Sociology 126

Health of the Public: Medicine and Disease in Social Context

Syllabus, Spring 2015

Course Description: This course examines the social causes and context of illness, death, longevity, and health care in the USA today. Who stays healthy and who falls ill? Who has a long life and who has a short one? What is a good death and why do so few Americans achieve it? What is good medical care, who gets it, and why? What role do physicians play in producing health in our society? To what extent do factors outside individuals’ control (factors such as genetics, geography, social networks, parental traits, or hospital quality) influence health and health care? Does socioeconomic inequality in society harm individual health? Do certain kinds of social networks or neighborhoods improve health? How do social factors get under our skin and literally become embodied? What are the collective constraints on individuals’ life prospects? What is the difference between an individualistic and a public health-oriented perspective on illness? And what issues of ethics and justice are raised by such questions? Would a different organization of society, different public expenditures, or different public policies matter? While exploring these questions, we will also consider how social scientists, biologists, epidemiologists, public health experts, and doctors address them—how they use theory to understand them and how they make “causal inferences” based on observational or experimental data. However, students are not expected to have in-depth knowledge of social science methods or statistics. The readings span the medical, public health, and social science literatures, and they reflect both qualitative and quantitative approaches. They also introduce new areas of “biosocial science” and techniques of “big data” as applied to health. In many ways, this course serves as an introduction to the field of public health.



Q: I’ve never taken a Sociology course before; will this be a good fit for me?

A: Yes.  If you have any interest in the relationship between health and society, the origins of health, and the role of medical care, it will be a terrific fit.  In past years, students from diverse majors, such as sociology, statistics, psychology, government, anthropology, economics, applied math, philosophy, biology, chemistry, physics, EEB, MBB, history of science, history, and many other majors have enrolled in the class.

Q: I’m interested in public health; will this be a good fit for me?

A: Yes.  The readings span the medical, public health, and social science literatures, and they reflect both qualitative and quantitative approaches.  In many ways, this course serves as an introduction to the field of public health.  The syllabus for the current year of the course is here.

Q: Does SOCY 126 count for Global Health?

A: Yes.  It is also Global Health 140.

Q: I understand that Dr. Christakis tinkers with the class?

A: Yes.  The class has been very well received by students for many years, and much of what students like about the class will remain unchanged – including the types of readings and the entertaining lectures that Dr. Christakis offers.  But we are going to try to make the class even more stimulating and enjoyable, if we can.

In 2011, Dr. Christakis tried a couple of innovations to make better use of the lecture time and increase student engagement with the material.  These innovations are modeled in part on processes advocated by Professor Eric Mazur; and they are prompted by the recognition that teaching should change as technology changes and as new research emerges regarding adult learning.  [For some more details, see E. Mazur, Peer Instruction, Prentice-Hall, 1997; and D. Bruff, Teaching with Classroom Response Systems: Creating Active Learning Environments, Jossey-Bass, 2009.]

Briefly, it is fair to ask what the objectives of lectures in a large class are.  On the one hand, lectures are a form of live entertainment, and there is an authenticity to being in a class and to watching someone lecture in real time that helps in the transfer of information.  Plus, having to get out of bed (even in the late afternoon!) and go to lecture concentrates the mind (though perhaps not as much as the prospect of being hanged).  And thinking about the material while taking notes facilitates learning for many people.  Dr. Christakis enjoys giving lectures (and, actually, he enjoys listening to lectures given by others).

On the other hand, in some cases, lectures can amount to nothing more than the ‘transfer of the professor’s lecture notes to the student’s notebook, without going through the minds of either.’  Plus, given new technologies such as cell phones and laptops, it is very tempting to students to come to class and spend time checking their Facebook pages or emailing their friends – which seems a waste of time for them, and a distraction to their neighbors.

Hence, the challenge is to make good use of scarce lecture time and to keep the benefits of lecture, while supplementing the experience in ways that exploit new technologies and make it more stimulating and instructive.

For example, during class, Dr. Christakis may sometimes employ two techniques to enhance information transfer from him to the students and among the students themselves.  One involves asking students conceptual questions during lecture and encouraging them to discuss their answers with their neighbors – a form of “peer instruction” that most students find both enjoyable and compelling.  Second, he may employ some new software in class that can be used to conduct and demonstrate social science experiments in real time.

You can watch the peer instruction technique in this brief, two minute video, if you are curious.

Q: Do many pre-meds take the class?

A: Perhaps 20-25% of the students taking the class are pre-meds.

Q: Do I need any statistical or quantitative background for this class?

A: No.  We’ll be considering how social scientists, epidemiologists, public health experts, and doctors use theory to understand health outcomes and how they make “causal inferences” based on observational or experimental data; however, students are not expected to have in-depth knowledge of social science methods or statistics.

Q: There was an equation in one of the slides and this scared me.  I know you’re going to pull something sneaky and ask me to calculate this for the midterms or final.

A: No.  We absolutely will not ask you to calculate Gini coefficients, Yitzhaki inequality measures, or age-adjusted mortality for the midterms or final.  We do, however, want you to understand the intuition behind these kinds of equations and why it might be useful for public health reasons to try and calculate things like this.

Q: Will the slides from the lectures be available online?

A: Yes.  We’ll periodically distribute the slides in pdf form online.  Probably, the slides will be uploaded a few days after each lecture, but possibly we will upload them in advance.  Moreover, the lectures are podcast; you can download the audio and files here, or subscribe in iTunes.

Q: Will Dr. Christakis discuss social networks in this class, and will this topic be covered too much or too little?

A: It will be just right.

Q: How much reading is there for the class?

A: Readings from books and articles average about 60 pages per session (range 15-200), or 120 pages per week.  It is our sense that this is not an excessive amount of reading; that this amount is typical of other undergraduate classes; and that the amount of time required (based on past student surveys) is not unreasonable.  Plus, we hope the carefully chosen and diverse readings will engage your interest and prompt deep understanding of the topics at hand.  We’ve got everything from sex to robots in the readings.

Q: Some of the readings look old, like from before I was born.  Why is that?  Is Dr. Christakis too lazy to update the syllabus?

A: The readings are from as early as 1971 and as recent as this year.  About 10% of the readings are updated each year, but some of them are either classics that are useful or fun to read, or are still the best examples of the topic of interest, and so they live on in the syllabus.  Not every old thing is clueless.

Q: Does Dr. Christakis really mean it when he says to come to office hours?

A: Yes.  Dr. Christakis is very engaged at office hours (Thursdays, 4:00–6:00 pm, at 17 Hillhouse, during the spring semester).  You can make an appointment or drop in.  Oftentimes, a conversation gets going among a bunch of students who have dropped by.  The best thing you can do to help Dr. Christakis get to know you and remember your name is to come to office hours.


Q: Can I take this course P/F?

A: Yes.

Q: I’m a senior, and will be working on my thesis this term, but it’s due around the time of one of the SOCY 126 midterms.  Can I have an extension?

A: No, sorry.  We recognize that a number of majors have theses due just before or just after our midterms.  Our sense is that experienced seniors are in a position to take the midterms and also complete their theses.

Q: The course looks interesting, but I don’t really have room in my schedule; can I audit?

A: Only if space allows, and with permission of the instructor; please defer to those around you who are taking it for a requirement.

Q: How are students assigned to sections?

A: We do online sectioning the first and second weeks of class.  Given all the moving parts, this is a difficult challenge, but we offer many section times and, in the end, virtually all students find a time that works well for them.  The TF’s for the class may come from several departments and schools in the university, and we often are fortunate to have some TF’s from the School of Public Health, the Medical School, and the Law School (though this varies from year to year).

Q: I’m a graduate student; can I take this class?

A: Yes. But graduate students taking SOCY 126 for credit should see the instructor in order to arrange different requirements.

Q: Why is there no course-pack?

A: Given very low demand in prior years, there is no course packet available for purchase.  Readings are available online, and also linked via the course website for you to print out.  A course packet of readings is also on reserve at Sterling Library.

Q: I have a class that overlaps completely with one of the weekly lectures.  Can I catch up via the podcast?  I promise to be diligent about keeping up with the work.

A: No.  We’re sorry, but this cannot be accommodated.

Q: What is the historical grade distribution?

A: Roughly 20-25% get an A, 20-30% get an A-, 15-25% get a B+, 10-20% get a B, and 5-10% get a C range grade.

Q: This is a large class with many sections. Are students in other sections being graded the same as me?

A: Previous students have been worried about consistency across TF’s in grading.  All TF’s follow the same grading procedure for evaluating your contributions to section.  We’ll also take any remaining TF variation between sections into account on your final grade.  Dr. Christakis is committed to personally reviewing all the grading and assigns final grades.

Q: How can I be sure my exam will be graded fairly?

A: Because we are obsessive, fanatical nuts.  We often study our own grading (courtesy of David Kim, MD, PhD) with the following results:

For example, one year, we have analyzed whether any irrelevant extra-textual factors predict students’ midterm scores.  For instance, we characterized answers as follows: we eyeballed the percentage of the page filled by each answer, ranging from 30-150% (yes, answers extending halfway down the verso); the mean was 86%.  Exactly 50% wrote in pen, 50% in pencil.  Only 15% wrote in cursive.  We also have wondered whether scores improved as grades moved through the alphabetized stack, which, if real and significant in magnitude, might suggest biased grading, e.g., increasing forgiveness as exhaustion/disillusion sets in among the graders.

Consider the following (econometrically suboptimal) linear model:

Question 2 score = %Page filled + Pen + Cursive + Alphabetical rank of last name

Where “Pen” and “Cursive” are indicator variables.

Coefficients:   Estimate    Std. Error   t value   Pr(>|t|)
(Intercept) 12.400229 0.956427 12.965 < 2e-16 ***
Pct_page  0.032386 0.010018  3.233 0.00135 **
Pen -0.273912 0.362933 -0.755 0.45095
Cursive -0.731711 0.497782 -1.470 0.14253
Alpha_pos  0.002791 0.001464  1.906 0.05748 .


Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 


Residual standard error: 3.269 on 331 degrees of freedom
88 observations deleted due to missingness)
Multiple R-squared: 0.04283,
Adjusted R-squared: 0.03127
F-statistic: 3.703 on 4 and 331 DF, p-value: 0.005773 

What would you guess is the effect of filling up more of the page?  There is one (p<.01), but it seems surprisingly small in magnitude, i.e., on average 0.8 points more for every additional quarter of a page of writing.  And thankfully it’s not, in general, the case that only long answers got good scores:

Figure 1

Thankfully, pen vs. pencil and cursive vs. print weren’t significantly predictive.  Alphabetical rank of last name had a borderline significant (p=.06) coefficient, so we now shuffle the exams.

Finally, it should come as a relief that a model based entirely on predictors that ought to be totally irrelevant has an adjusted R-squared of 0.03.  Alas, many papers are published in the social sciences with equally low R-squared values.

What is the effect of student year (freshman/sophomore/etc.) on course grade?  SOCY 126 draws impressively evenly from the full spectrum of the collegiate career.

Figure 3

And the mean course grades, by year in college?  With dazzling, poetic symmetry, they are 87.3, 88.4, 88.4, 87.3.

Let us hammer this point home:

Figure 3