Term Paper

FEEDBACK ON YOUR ANALYSIS TASKS SO COMMON I DIDN’T TRY TO WRITE IT ON EVERY PAPER:

  • Always report p-values. This means you’ll want to do at least some regressions, since that is the easiest way to get p-values. Report the raw p-values, then do the Benjamini-Hochberg FDR adjustment for multiple hypothesis testing if appropriate. (It is confusing if you don’t also report the raw p-values.)

  • Always give the full details of the wordings of the questions and the response options. You can always get these by doing the survey again, but when I am doing that sort of thing I don’t give real answers to the questions, I just click anything until I get to the questions I wanted to look at.

  • I said I love scatterplots, but there is an exception: when one of the two variables has only a few possible values, box plots for the other variable given each of those few possible values are a better way to show the relationship. Note that box plots are a lot like bar charts—but they have more total information in them than bar charts.

  • If you have income in the regression, always use log(income). When it is income ranges (bins) that people say, you should use log(midpoint of the range) as log(income). Using non-logged income is almost guaranteed to get you weird results. And using bin number gives a coefficient that is hard to interpret.

  • If you have log(income) in a regression (and I think it will be household income—all the adult incomes should be counted), I highly recommend using putting log(household size) in as another variable in the regression. That makes sure that you are accounting for a given amount of income being spread over more people while being fairly agnostic about economies of scale in the household.

  • Make sure to discuss causality and to discuss causality in the context of your particular analysis, not just in general terms. What are the likely biases? What is their likely direction? What are some things that are possible but that you don’t think are issues in your particular case?

  • Carefully use non-causal words where you aren’t actually claiming causality. You don’t want to use causal words until you are really ready to discuss causality.

Due by 11 PM Wednesday, April 30, 2025.

Your term paper should incorporate a revised version of what you wrote for your Analysis Task and be 4-6 pages longer than that revision of what you wrote for your Analysis Task. That is, the rest of your paper beyond what you wrote for the Analysis Task should be 4-6 pages.

Here are some of the main things I’ll look for in your term paper:

  1. Write well: have a thesis statement and a theme that you follow through

  2. Critique a paper in the academic literature on topics related to things we have discussed in class. Don’t choose one of my papers. But the references lists in my papers on happiness and of the other papers listed below are good places to find a paper.

  3. (Drilling down on point 2.) Assume that the paper has flawed statistical interpretation. (In particular, results are typically overinterpreted to make them sound more exciting.) Do better on the statistical interpretation front. Use all of the things you learned about statistical interpretation in class that are relevant. (I’ll subtract points if a statistical interpretation principle was clearly relevant to what you are critiquing or to your own analysis and you don’t discuss it and add points if you do a great job discussing an issue.) The goal is not to solve everything but to show your awareness of the issues and do what can readily be done to think about what the issue implies. (For example, it is great if you can say which direction a bias is.)

  4. Weave in a revision of your analysis task. The integration of this with your critique of a paper in the literature doesn’t have to be perfect, but it is a plus if you can make the critique of a paper in the literature and your own statistical analysis fit together with a theme.

  5. Be timely: because of the exigencies of making grades for this and my other course, it is especially important that I get these term papers by the due date. However, I will still accept them, but with some points off, up to two days late.

The main idea for the term paper is to discuss an academic journal article on well-being skeptically. Chat GPT and other AI are great ways to find academic articles related to the analysis you did, though if you have real trouble finding something closely related, more distantly related is fine.

When I ask ChatGPT for a book or article on the topic, I use a prompt like this: Please find 10 articles on [TOPIC}. You often hallucinate articles and books that don’t exist, so please check online to make sure each one of them really exists and give me a link for each.

How to be skeptical. Here are some ways you might want to be skeptical (and there are many others):

  1. Scale-use differences might be creating an illusion. (How?)

  2. There is likely to be statistical bias relative to what the author or authors seem to be claiming or implying. (Make sure to explain which direction you think any story of possible statistical bias would bias things. Is the estimate in the paper likely to be higher than the truth or lower than the truth? What does that say about the truth?)

  3. A result that has a nominal p-value of 5% (t-statistic of 2 or so) really has about a 50% chance of being spurious, as indicated by replication studies. (By contrast, a result with a nominal p-value of 1/2 % (t-statistic of 3 or more) has only about a 5% chance of being spurious, and so is relatively trustworthy. Here though, “trustworthy” had to be taken in a narrow sense. Something is probably going on with that coefficient, but what is going on may be very different from what the authors claim. (See for example the rest of this list of reasons to be skeptical!)

  4. Happiness is not the same as utility. As my coauthored papers “Utility and Happiness,” “What Do You Think Would Make You Happier? What Do You Think You Would Choose?” “Can Marginal Rates of Substitution Be Inferred from Happiness Data? Evidence from Residency Choices” and “Beyond Happiness and Satisfaction: Toward Well-Being Indices Based on Stated Preference” suggest, there are many distinct meaning of happiness. Among them are:

    • utility

    • feeling happy

    • what people report on a survey about their happiness, life satisfaction or position on the ladder of life (all of which have a lot of data available)

    • Aristotelian noble happiness, often called “eudaimonia” or “eudaemonic well-being” in the literature

  5. Other theoretical issues

In Addition to Getting Help from AI, There Are a Lot of References in Your Textbook to Papers that Would Be Good to Critique. And Here are Some Papers with References Lists in which You Can Find a Paper to Critique (Note: Don’t Critique a Paper with Miles as a Coauthor—those papers are only included because they have highly relevant references lists):