Analyse where values enter social inquiry, distinguishing epistemic from non-epistemic values and assessing the threat of bias to social-scientific knowledge

What this dot point is asking

SEAB wants you to analyse the role of values in social inquiry: to identify where the inquirer's values enter the research process, to distinguish the values that belong in good inquiry from those that threaten bias, and to assess whether value influence necessarily undermines social-scientific knowledge. This builds on the objectivity debate and connects to the fact-value distinction in ethics. Your task is to map the entry points of values and to judge which are legitimate.

The answer

Where values enter inquiry

Values enter social research at several distinct stages, and conflating them causes confusion. They enter in the choice of topic (what is judged worth studying), the framing of the research question, the choice of concepts and measures (often value-laden, like welfare or deviance), the conduct of the research (ethical limits on how subjects may be treated), the interpretation of data, and the use to which findings are put. Some of these involvements are clearly legitimate and unavoidable; the controversial cases concern the interpretation of data and the acceptance of conclusions.

Epistemic and non-epistemic values

The key distinction is between epistemic and non-epistemic values. Epistemic values are those internal to the pursuit of knowledge: accuracy, empirical adequacy, breadth of scope, simplicity, internal consistency, fruitfulness, and predictive success. These legitimately and unavoidably guide theory choice and inference, because they are part of what makes a belief well grounded. Non-epistemic values are moral, political, social or personal commitments not internal to the search for truth, such as a preference for a particular policy outcome. These threaten bias when they determine conclusions, because then findings track the researcher's preferences rather than the evidence.

The threat of bias

The genuine danger is that non-epistemic values shape what evidence is gathered and how it is read, so that the conclusion merely confirms what the researcher wanted to find. This is bias, and it undermines objectivity in the epistemic sense. The history of social science offers cases where prejudice masqueraded as finding, which is why the threat must be taken seriously rather than waved away. The question is whether this threat is escapable.

The argument from inductive risk

A subtle complication shows values cannot be confined to the choice of topic. Because accepting or rejecting a hypothesis carries practical consequences, deciding how strong the evidence must be before accepting a claim involves a value judgement about the costs of error. Wrongly declaring a substance safe (a false negative on harm) and wrongly declaring it dangerous (a false positive) have different costs, so how much evidence counts as enough depends on what is at stake. This argument from inductive risk shows that non-epistemic values legitimately enter even the acceptance of hypotheses, not merely the framing of questions.

Managing value influence

Granting all this, value influence need not undermine knowledge, provided it is managed. Three safeguards do the work. Transparency: researchers declare their value commitments so others can assess possible bias. Methodological safeguards: blinding, controls, random sampling and pre-registration of hypotheses limit the play of preference. Pluralism: a community of researchers with diverse values subjects each other's work to criticism, so individual biases tend to cancel rather than accumulate. The defensible conclusion is that values necessarily enter social inquiry but do not necessarily undermine its objectivity, as long as epistemic values govern inference and non-epistemic values are made explicit and checked.

Examples in context

Example 1. Setting the bar for a safety claim. A regulator deciding whether the evidence is strong enough to declare a chemical safe must weigh the cost of a false reassurance against the cost of a needless ban. This is a value judgement about the costs of error, and it legitimately affects how much evidence is demanded. The case illustrates the argument from inductive risk: even the acceptance of a factual hypothesis can properly involve non-epistemic values, made explicit.

Example 2. Pluralism cancelling bias. A contested question in economics is studied by researchers across the political spectrum, each scrutinising the others' assumptions and methods. Where one researcher's preference might skew an interpretation, critics with different commitments expose it. The collective result is more objective than any single value-laden study, illustrating how a plurality of researchers manages bias without pretending individual inquiry is value-free.

Try this

Q1. List four stages at which values enter social inquiry. [6 marks]

• Cue. Any four of: choice of topic, framing the question, choice of concepts and measures, research ethics, interpretation of data, and use of findings.

Q2. Explain the argument from inductive risk and what it shows about values in inquiry. [8 marks]

• Cue. Accepting or rejecting a hypothesis has practical costs, so deciding how much evidence is enough weighs the costs of error, a value judgement; this shows non-epistemic values legitimately enter even the acceptance of hypotheses, not just topic choice.

Q3. State two safeguards that help prevent values from biasing social-scientific conclusions. [6 marks]

• Cue. Any two of: transparency about value commitments, methodological safeguards (blinding, controls, pre-registration), and a plurality of researchers whose mutual criticism cancels individual bias.