Critique and analysis of academic work

¶ … Peer and Media Influences to the Development of Purging Behaviors Among Preadolescent and Adolescent Girls by Field et al. discusses the relationship between the two factors (peers and media) on whether or not young girls will development purging habits in order to look like women they see on television, magazines and other popular media outlets. The author's reason for conducting the study was to find out more about how these young girls develop poor body images which lead to poor eating habits. It appears that the results of this study could be used for further study in the prevention of the causes of eating disorders among your girls.

The study consisted of selecting the daughters of women between the ages of nine (9) and fourteen (14). The women who were selected were participants in a program called the Nurses' Health Study II. Surveys were sent to the young girls to fill out that used information such as the girls' height, weight, and age at menarche among other questions. The survey also asked questions related to how thin the survey takers friends were, whether or not they dieted or used laxatives to control their weight, and whether or not the media and friends had a strong influence on how they viewed their appearance.

The way in which the size of the population was determined is not explicitly discussed, but it is inferred that the number of participants chosen had a direct correlation with the number of women participating in the Nurses' Health Study II. Since the children of the women in this group were the ones chosen, it is assumed that this is how the population size was determined. Because the participants were chosen from one particular group of women, bias definitely plays a factor. Gerhard says that researchers must know the difference between the study population and the source population. The study population is comprised of participants included in the final study, but the source population refers to where the study population originated (2008, p. 2160). The researchers did not consider the source population for this study and when straight to a predefined study population which could make the results of their research skewed.

The women chosen in this study should have included more variety and should have come from various backgrounds. If this had been done, the outcomes would have more than likely been different. They should have chosen the daughters of women from all walks of life and not just the daughters of women who were participants in the Nurses' Health Study program. It can be assumed that the women who participated in this program had many things in common. It stands to reason that their daughters would probably have many things in common as most people from similar backgrounds do. It would have been wise to reach out to various communities and other programs where women were participants and had young daughters. In this way the researchers could have gotten a more diverse outcome from their survey.

Because there appears to be bias in the selection of participants, the information collected from them is probably biased by default. The researchers were looking for ways in which these young girls are influenced that causes them to develop bad behavior and habits such as taking laxatives and dieting at such an early age. If they had cast a wider net and got a more diverse sampling of participants they may have found that body image is not an issue with girls who are from different backgrounds than what they have chosen to study. Miller et al. say that researchers should always be aware of selection bias and that is should be an ongoing process whenever a hypothesis is formulated (2001, p. 38).

Even though the opinion is that the study was biased, the researchers did a good analysis with the information they received from the young girls. The researchers focused in on the results of the questions that had would have the most impact on the study. They seemed to be most interested in the age of the girls and the Tanner stage of pubic hair. It is assumed that the researcher felt that the lack of or presence of pubic hair in the girls signified the onset of puberty which is the stage of life that the study is most concerned with. Of the almost 7,000 girls surveys, only a small percent exhibited behavior indicative of being overly concerned with body image.

The interpretation of the data is consistent, but does not totally support the unstated hypothesis that the media as well as peer pressure play a large part in how young girls feel about themselves and their body image. The conclusion they came to was that the media and peers have an influence on young girls and body image and that the media should use more models and women the look normal or like most of the young girls themselves so that eating disorder prevention programs can be more effective. The reality is that only one percent (1%) of the girls showed signs of behavior that was indicative of being affected by their peers and the media. However, a large percentage of this 1% did not show signs of improvement one year after the study.

The researchers received an excellent response rate from those surveyed. However, they have actually done a disservice to the entire study by not being more diverse in sample selection and casting a wider net. In addition to casting a wider net, the results of the research were based solely on responses to the survey. A better and more accurate response to many of the questions on the survey could have been received if the researchers had done interviews (with at least one parent present) or even focus groups. Though not a fact, sometimes with surveys people may not be as truthful as they would be face-to-face. Also, if a person is being interviewed there are certain things the interviewee can do to make the person more relaxed and at ease for more accurate answers.

In a focus group the researchers can use tactics to make the girls open up and feel comfortable around their peers. All it takes is one person to be candid about their experiences and soon others will open up and discuss their true feelings. If the researchers had done more than surveys as well as broadening their selection of participants, they would have received not only a much better, but more accurate response. The researchers state that their study was consistent with one that was done a year ago. In theory this is true, but it still does not mask the fact that there is considerable selection bias with this study. When selection is done in a nonrandom manner, bias occurs (Winship and Mare, 1992, p. 327).