Many important decisions must be made before their consequences are known. A city may need to strengthen a bridge before engineers can determine exactly how long it will remain safe. A farmer may choose what to plant without knowing how much rain will fall. A company may introduce a new product without knowing how customers will respond. In such situations, decision-makers are not completely uninformed. They usually have observations, previous experience, measurements, models, or expert judgments. Yet none of these can guarantee what will happen. This creates a common problem: evidence can reduce uncertainty without eliminating it. The challenge is therefore not simply to obtain information, but to decide what the available information justifies. Consider a town located beside a river. Historical records show that serious floods have occurred there, although not at regular intervals. After several relatively dry decades, local authorities must decide whether to build an expensive flood barrier. Engineers can study rainfall patterns, river levels, soil conditions, and previous floods. They may estimate the possibility of severe flooding during the next several decades. However, they cannot say with certainty whether a major flood will occur next year, in twenty years, or not during that period at all. If officials demand certainty before acting, they may wait forever. On the other hand, uncertainty does not automatically justify immediate action. Building the barrier would use resources that could otherwise support schools, transportation, or health services. The relevant question is not simply, “Will there be a flood?” It is closer to, “Given what we currently know, how serious are the possible consequences, how likely do they appear to be, and what are the costs of acting or not acting?” This illustrates the relationship between uncertainty and risk. Uncertainty refers to limits in what is known. Risk involves possible outcomes and their consequences, often considered together with estimates of how likely those outcomes are. The two concepts are related, but they are not identical. Imagine two maintenance problems in a university building. There is a 40 percent chance that an old projector will stop working during the semester. Replacing it in advance would be inconvenient but inexpensive. Engineers also estimate a 2 percent chance that part of a ceiling structure could fail. That probability is much lower, but the potential harm is far greater. Focusing only on which event is more likely would ignore an essential part of the decision. Evidence helps people estimate such possibilities, but evidence varies in quality and relevance. Ten years of carefully recorded measurements may provide a stronger basis for judgment than a few informal observations. At the same time, more information is not always more useful information. A large collection of data from conditions very different from the present situation may be less informative than a smaller set of directly relevant observations. Evidence can also support more than one reasonable interpretation. Suppose a new teaching method is tested in several schools and most participating students improve their results. This may suggest that the method is effective. Yet questions remain. Would the same improvement occur in schools with different resources? Did students improve because of the method itself, or because teachers received additional training? How long did the improvement last? The study can provide meaningful evidence without answering every question. Problems arise when uncertainty is treated as evidence that nothing is known. If researchers cannot guarantee that a method will work everywhere, it does not follow that their study tells us nothing. The opposite mistake is also possible: a promising result can be presented as if uncertainty has disappeared. The way information is presented can further influence decisions. Imagine that a medical screening program correctly identifies a condition in 90 percent of people who have it. That figure may sound reassuring. But someone evaluating the program would also need to know how often it incorrectly identifies the condition in people who do not have it and how common the condition is. A single accurate percentage may create an incomplete impression when separated from the information needed to interpret it. Framing matters as well. A treatment described as having a 90 percent survival rate may be received differently from one described as having a 10 percent mortality rate, although the numerical relationship is the same. Neither description is false. However, one directs attention toward survival and the other toward death. Responsible communication therefore requires showing both what is reasonably well supported and what remains uncertain. Decision-makers must also consider whether waiting for more evidence has consequences. Additional research can be valuable, but information requires time and resources. If the town postpones its flood barrier while collecting another ten years of environmental data, future estimates may improve, but the town remains exposed during that period. In other situations, delay may be appropriate. If the consequences of waiting are minor and new evidence is likely to change the decision significantly, postponing action may be reasonable. The value of additional information therefore depends partly on what can happen while it is being collected. Different people may also make different decisions while accepting the same evidence. A small business owner considering expansion may accept an estimate that demand has a 60 percent chance of increasing. One owner may consider that sufficient to justify investment, while another may decide that the financial consequences of being wrong are too severe. The disagreement may concern acceptable risk rather than the evidence itself. Evidence, then, does not make decisions by itself. It can describe possibilities, reduce uncertainty, and challenge incorrect assumptions. It cannot always determine how much risk should be accepted or which consequences matter most. Finally, the quality of a decision should not be confused with its outcome. Suppose a rescue team chooses the route that all available information indicates is safer, but an unexpected landslide blocks it. Another route happens to remain open. The unfavorable result does not necessarily show that the original reasoning was poor. Similarly, someone may ignore strong warnings, take an unnecessary risk, and suffer no negative consequences. A fortunate outcome does not automatically make the reasoning sound. Most real decisions exist between complete ignorance and complete certainty. Good decision-making does not require eliminating uncertainty, because that is often impossible. It requires understanding what the evidence supports, recognizing what remains unknown, considering the consequences of different outcomes, and making choices whose reasoning can be defended even when the future turns out differently than expected. # Reading Comprehension Test **Instructions:** Read each question carefully and choose the option that is best supported by the text. ### 1. According to the text, what is the main challenge when people make decisions under uncertainty? A. Finding enough information to predict exactly what will happen B. Determining what conclusions and actions the available evidence reasonably supports C. Avoiding decisions until all important uncertainties have been removed D. Choosing the option that has produced good results in similar situations ### 2. In the discussion of the town and the flood barrier, why does the author replace the question “Will there be a flood?” with a more complex question? A. To show that predictions about natural events are usually unreliable B. To suggest that public authorities should focus mainly on financial costs C. To demonstrate that historical records cannot provide useful evidence about future events D. To include likelihood, consequences, and the costs of different actions in the decision ### 3. A laboratory has a 30% chance of losing one day of work because of a computer failure and a 3% chance of losing several months of research because of a storage-system failure. According to the reasoning in the text, what should the laboratory consider? A. Both the probability and the possible consequences of each failure B. Only the computer failure, because it is more probable C. Neither failure until more precise probabilities become available D. Only the storage failure, because rare events should always receive priority ### 4. What does the author mean by saying that “more information is not always more useful information”? A. Large amounts of data usually make decisions unnecessarily complicated B. Informal observations can sometimes replace systematic measurements C. Relevant evidence may be more informative than a larger amount of evidence from different conditions D. Decision-makers should stop collecting evidence once they have enough data to act ### 5. Researchers find that a workplace training program improves performance in one company, but the employees also received new equipment during the study. Which description best fits this situation according to the text? A. The evidence is useless because more than one explanation is possible B. The evidence is meaningful, but uncertainty remains about what caused the improvement C. The evidence proves that the program works whenever new equipment is provided D. The evidence should be accepted only if the same employees repeat the program ### 6. A coastal community can spend two years collecting better storm data before deciding whether to strengthen its defenses. During those two years, however, existing defenses will remain vulnerable. Which principle from the text applies most directly? A. Historical evidence becomes less reliable whenever conditions may change B. Communities facing serious consequences should always act immediately C. Additional evidence is valuable only when it completely removes uncertainty D. Waiting for better information must be considered together with the consequences of waiting ### 7. How does the example comparing the projector and the ceiling structure develop the distinction between risk and uncertainty? A. It demonstrates that inexpensive problems generally involve more uncertainty than dangerous ones B. It shows that uncertainty disappears when numerical probabilities can be calculated C. It shows that an event with a lower probability may still deserve serious attention because its consequences are greater D. It suggests that high-probability events should be addressed before low-probability events ### 8. A manager receives a large international dataset about consumer behavior and a smaller study of customers in the manager’s own region. The international data were collected under very different economic conditions. Based on the text, what should the manager avoid assuming? A. That the larger dataset is automatically more useful because it contains more observations B. That economic conditions can influence how evidence should be interpreted C. That the smaller study contains any useful information D. That evidence from local customers can ever support a general conclusion ### 9. Why can describing a treatment as having a “90 percent survival rate” rather than a “10 percent mortality rate” affect a decision? A. One description is statistically more accurate than the other B. People can understand positive percentages more easily than negative percentages C. The descriptions provide different probabilities for the same outcome D. The presentation can direct attention toward different aspects of equivalent information ### 10. What function does the medical-screening example serve in the author’s argument? A. It shows that medical decisions involve more uncertainty than decisions in other fields B. It illustrates how an accurate piece of evidence can still create an incomplete impression when important context is omitted C. It proves that percentages should not be used when communicating uncertainty D. It introduces an exception to the claim that relevant evidence improves decisions ### 11. The flood-barrier example appears both when the author discusses uncertainty and when discussing the value of additional information. What relationship between these discussions is most strongly supported? A. A decision must consider both current uncertainty and what may happen while efforts are made to reduce it B. Because uncertainty cannot be eliminated, collecting additional evidence has little practical value C. The main purpose of gathering more information is to determine the exact time at which an event will occur D. Expensive preventive action is justified whenever additional information requires many years to collect ### 12. Two business owners accept the same estimate that demand has a 60% chance of increasing but choose differently about expansion. What broader implication does the text draw from situations of this kind? A. Numerical estimates cannot contribute meaningfully to practical decisions B. People who accept the same evidence should normally reach the same decision unless one misunderstands it C. Evidence becomes unreliable when people disagree about what action to take D. Decisions can differ because evidence informs choices without determining how consequences and acceptable risk should be valued ### 13. A public agency reports that a new safety program reduced accidents in several participating workplaces. It presents the reduction as strong evidence for expanding the program nationwide. However, the participating workplaces also received additional staff training, the study did not include workplaces operating under substantially different conditions, and the agency’s public report emphasizes the reduction without discussing these limitations. Which evaluation best integrates the author’s treatment of evidence, uncertainty, and presentation? A. The reported reduction should be rejected because evidence with unresolved alternative explanations cannot contribute to a decision. B. The result provides relevant evidence, but its causal interpretation, applicability to other settings, and presentation should all be examined before it supports a broader decision. C. The result is sufficient for nationwide expansion because uncertainty about causes matters less once an improvement has been measured. D. The main weakness is the way the result was presented; if the limitations were communicated clearly, the evidence would establish that the program itself caused the reduction. ### 14. A transportation agency must choose whether to repair a tunnel now. Current inspections indicate a small possibility of a serious structural problem. A more precise inspection will be available in six months, but the tunnel must remain in use until then. Which approach is most consistent with the criteria developed in the text? A. Repair the tunnel immediately because severe possible consequences always justify preventive action. B. Ignore the estimated probability and decide according to whether previous tunnel repairs were successful. C. Compare the present evidence, possible consequences, costs of action, risks during the waiting period, and likely value of the additional inspection. D. Wait for the new inspection because decisions should use the most precise evidence obtainable. ### 15. A committee chose Option X because the available evidence indicated that it had a lower probability of serious harm than Option Y. Option X nevertheless produced a serious loss, while another organization chose Option Y and experienced no loss. A reviewer concludes that the second organization therefore made the better decision. Which evaluation is most consistent with the text as a whole? A. The conclusion is justified because actual outcomes provide stronger evidence about decision quality than estimates made beforehand. B. The conclusion is justified if both organizations had access to the same numerical probabilities, because identical evidence should produce identical choices. C. The conclusion cannot be assessed unless it is known whether either organization collected additional information after making its decision. D. The conclusion is not established by the outcomes alone; the decisions should be assessed using the evidence, uncertainty, possible consequences, and reasoning available when each choice was made. [[Respuestas del examen de comprensión - Inglés]]