Test: Sociocultural factors in obesity
The following assessment is something that I give to my students to test their critical thinking skills. I provide the research summaries for them and then ask them to write three paragraphs in response to questions that show critical thinking.
Two of the studies below I have done in class with them - one is not familiar to them.
I use a revised critical thinking rubric to award up to 7 marks for each response. Student samples are included below.
A shareable student version of this page can be accessed here to share model responses.
Study 1. Martinez-Gonzales et al (1999)
A large-scale study of 15.239 people was carried out in the European Union to determine the relationship between sedentary behavior and obesity. The sample was a stratified cluster sample - that is, stratified for each of the 15 countries - made up of males and females ages 15 and older.
Professional interviewers carried out structured interviews in the homes of the participants. The participants were asked about the amount of time they spend in various physical activities. A sedentary lifestyle was assessed by self-reported hours spent sitting down during leisure time. In addition, BMI measurements were calculated.
The researchers found a direct correlation between those who had high amounts of sedentary time (35 hours per week) and obesity and an inverse correlation between those with high amounts of physical time and obesity.
Study 2. Currie et al (2010)
Currie et al (2010) carried out a correlational study in the USA with data from four states (California, Michigan, Texas, and New Jersey) comprising roughly 3 million children. Her research found that among ninth graders, a fast food restaurant within 0.1 miles of a school results in a 5.2 percent increase in obesity rates. Non-fast food restaurants showed no significant correlation with weight outcomes.
Study 3. Forslund et al (2005)
Forslund et al. (2005) undertook a cross-sectional study in 22 medical centers in Sweden, to investigate how snacking may influence weight. They used questionnaires to compare the energy intake (estimated in kilocalories) of 4259 obese men and women with 1095 controls. They were interested in how much of the energy came from meals and how much came from snacking. Snacking was defined as eating between the main meals. The participants had a physical exam and blood tests were taken.
The obese participants ate on average six times a day, compared to five times in the non-obese group. Obese participants were also more likely to eat later in the day than the non-obese. The researchers noticed that, overall, obese participants reported a significantly higher energy intake than the control group.
Snacking was more frequent in the obese group, and women were more frequent snackers than men. Generally, energy intake was more likely to come from sweet, fatty food choices in obese frequent snackers. The non-obese participants ate more healthy snacks.
Questions
- Which study do you feel is the best evidence for a potential cause of obesity? Write a paragraph that explains the strengths of the study and why it is better evidence than the other two studies.
- Which study do you think is the weakest of the three studies? Justify your position by discussing the limitations of the study compared to the other two studies.
- Explain at least two obstacles to studying the role of sociocultural factors in obesity.
Assessment
To assess each response, I use the following edited IB critical thinking rubric. I award up to 7 marks for each response for a maximum of 7 marks. The mark bands for this assessment are in the table below.
Markbands
| 7 | 6 | 5 | 4 | 3 | 2 | 1 |
| 19 - 21 | 16 - 18 | 13 - 15 | 10 - 12 | 7 - 9 | 4 - 6 | 0 - 3 |
Student sample: Question 1
Out of all 3 studies, Martinez-Gonzales et al (1999) provides the strongest evidence for a cause of obesity. First of all, the study uses a large sample size of 15,239 and a stratified cluster sample across 15 countries, with both genders represented. This indicates that the findings may be generalizable to the population it was drawn from. Additionally, the study used a structured interview and a highly standardized procedure, meaning that it is easily replicable and thus more reliable. On the other hand, this study has potential issues of construct validity. The BMI is a highly problematic measure of obesity. A person who is healthy could actually score high on BMI because of muscles.
It is also difficult to measure the variable of a "sedentary" lifestyle, and thus that measurement may lack construct validity. The study defines a highly sedentary lifestyle as 35 hours per week. Over the two decades since this study was carried out, our definition of highly sedentary may have changed, and thus this study would need to be replicated again in a more recent time. This study alone does not address the interaction of environmental factors that may contribute to obesity, as well as culture class, and gender. While the study may have developed a strong correlation, this is not an indicator of causation; a sedentary lifestyle may be a symptom of obesity but perhaps not the cause. However, with these limitations in mind, the study still provides better evidence for the cause of obesity in comparison to Carrie et al and Forslund et al. Currie et al's study is correlational, unable to establish solid cause and effect, however, it is even weaker in arguing for the cause of obesity - it doesn't even look at the amount of fast food that participants consume, but simply the proximity of a fast-food restaurant to their school. Forslund et al's cross-sectional study argues for eating habits as the cause, but it assumes that all obese people share similar characteristics and all calorie intake is equal - which has been disproved.
This is a good response. The comments relevant to temporal validity are not very strong. The clarification about the correlational nature of the study is an important aspect of the response. This response was awarded a 6.
Student sample: Question 2
The weakest study out of the three is Currie et al's (2010) study. The results of the study suggest a correlation between obesity rates, and how close a fast-food restaurant is, this makes it vulnerable to bidirectional ambiguity. The study assumes that access to fast food influences obesity, the issue with this is that you cannot establish whether or not it was solely the fast-food access that influences the obesity rates of the children, or if there were other co-founding variables such as socio-economic class, culture, or stress. This also questions the construct validity, you cannot establish to what extent is this study really measuring the influence of fast food or diet on obesity since it only looks at the distance between the fast-food chain and school. Although the study consisted of a sample size of 3 million, the sample was vulnerable to sampling bias as the participants all were from the US. The eating culture of the US differs from the rest of the world, therefore you cannot generalize the findings of this study to kids outside of American culture. The results suggest only a 5.2 percent increase in obesity rates, this questions the extent to which access to fast food actually causes obesity.
The comment about bidirectional ambiguity is not correct. We cannot say that we do not know if the presence of fast food led to obesity or if the presence of obesity led to fast food. This needed more clarity. The comment about construct validity is also not correct. The fact that the proximity of restaurants that were not classified as "fast food" is a significant omission in the response. The sample size is relevant, but the comments about culture are vague and not really linked to why this may influence our ability to make a link between fast food and obesity. The actual findings of the study are not clearly understood. This response was awarded a 4.
Student sample: Question 3
One of the obstacles that can have an effect when studying the role of sociocultural factors in obesity is the problem with generalization. This is because a lot of the time studies are specific to one culture, and tend to not be cross-cultural. This is an obstacle because different cultures have different norms for which they consider what is healthy and what is not. Therefore, it would not be possible to apply results from a study conducted on solely participants from the US to people that live in Japan, for example. This is because the people that live in the US have different norms, and even, for example, their portion sizes, or what they eat is different from the people that live in Japan. This means that a lot of the results gathered from sociocultural studies have a problem with generalizability unless they take this factor into account. Another obstacle that comes with studying the role of sociocultural factors in obesity is that it is often difficult to draw a direct cause-and-effect relationship. This is because a lot of the studies have a problem with bidirectional ambiguity, meaning it is not sure if the variable x causes y or if y causes x. Especially when supported by the factor that a lot of these studies are not longitudinal, it is difficult to establish this relationship and to truly know which variable has a direct effect on the other, and the direct effect on obesity.
The first example is not really an obstacle to research. There are universal norms of health. The definition of obesity has been established by the WHO. The comments made are not relevant. The second half of the response is better but could be more developed. This response was awarded 3 marks.