Study Guide

An Epidemic of Uncertainty: Navigating HIV and Young Adulthood in Malawi

Note: This study guide was developed jointly by Jenny Trinitapoli, Emily Williams, and Stephanie Achugamonu with the intention of making it easy for instructors to assign the book (or portions of it) in graduate or undergraduate courses. For each chapter, we provide three resources: comprehension questions designed to gauge how well students understood the reading, discussion questions designed to provoke open-ended conversation and debate in a classroom, and exercises that could be completed outside of class, as graded or ungraded assignments. We hope instructors take what is useful from these suggested questions and prompts. If you are using An Epidemic of Uncertainty in class, please write to share what worked well (and what didn’t!) so we can keep updating this guide.

Chapter 1: Introduction: Surveying the Shadows of Uncertainty

Comprehension Questions:

  1. How does the author characterize the relationship between HIV and fertility? How is this characterization supported (with data)?

  2. Define the Thomas Theorem and explain how it relates to HIV.

  3. What are the three core concerns of demography?

  4. Explain “vernacular theory” as applied in this chapter.

Discussion Questions:

  1. This book will explore how uncertainty impacts relationship and fertility patterns amid an HIV epidemic in sub-Saharan Africa. What other contexts do you know of where uncertainty may be playing an important role in demographic processes?

  2. Consider the quote from Veronique Petit that “data are only meaningful if demographers use appropriate theoretical resources to provide an interpretive model.” Do you agree? How do researchers balance theory and evidence when constructing an interpretive model?

  3. What demographic processes do you see unfolding in your own community? What are the most common causes of death? Are young adults getting married earlier or later than their parents? Having more children? Fewer? Where do you think your perceptions come from?

  4. What are models good for in demographic analyses?


  • For next week, look for examples of the Thomas Theorem at work in your own social circles. Write a fieldnote-style journal entry about the consequences of a perception, regardless of whether the perception is accurate.

  • Pick one of the four demographic concepts listed on page xxxx with which you are not already familiar. Using scholarly resources (library databases), what can you find out about the origins and development of this concept? Who advanced it and why? (What puzzle were they trying to solve?) Has it been used or supported in further studies? Has it been challenged, and by whom? (Use backward and forward citations to explore.)

  • Find three people who are not from your immediate social circle (grew up in a different town or state, different age, different socioeconomic background) and ask them each five of the following questions (does not have to be the same five for each person). Jot notes about their answers; then write a brief summary (3-4 paragraphs) summarizing what you learned.

  1. What do you think are the top three reasons people get married?
  2. What do you think are the top three reasons people have children?
  3. What do you think are the top three reasons people get divorced?
  4. Do you, yourself, ever worry about HIV or other STIs?
  5. What do you notice about marriage patterns in your social circles? Are they changing these days?
  6. Do you have any concerns about population change?
  7. Do your friends want children?
  8. What would your family think if you got married/divorced next week?
  9. Try to guess three of the top five causes of death in the United States today among young adults ages 18-30.
  10. I know it’s a heavy thought, but if you were to die within the next year, what do you think the cause of death would be?

Chapter 2: Ten Years in Balaka: The Excellent and Imperfect Data of Longitudinal Studies

Comprehension Questions:

  1. What gap in the knowledge base was the TLT study designed to fill?

  2. How did the TLT cohort experience the HIV epidemic and the simultaneous political landscape in Malawi? Why is this age-group’s experience particularly interesting?

  3. Why do so many Malawians see HIV and democracy as “part of the same shift?”

Discussion Questions:

  1. (Instructor: Provide population pyramids from Malawi, Bangladesh, U.S., Italy, and Japan from 2020 OR have students create these using the amazing INED tool.) Compare and contrast these population pyramids. Describe the age structure and contrast these distinct realities. Are these populations likely to face the same sources of uncertainty? Why/why not?

  2. Consider the last time you moved or visited a new place, near or far. What cultural differences (large or small) did you notice? What specific interactions drew your attention to these differences? And how did you notice it as part of a pattern? What questions would you have liked to ask someone who had lived in the area for their whole life? To another newcomer?

  3. This chapter describes several different policies that organize HIV testing and treatment, neonatal care, and public education in Malawi. What does this policy backdrop reveal about the country’s priorities? Their constraints? On what other basis might goods be distributed in poor countries? What about in rich ones like the US? What gaps in policy makers’ minds can you identify in this narrative?


  • Keep a journal like Gertrude’s. For 1-2 weeks, record your everyday conversations around a chosen topic (politics, family planning, or health care). You can include online chatter as well - posts written or shared by people you know personally, ideally those that generate conversation.

  • Listen for exchanges that represent population chatter about demographic processes. Record at least 10 such exchanges, and explain what is happening in the exchange that marks it as part of a demographic process.

  • Suppose you wanted to know how online dating has impacted young adults’ perception on relationships and marriage in England. Without budget constraints, design a study that would allow you to investigate this question. Ethical standards apply. How would you recruit respondents? What questions would you ask? How would you conduct your surveys (in what setting)? How often would you survey your respondents? For how long would you follow them? (Bonus: What imperfections would you predict arising in your data? How could you prevent or address them? Who would fund such a study and pay you to execute it?)

  • Using Figure 2.3 as inspiration, draw a Lexis-style figure with a lifeline for your birth cohort and lifelines for your youngest parent and grandparent. Drop vertical lines to indicate the most salient political events that occurred during your lifetimes (label these at the top) and more lines to represent major personal events that influenced your own life (label these at the bottom). How would you narrate the most important intersections in your personal Lexis diagram? Write 1-2 paragraphs summarizing.

Chapter 3: Uncertainty Demography

Comprehension Questions

  1. In your own words, explain the difference between an independent variable and a dependent variable. What does the author mean by micro and macro as described in Figure 3.1? Relate these to the distinction between personal and aggregate experiences of uncertainty.

  2. Contrary to other scholars, Trinitapoli argues that uncertainty is measurable. What does this mean, exactly? And what does it mean for social scientists to find ``close proxies for uncertainty” in their datasets. Extending this to a discussion question: what unique ideas do you have about measures that might serve as good proxies for uncertainty, overlooked here?

  3. The author discusses several papers that leverage variance in their attempt to measure uncertainty. What is variance in the statistical sense? How does it relate to a mean/average? And why does the author believe that variance is often a good proxy for uncertainty?

Discussion Questions

  1. What are the most profound uncertainties in your society today? What kinds of uncertainties dominated your childhood (approx. ages 10-15)? Have any of these been resolved in the interim?

  2. Point 10 that follows Trinitapoli’s definition of uncertainty claims that “uncertainty can be fruitful – a productive condition that drives innovation and creativity.” In pairs or small groups, discuss this claim. Who agrees? Disagress? What examples can you give of the fruitful side of uncertainty? What would happen to our live if it were eliminated completely?

  3. According to Point 8 that follows the definition: How can one rigorously distinguish one historical time period as being more or less uncertain than another. Who decides and how? Are these claims well supported? Or simply asserted?

  4. How does time factor into the very concept of uncertainty, both in the material covered in this chapter and in your own life experience?


  • Before reading the chapter, write a definition of uncertainty based on your current understanding of the concept. Try to be precise, but write your definition using your own words and plain language. After reading the chapter, revise your definition to improve what you started with. What do you love about your definition? And what, if anything, is lacking?

  • Book production takes time. Given a publication date in late 2023, the author probably finished revising this chapter sometime in 2022. Using library resources (i.e., PubMed, Web of Science, other scholarly databases), do a keyword search for uncertainty + demography or uncertainty + fertility (or some other permutation) to locate academic articles related to uncertainty and demography published since the author would have written this chapter. Choose 2 articles to read for mastery; write a rhetorical precis and include a proper bibliographic citation. In addition, be sure to mention where each paper would fit in the 2x2 table of micro vs. macro, dependent vs. independent variable. How do the findings in each paper you chose square with the key papers discussed in Chapter 3?

  • With a partner, use your laptops to seach “Uncertain Times” in your favorite search engine. Spend about 10 minutes browsing the variety of pages that come up through such a search. Discuss the use and misuse of the term on the internet. Has the phrase been cliched beyond repair? From your browsing, which page uses the phrase with the most intellectual integrity? Which the least?

Chapter 4: The Scope of HIV Uncertainty

Comprehension Questions

  1. What is the value of the beans method? How does it improve on simply asking respondents to answer questions about their perceptions of risk either verbally or in writing/through multiple choice?

  2. Describe in your own words the distributions depicted in Figure 4.1 and how the perceived likelihood of HIV infection shifts over the three time horizons.

  3. How does the presence of HIV uncertainty impact fertility decisions and health outcomes in this sample? How was this measured in the study?

  4. Compare and contrast the hazard models in Figures 4.4 and 4.5. What do these models represent? What is important about the differences between the two figures?

Discussion Questions

  1. In the beginning of this chapter, the author claims that focusing on serostatus alone will mislead researchers who are trying to understand human behavior. Why does she say this? Do you agree? If you’re a health care worker, a patient’s serostatus is an important piece of information for you to have. Why? Now imagine you are a researcher studying human behavior; under what conditions would you prioritize perceptions over serostatus if your goals was predicting future behavior?

  2. Several of the results shared in this chapter are presented separately for men and women. How does gender factor into the study design? What, if anything, surprised you about the gender differences documented so far in the TLT study? How do questions about HIV and uncertainty apply differently to men and women?

  3. Consider the kinds of uncertainties you experience in your own life. Might any of these uncertainties be influencing your own health? Shaping major life decisions? Have you ever noticed these relationships before?

  4. In Figure 4.2, the author “bins” the beans responses to create analytic groups: infected, vulnerable, uncertain, acutely uncertain, and insulated. Would you create any other groupings out of the pyramid in Figure 4.2? If so, how and how would you define it? And if no additions, why not? What makes this representation comprensive?


  • Design a few beans questions of your own. Write 5-10 questions that could be asked on a scale of 0-10 where respondents answer by assessing the likelihood of an event occurring on a particular time horizon. Break into groups of 2-4 and take turns interviewing your classmates. Extra challenge: organize the questions as a survey researcher would, beginning with casual questions and gradually broaching the more sensitive topics. How would you engineer the conversation to put respondents at ease?

  • Trinitapoli and Yeatman explain that HIV uncertainty in Malawi is associated with early or accelerated childbearing. Was this finding expected or surprising to you? In small groups, research examples of other cases of uncertainty related to human survival and well-being (ex. economic crises, political conflict, famine, etc.) and try to determine whether this pattern of accelerated childrearing is specific to HIV and Malawi or something broader. Can you find cases where uncertainty operates differently, perhaps delaying child bearing?

Chapter 5: HIV Uncertainty & the Limits of Testing

Comprehension Questions

  1. Describe the Health Belief Model. What is it, how has it been used in previous research, and how does the model stand up against the data presented in this chapter?

  2. How has the “culture of testing” in Balaka evolved since the diagnostic technologies scientists and clinicians had to rely on in the 1990s?

  3. Which variables are most strongly correlated with HIV-related uncertainty? Are they the same or different compared to predictors of experiencing uncertainty immediately following testing?

  4. Compare and contrast the aggregate view of HIV-related uncertainty displayed in Figure 5.3 to the within-person change depicted in Figure 5.4. How do each of these figures frame the data? What question does each figure answer?

  5. Explain in your own words what is going on in Figure 5.5.

Discussion Questions

  1. Consider other conditions that cannot easily be detected or confirmed without some kind of test - early pregnancy, malaria, cancer, etc. Do you believe testing for these conditions is similarly (in)effective at resolving uncertainty? What factors might change the relationship between testing and uncertainty for each condition?

  2. Based on the evidence presented in this chapter, what advice would you give to the District Health Officer in Balaka if she told you she wanted to reduce HIV uncertainty (and its negative consequences) in her district?

  3. Consider the “View from the Borehole” interlude. If you had been present during this conversation, what follow-up questions would you have asked the speaker? After reading the story, what are you left feeling curious about or inclined to investigate further? How does the content of this conversation - what is shared, what is asked, and what is not - illuminate distinct priorities and patterns in the collective thinking of this community? What does the ethnographic view add to what was already established by the three survey-based views offered earlier?

  4. Are there any potential predictors of HIV-related uncertainty might the author have overlooked? What additional correlates of HIV-related uncertainty would you find interesting to investigate?

  5. Consider the top three sources of major uncertainty in your life. Is there a “test” or any kind of intervention that would eliminate or even abate these uncertainties?


  • Imagine you are working for a communications team designed to promote population health. Design a “Know Your Status” poster or cartoon for a condition other than HIV. Explain why you chose the visuals and slogan that you did. How would you disseminate this campaign in your [a relevant] community?

  • (In pairs or small groups) Design a mini-study. Come up with a question that you could ask twice (before and after) and an intervention that would provide each respondent with new information or a chance to reinterpret their answer. Example: “How much you weigh?” before and after measureing weight with a scale. Analyze and interpret the results. Did the new information change your respondents’ answers?

  • Take some inspiration from Kelvin Maigwa: Sketch a political cartoon that fits the themes of testing, uncertainty, and population-related phenomena. What makes your cartoon funny? Provocative? Complicated?

Chapter 6: Relationship Uncertainty & Marriage Instability

Comprehension Questions

  1. How is “marital experience” defined in this chapter? How is HIV patterned by marital experience in this population?

  2. What evidence in this chapter supports the assertion that relationship uncertainty is prevalent in Balaka?

  3. Describe the process of “marital shopping” and its implications for the spread of HIV.

Discussion Questions

  1. Examine figures 6.4 and 6.5 in pairs and discuss any aspects of these graphs that leave you with unanswered questions.

  2. Given what this chapter demonstrates about the levels of trust men and women have in their partners, what gender norms would you expect to observe in Balaka? What behaviors or attitudes would you predict? Are there any behaviors you have read about in EOU or know about that contradict the data presented in this chapter?

  3. If you were a young adult navigating relationships in a community patterned similarly to Balaka, what bundle of strategies would you rely on to achieve relationship satisfaction while minimizing your risk of contracting HIV?

  4. What is the significance of Patrick’s story in contrast to Njemile’s?

  5. Consider the discussion of the RBF4MNH policy in “Population Chatter at the Salon.” What do these perceptions of public policy tell us about how patients, healthcare workers, and policymakers understand (or not) one another?


  • This is likely to have been your first exposure to the idea of decomposition analyses. Without responsibility for conducting any analyses yourself, get into pairs and try to come up with 2-3 examples of social change that may be driven by compositional effects, rather than changing rates. Then, using library resources, do a search for scientific articles to determine if any scholars have used decomposition analyses to answer the question. What did they find?

  • TLT respondents were asked if they agreed or disagreed with the statement “These days, most married men are faithful to their wives.” They were also asked if they thought their current partner had other partners on the side at any point in the relationship and given 4 possible responses to choose from — “Yeah, I know s/he does”, “I suspect so,” “One can never know what another person does”, and “Probably not”. Ask these two questions to your classmates and summarize your findings. Is there a general consensus of unfaithfulness similar to the one reached by TLT respondents? How can this perception of men affect the quality and length of romantic relationships?

Chapter 7: Call the Ankhoswe

Comprehension Questions

  1. Who are the ankhoswe and what role do they play in marriage stability in Malawi?

  2. Why did the TLT researchers ask respondents about their best friends’ extramarital relationships and not just their own? Locate the 2011 paper by Yeatman and Trinitapoli to learn a little more about the merits of this approach. What are some critiques?

  3. How has the proliferation of mobile phones changed patterns of relationship mediation in Balaka?

  4. What factors does the author show to be strongly associated with exacerbated relationship uncertainty?

  5. What steps constitute the marriage process in Balaka? What is significant about which steps couples complete/pursue? How does the marriage process tend to unfold in your family’s culture?

Discussion Questions

  1. Explain the conceptual link between a person’s own uncertainty about their HIV status to their knowledge of a friend’s sexual behavior.

  2. Who are the salient others in your own relationships? Are you a salient other for anyone else in your life? Do you feel control over who plays this role in your life or for whom you take it on?

  3. Consider figure 7.2. What trends do you notice in the TLT data over time? What might you ask or theorize about these trends?

  4. If HIV were eliminated in this population, would you expect relationship uncertainty and marriage stability to follow the same patterns into the future? How might they be different?


  • For next week, take notes on any comments, messages, or suggestions you encounter that relate to decisions about family formation: does your family say anything directly to you about what they expect or hope for? Do you see the shadow of the future looming over discussions with friends in explicit or implied ways? Do you discuss a potential/future family with friends, or come across ads or billboards encouraging certain behavior or outcome?

  • The author conveys the challenges of generating accurate estimates of sexual behavior from self-reports in surveys. To obtain more accurate estimates about the prevalence of concurrent partners in Malawi, Trinitapoli and Yeatman asked TLT respondents to report on their best friend’s relationships. What other strategies have researchers utilized to get better estimates for research topics that are underreported, such as sexual behavior and drug use? Find 2-3 research articles that discuss these strategies and share your findings in small groups.

Chapter 8: Ultimate Uncertainties and the Mortality Landscape

Comprehension Questions

  1. What is noteworthy about the different ranges of the y-axes in figures 8.3 and 8.4?

  2. How are HIV status uncertainty and perceived mortality related in the TLT study?

  3. Explain how figure 8.6 challenges the traditional demographic approach to views of HIV risk.

Discussion Questions

  1. How does your own perception of the mortality landscape compare to the experience of the cohort of young women in Balaka elaborated in this chapter? Consider the frequency of funerals in your life, the causes of death that concern you, and how risks are communicated and mitigated in your world – both by the people you interact with directly in your community and through the more anonymous external sources such as media advertisements and government messaging.

  2. Given the recent changes to the mortality patterns in Malawi, do you think this uncertainty around the mortality landscape will persist? For how long? What might it take for perceptions of mortality to stabilize? How long does it take for the implications of a changing mortality landscape to be perceived and the resulting impacts on other institutions to be respond in turn?

  3. Discuss the “false equivalences” drawn in order to compare incidence and perceived risk across HIV, malaria, and food insecurity. What are the challenges in comparing these three maladies? Are you convinced that the author’s attempt to standardize the experiences and measure all three at once is useful for clarifying the perceptions of HIV and uncertainty in this population? Why or why not?


  • Using data from the World Population Prospects database, create death distribution graphs from life tables for three other countries in/around 2020. Analyze these in comparison to figures 8.3 and 8.4. Come up with 5 empirical questions about demographic patterns in the countries you chose drawn from these comparisons.

  • Since the onset of COVID-19, gains and losses in life expectancy have been a topic of renewed interest in the popular press. Find two articles in the popular press that discuss changes to life expectancy, healthy life expectancy, or years lived with wellbeing. Skim them with the purpose of analyzing the fit between argument and evidence. Is life expectancy the right metric for supporting the argument? What other dimensions of the mortality landscape might the reporter have productively considered or explored?