This chapter teaches students how to identify, understand, and evaluate different types of evidence. Students will learn that evidence is not a single thing — it is a category that includes facts, examples, statistics, experiments, observations, documents, testimony, expert opinion, visual materials, artifacts, and data sets. They will learn that different research fields rely on different kinds of evidence, and that the strength of any evidence depends on how well it fits the research question it is meant to answer.

By the end of this chapter, students should be able to recognize evidence in a source, name the type of evidence being offered, assess whether that evidence fits the claim being made, and explain why evidence that is strong in one context may be weak or irrelevant in another.

Every piece of research writing, every academic argument, every journalistic report, and every persuasive message involves at least two things: a claim and evidence.

A claim is a statement that something is true, that something happened, that one thing caused another, that a trend exists, or that a policy works. Claims can be descriptive, analytical, interpretive, or argumentative. Evidence is the information offered to support a claim. It shows that the claim is grounded in something beyond the claimant's opinion or desire, and gives readers a basis for evaluating whether the claim is worth accepting.

Research is the work of connecting evidence to claims responsibly — finding evidence that genuinely supports the claim, representing it accurately, acknowledging its limits, and being honest about what remains unknown.

The evidence-claim match. Even genuine evidence can be misused if it does not actually support the claim attached to it. A study showing that students who eat breakfast perform better on standardized tests is real evidence — but it does not, by itself, prove that a school's free breakfast program caused the improvement. Other factors may explain the pattern. The evidence is genuine; the claim is too strong for what the evidence shows.

Different academic fields have developed different standards for what counts as good evidence, how it should be collected, and how it should be interpreted. These differences are not arbitrary — they reflect what kinds of questions each discipline asks and what methods are most reliable for producing trustworthy answers.

In the natural sciences, a single study rarely settles a question. Findings accumulate through multiple independent studies using varied methods. In the social sciences, full experimental control over human populations is often impossible, so researchers rely on surveys, quasi-experiments, and statistical controls while acknowledging the limits those methods carry. In the humanities, evidence is typically interpreted rather than measured — two scholars reading the same document may reach different but both legitimate conclusions, as long as their readings are grounded in the text and responsive to alternative interpretations. In applied fields such as medicine, law, and public policy, evidence standards from multiple disciplines combine, and the consequences of acting on weak evidence can be serious.

Knowing which discipline you are working in tells you what kind of evidence to look for, what methodological language to expect, how to judge whether evidence fits the claims made, and how to present evidence in your own writing.

A fact is a specific, verifiable piece of information confirmed through reliable sources — dates, measurements, documented occurrences, counts. Facts anchor claims in verifiable reality, but they require context to be meaningful. "The United States incarcerates more people per capita than any other country" is a fact, but understanding what it means requires knowing how incarceration is measured and counted, and what policy factors contributed to the figure. Facts without context can be accurate and still mislead.

An example is a specific instance used to illustrate a claim. Examples make abstract claims concrete, but a single example cannot establish a pattern, a typical case, or a cause. The most common misuse of examples is cherry-picking — selecting only examples that support a desired conclusion while ignoring those that complicate it.

Statistics summarize patterns across large numbers of cases, making them powerful evidence — but numbers create an impression of precision and objectivity that can mislead when context, collection methods, or limits are hidden.

An experiment manipulates one or more variables under controlled conditions to observe the effect on another variable, isolating cause-and-effect relationships. The randomized controlled trial (RCT) — in which participants are randomly assigned to a treatment or control group — is considered the gold standard for causal evidence in medicine, public health, and social policy, because random assignment makes the two groups comparable on both measured and unmeasured characteristics.

Not everything can be studied experimentally — it would be unethical to randomly assign children to poverty, for instance. For these questions, researchers turn to observational research: cohort studies, case-control studies, cross-sectional surveys, ethnographic observation, and systematic reviews that synthesize multiple studies.

When you encounter experimental or observational evidence, ask: What was manipulated or measured? Is the sample representative of the population the conclusion is about? How large was it? What alternative explanations did the researchers consider? Have the findings been replicated?

A document is any record — official, institutional, personal, published, legal, or diplomatic — created at or near the time of the events it describes. Documentary evidence is powerful because it is close to the event and reveals what the writer thought, knew, and felt at the time.

Testimony — a person's account of what they experienced, witnessed, did, or believed — appears in oral history, journalism, legal proceedings, and qualitative research. It can reveal subjective experience and reasoning that no document records, but human memory is fallible: people misremember, reinterpret the past, and are influenced by what they have heard from others. Uncorroborated single eyewitness accounts are among the weakest forms of evidence for factual claims, even from sincere people. Testimony is strongest when corroborated by other testimony and by documentary evidence.

Expert opinion is legitimate evidence when a question requires specialized knowledge, when an expert synthesizes a large body of research a non-expert could not practically evaluate alone, and when the question is one on which experts with relevant credentials broadly agree. Expert consensus — the overwhelming agreement of relevant experts — represents collective evaluation of evidence by those best positioned to assess it, and should not be treated as equivalent to a single contrarian voice.

When evaluating expert opinion, ask: Is this person a genuine expert in the relevant domain? Are they operating within that domain? Does their view represent the field or an outlier position? Is their conclusion supported by evidence they cite? Have conflicts of interest been disclosed?

The most important question about any evidence is not just whether it is accurate, but whether it fits the question being asked.

Find one source on a research topic of your choice using the MGU Library catalog, a library database, or a reputable news source. Read it carefully, then record:

Choose one topic — the health effects of urban green space, the causes of declining voter turnout, the representation of race in American film, or the relationship between poverty and academic achievement. Search for two sources: one from a social science or natural science database, and one from a humanities source or journalism outlet.

Use the MGU Research Note Model — Source, Summary, Key evidence, Quotation, Paraphrase, Usefulness, Limits, Citation information — every time you record evidence from a source. Here is a model entry.

Misinformation rarely admits it has no basis — it dresses up weak or fabricated support as evidence. Recognize these common evidence imposters:

When introducing evidence in your own writing, signal its type and source clearly so readers can evaluate it for themselves.