Friday, October 2, 2015

Kids who Breathe More Pollution Have Lower Grades

A child plays near the Valero Oil Refinery in the Manchester
neighborhood of Houston, TX. (Source: Earthjustice)
The link between air pollution and increased respiratory disease in children has been well studied.  There are still questions, however, about how such pollution may impact a child's academic performance. While some studies have looked at the link between air pollution exposure and academic success, these studies have been limited by their resolution, primarily using school-level data which can miss the effects of individual factors, like socioeconomic status or parental education, on school performance.

A new study published in the journal Population Environment looks at this problem at the individual level, and demonstrates a link between exposure to air toxic pollution and a lowering of GPA when controlling for several factors also known to impact school performance.  The study used data from EPA’s National Air Toxics Assessment (NATA), an inventory which estimates concentration of air toxic pollutants, as well as household and school performance information gathered through mail surveys.  Even when controlling for a variety of factors (economic deprivation, mother’s education, age of mother at birth, race/ethnicity and mother’s English proficiency), seven of the eight NATA risk variables were significantly negatively correlated with a child’s GPA. In other words, all other things being equal, children exposed to higher doses of toxic air pollutants will have lower GPAs than their peers.  The online publication, Mother Jones, covered the study in its recent article “Kids Who Breathe More Pollution Have Lower Grades”

In general, the Mother Jones does a good job summarizing the main points of the study and conveying the conclusions of the research.  The final paragraph of the MJ article includes a quote from the study’s author, Sara Grineski: “The finding that there is a significant association between residential exposure to air toxins and GPA at the individual level is both novel and disturbing…these findings provide another piece of evidence that should inform advocacy for pollution reduction…”.  This really is the main point and conclusion of the journal article and MJ does a good job to leave the reader with that sentiment.

The MJ article, however, fails to discuss the limitations of the research. It is really difficult to estimate individual level exposure without direct monitoring and the study’s data sources leave a lot of room for error. The EPA’s assessment of air toxic concentration is based on calculation and modeling, meaning that it is likely a close, but not exact estimate of ambient air toxic concentration. Further, the study assumes that all children living within a census block experience the same exposure, which is also likely untrue.  Concentration at individual houses will be variable due to weather patterns, some children may spend less of their time at home, some may spend more of their time playing outside where they are exposed at higher rates, etc. While the study mentions these limitations, the Mother Jones article does not include information about the data used in the study. Without looking at the original article, the reader would likely assume that the data comes from direct pollution measurements and that exposure estimates for individual children are exact.

The article links to the MJ cover story about a growing body of research linking ultra-fine particle pollution to dementia and Alzheimer’s disease.  While this provides the reader more information about how pollution may impact student’s GPA, it also implies that the journal article demonstrates a clear link between air pollution and lowered neurological function. The journal article only shows that higher exposure is correlated to lower GPA. This could mean that breathing more pollution is directly worsening school performance by impacting cognitive function, but it also could mean that the children are missing more days of school due to other pollution-related illness. By framing this study in terms of the Alzheimer’s work, the MJ article draws a conclusion not demonstrated by the research.

While they don’t get all the nuance of the data, and perhaps go beyond what the study’s authors have proven, the MJ article does a good job presenting the information in an accessible and engaging way and highlights an important area of environmental justice research that ought to be better understood by the general public.

14 comments:

  1. We discussed ethics briefly in class, but I think it is important to note that the researcher, Sara Grineski, found the results "disturbing". This study was observing a trend to find a correlation, and did not involve an experiment to determine a causation. However, I think the ethics of the situation should be evaluated by the researcher. These children are placed into a situation of which they have no control, and the research presented in this article highlights detrimental effects to their academic success. I think studies like this should be able to invoke policy change, or induce further studies that will help to address the issue.

    ReplyDelete
    Replies
    1. I think you're right! This study (and a lot of EJ research generally) aims to understand relationships, and the many ways that children may be impacted by air pollution. We don't need to know that air pollution is lowering student's GPAs by .031 exactly, just that it is lowering them.
      On the question of ethics in studies like these, I wonder if any information on the study's findings were given to the participants. You could argue that, because the children were not being exposed to more pollution as a result of the study, there are no issues with ethics. But is it ethical to not provide the findings to the family so that they can make informed decisions about their exposure? Especially since there would be no research if it weren't for them?

      Delete
  2. I agree that the MJ article does presents the correlation between pollution and GPA very well, butut the MJ article provides no reason to question the results. It is pointed out that some adjustments were made to compensate for many variables, but it was not mentioned that the grades reported to the parents are inherently inflated by the educational administrators.

    ReplyDelete
    Replies
    1. You're right that the MJ article doesn't bring up the issue of grade inflation mentioned in the journal article. Although, the researchers note that this is probably dampening the negative correlation with pollution (when there's less variability overall in GPA, no factor can account for significant GPA drops).

      Delete
  3. I think it's interesting to see the social impacts of pollution alongside the physical ones, however I wonder at the correlation of air pollution to grade point averages when the root of the issue is pollutant-caused disturbance of neurological function in children. Out of genuine curiosity, why do researchers perform studies for correlation to "softer" issues like grade point averages (rather than health-related, or hard science issues)? Do these more "personable" studies strike closer to home with the general public or government groups? Does this make the root issue (in this case, air pollution) more likely to be acted upon, or does this give more awareness on the issue?

    ReplyDelete
    Replies
    1. I think this is an interesting point. It does seem plausible that the reason these kind of topics are investigated because it connects the public. It would seem easier to convince people that the children are performing worse in school than discussing neurological function in a bunch of "fancy" science lingo. Remember that a large majority of the population is not going to be science literate. In order to bridge that gap, focusing on topics such as GPA might be your only way.

      Delete
    2. So, I think that studies like this are important for teasing out the intersection between environmental health, poverty, race and inequality. I would caution against dismissing these types of studies as non-scientific and argue instead that they are looking at how science relates to social factors. The reality is that pollution does not just impact physiology, it impacts productivity (via missed days of work or school), opportunity (lead exposure in children can cause behavioral issues which hold them back academically and professionally), and economic success (if you live near a heavy polluter, your property values plummet). Understanding how pollution impacts biology is very important, but we also have to understand how exposure can impact someone over a lifetime which is what this study gets at.

      Delete
  4. The scientific article's conclusion that the effect of air pollution on children gpa was mild and not noticeable at an individual model amazed me. I wonder if this study is reproducible in more high pollution areas like China. What about the effects on standardized scores, which seems like a better variable to choose from because grades are a bit more subjective based on the school and teacher. Overall, the mother jones article did an okay job but I thought they should of mentioned like you, Katie, the fact that the GPA drop could be attributed to missed school days.

    ReplyDelete
  5. Overall I think the Mother Jones did an adequate job summarizing the study. The study itself had a pretty thorough description of limitations of the data and conclusions, and it could have been more transparent to include some of the larger issues in the Mother Jones study (like the most recent USEPA NATA data was from 7 years previous to the study, and a 30% response rate on the survey). However, I understand a discussion of those factors would only weaken the argument.

    In the discussion section of the study, the authors mention that the relative effect on GPA is mild, and because of the mildness it will likely go unnoticed. They then extrapolate the effects of early low academic performance throughout the lifetime of a person. This seemed like a bit of reach for me, and I agree with the authors that a longitudinal study would be a good follow-up to see the long-term, non-health related effects of air pollution.

    ReplyDelete
  6. My thought process was similar to Olivia's. If there is already some data (at least preliminary data) that ultra fine particulate is messing with the human brain in some way, shape or form why are the studies being done so "superficial"? I also agree with Yonas on grades being a bit more subjective depending on teachers and schools. In my opinion, a 0.031 difference can be attributed to this subjectiveness. From my experience with my past teachers and as a GSI, each person is their own small universe so some variables may not be kept constant in that way.
    PS: I just love how these writers pick out the most fitting quotes from the authors to highlight the point of their article.

    ReplyDelete
  7. Good post. The scientific article cann't build a relation between low GPA and air pollution without ruling out other reason which may also result in low GPA. Besides, the article can do better if they can measure the exposure level of children instead of using modeling and assumption with some portable direct reading monitors.

    ReplyDelete
    Replies
    1. They have done relatively good job controlling for other variables (mother's education level, poverty, etc). And I agree that personal exposure monitors would be ideal, but they are often cost-prohibitive. This area of research is generally very underfunded, and it's important not to dismiss the findings of a study because it did not have access to an ideal data set.

      Delete
  8. A lot of people seem to be against the modeling system used but this seems difficult to address otherwise. I've no expertise in this field but I'd like to hope that a incorrect modeling system wouldn't have gotten through peer reviews. I'd like to assume the methods used in this paper for data collection were in line with the field and appropriate for this particular study.

    That said I find the 0.031 difference in GPA too small to draw any major conclusions. I'd suspect the derivation for such a complicated metric as GPA is is too large for such a minor change to be significant.

    ReplyDelete
  9. I have issue with the scientific paper drawing such conclusions upon the method applied and the results obtained. I don't think there are any definitive conclusions that could be made from the study done. I would like to see proof that the models used can achieve the accuracy they are claiming

    ReplyDelete