Questions: The Role of Values in Science

Question 1

A pharmaceutical study's researchers set a significance threshold that prioritizes avoiding false negatives (missing a dangerous drug) over avoiding false positives (wrongly flagging a safe drug). Is this a legitimate or illegitimate role for values in science?

• A. Illegitimate — any non-epistemic influence on statistical thresholds compromises scientific objectivity
• B. Legitimate — choosing acceptable error rates based on the real-world consequences of different kinds of mistakes is exactly the indirect role that values are permitted to play
• C. Illegitimate — significance thresholds are matters of statistical convention and must not vary by context
• D. Legitimate — but only because this is an epistemic value (accuracy), not a non-epistemic one

Answer: (B)

Heather Douglas's distinction between direct and indirect roles for values is decisive here. A direct role — where values determine whether evidence supports a hypothesis — is illegitimate: it overrides the evidence. An indirect role — where values shape what level of risk of each error type is acceptable — is legitimate and unavoidable. In drug safety testing, the asymmetry between the harms of false negatives (missed dangers) and false positives (wrongly rejected safe drugs) is a value judgment that appropriately influences the threshold. This is not 'compromising objectivity' — it is making explicit a value commitment that is otherwise hidden.

Question 2

What is the key distinction Heather Douglas draws about how values may and may not legitimately enter scientific reasoning?

• A. Values may influence which questions get studied but not how data is analyzed
• B. Values may be used by individual scientists but not by scientific institutions or journals
• C. Values may play an indirect role (shaping acceptable error tolerance) but not a direct role (determining whether evidence supports a hypothesis)
• D. Epistemic values may influence theory choice but non-epistemic values may not

Answer: (C)

Douglas's distinction is between direct and indirect roles. The direct role — values determining whether evidence counts as support for a hypothesis — is illegitimate because it allows non-evidential factors to override the evidence. The indirect role — values shaping how much risk of each kind of error is acceptable — is legitimate because it governs the decision context, not the evidence itself. The distinction matters practically: a researcher who ignores contradictory evidence for political reasons (direct) is acting badly; a researcher who sets a high bar for safety conclusions because the stakes of being wrong are severe (indirect) is acting responsibly.

Question 3

Recognizing that non-epistemic values shape scientific practice implies that science is merely another form of politics, and that one value-laden account is as epistemically good as any other.

• T. True
• F. False

Answer: False

This is the relativist conclusion that the value-in-science literature explicitly argues against. Science's intersubjective methods — replication, peer review, empirical testing, the self-correcting community — give it genuine epistemic authority that pure political opinion lacks. Acknowledging that values influence funding priorities, theory choice, and error thresholds does not erase this authority. The correct conclusion is the opposite of relativism: because values inevitably enter scientific practice, transparency about value commitments is a prerequisite for doing more honest, more accountable science — not a concession that science is arbitrary.

Question 4

Underdetermination of theory by evidence is relevant to the question of values in science because it creates gaps where background assumptions — which reflect social values — can influence theory selection without overriding the data.

• T. True
• F. False

Answer: True

Underdetermination means that the available evidence is consistent with more than one theory. In those gaps, scientists must choose — and they typically do so using background assumptions about what kinds of explanations are plausible, what prior work is reliable, and what counts as a satisfying solution. Helen Longino argues these background assumptions are not value-neutral; they reflect social and contextual values embedded in the scientific community. This is not a failure of science — it is an unavoidable feature of evidential reasoning in complex domains. Making those assumptions explicit is what responsible science demands.

Question 5

What is the difference between epistemic values and non-epistemic values in science, and why does the distinction matter for evaluating scientific objectivity?

Model answer: Epistemic values are internal to science's core aim of producing reliable knowledge: simplicity, accuracy, empirical adequacy, consistency, and explanatory breadth. Preferring simpler or more accurate theories is not a bias — it is constitutive of good scientific reasoning. Non-epistemic values (moral, political, social, economic) come from outside this core. The distinction matters because the ideal of value-free science specifically targets non-epistemic values: science should not let political preferences determine whether evidence supports a hypothesis. But recognizing that non-epistemic values inevitably enter science — through funding priorities, framing choices, acceptable error rates — shows that the goal of value-free science must be qualified and made more precise, not abandoned.

The distinction also explains why value-laden science is not automatically bad science. A study that explicitly acknowledges its error-tolerance choices, its funding sources, and the value judgments embedded in its framing is doing better science than one that claims false neutrality. The philosophy of science has moved from 'science should be value-free' to 'science should be transparent about its value commitments' — a more honest and achievable ideal.