandom, purposive sampling method should be employed. Using a coding scheme and carefully categorizing the data would help answer our research question. Reliability and validity are a significant part of content analysis and would ensure unbiased results. The target group of this research question would be tattoo parlors and their parking lots. For our collection process to be successful, we would need to solicit individual drivers for a brief demographic profile of their tattoos. “Tattoos, as a visual means to separate oneself from the normalized culture, can be thought of as a loaded choice, a choice that draws attention to oneself intentionally” (Bell, 1999, p.54). One way to achieve a good response would incorporate a variety of methods: place ads in local magazines, put flyers in local parlors, and ask for volunteers to participate in the study. Using the modified coding sheet would enable us to ga...