Using the article, "Differential Effects of a Body Image Exposure Session on Smoking Urge Between Physically Active and Sedentary Female Smokers" by Nair, Collins, and Napolitano (2013), how...


Using the article, "Differential Effects of a Body Image Exposure Session on Smoking Urge Between Physically Active and Sedentary Female Smokers" by Nair, Collins, and Napolitano (2013), how would one evaluate the article and critique the statistical analysis employed in the study? How would one examine the assumptions and limitations of the study?

Expert Answers
Tamara K. H. eNotes educator| Certified Educator

The article titled "Differential effects of a body image exposure session on smoking urge between physically active and sedentary female smokers" reported a study conducted to examine the correlation between female smokers' urge to smoke, their body image, and physical activity. The authors of the study took a sampling of 16 "sedentary" women, meaning women who don't typically exercise, and 21 women who exercise regularly, all of which who had smoked less than or equal to five cigarettes a day for the past 6 months. The authors then took the sampling and exposed them to body images with the intention of making them feel concerns about their own body image and weight in order to observe whether or not the women felt the urge to smoke after being exposed to the body image. The results showed that more of the physically inactive women self-reported feeling the urge to smoke plus were faster to take the first puff. In contrast, fewer physically active women self-reported feeling the urge to smoke but there were no similar decreases in how long it took the active women to take the first puff. The scientists concluded from their study that body image has a great deal to do with smoking urges, and women smokers concerned with weight may feel fewer urges to smoke as they increase physical activity. As full access to the article online is limited, below are some ideas to help get you started on your assignment.

To conduct a statistical analysis, you want to identify the hypothesis that was tested and assess whether or not the researchers sufficiently analyzed the results of the study to either validate or invalidate their hypothesis. The first step is to identify either of two hypotheses, called the null hypothesis  and the alternate hypothesis. The null hypothesis is a statement about a sample that can be assumed to be true until proven incorrect; an alternate hypothesis is a statement that contradicts the null hypothesis and will be accepted once the null hypothesis is rejected. In this study, the researchers recognized as true the understanding that women often use smoking as a faulty method of controlling weight and also saw that physical activity and smoking can have the same perceived benefits, such as "enhanced mood, reduced anxiety, and weight control" (Nair, Collins, Napolitano). Therefore, they claimed as their null hypothesis that a woman's body image will also have a direct effect on a woman's smoking habits, influencing her to smoke more. So, as you assess the results of the study, this is the hypothesis you want to keep in mind to see how well the researchers analyzed the test results of the hypothesis.

You next want to identify how the study was designed. As stated above, we already know groups of both physically-active- and inactive-smoking women were tested and how many in each group. We also know that the researchers exposed each woman in both groups to a body image. As you read the article more fully, you'll want to see if the authors' go into greater details about exactly how they exposed the women to body images and make a note of that.

You next want to state how the statistical test was conducted, and you have several options, such as a two-way ANOVA, a mixed model ANOVA, or a test with factor A fixed and factor B random. We already know that the authors tested two groups. You also want to know that a two-way NOVA is a comparison between two groups with two independent variables to see if there is a relation between the independent and dependent variables. For example, one might conduct a study to see if both gender and educational level plays a role in test anxiety for university students. In such a test, the two independent variables would be gender (male vs. female) and education level (undergraduate vs. graduate) while the dependent variable would be test anxiety ("Two-way ANOVA in SPSS"). In our authors' study, all of the test subjects were women smokers, so smoking women is the dependent variable. The authors also comparing whether or not body image influenced the urge to smoke for both physically active and inactive women; therefore, our two independent variables are body image and physical activity vs. inactivity. Therefore, we know that our statistical test was conducted as a two-way ANOVA.

You will next want to state what the conclusions of the test were and how the conclusions relate to your hypothesis. Finally, you want to critique the statistical methods used. Ask yourself a series of questions: Do you think the two-way statistical test was the only way to conduct the study? Do you think the presentation of the body image was telling enough? You'll be able to think of many other analytical questions to pose to yourself as you deeply study the material.   

Your next task for the assignment is to state any assumptions the study was based on or any limitations the study had. To be able to do this, you first want to figure out if the research was quantitative or qualitative and exactly what the limitations of both quantitative and qualitative research are. Quantitative research generalizes data that results from testing a study sample, while qualitative research looks at underlying reasons and motivations behind a proven hypothesis. Since our authors' study tested a sampling of subjects and the authors drew conclusions based on such things as the number of women to self-report the urge to smoke and the length of time it took the sampling of women to take the first puff, we can say that the research is more quantitative than qualitative. All quantitative studies assume that the measured results can either prove of disprove a hypothesis but also "do not account for the subjective nature" of the decisions made based on the data ("Evaluating Research Methods"). You might consider that one limitation to the study is that the researchers observed fewer self-reports of the urge to smoke in physically active women but not decreases in the amount of time it took physically active women to take the first puff. The researchers concluded from this that the physically active women felt fewer urges to smoke, but the data may be inconclusive if the women who did not self-report the urge to smoke also did not take a puff of smoke at all.