Scientific Gullibility

Abstract

If “gullible” means “too easily persuaded,” then “scientific gullibility” means “too easily persuaded by data or reasoning that does not actually justify that conclusion.” Do standards exist for reaching valid conclusions from logic and data, and, if so, can it be shown that scientists systematically violate them? This chapter argues that the answers to both questions are “yes.” Some standards are common across many types of psychological research. For example, small samples are notoriously unreliable, so that conclusions based on such studies should be treated as extremely preliminary and tentative. An even more basic standard is that claims about facts require data. Despite these standards, even the most seasoned researchers may too readily reach conclusions that violate them. Researchers may engage in motivated reasoning, easily accepting evidence that supports preferred conclusions and intensely scrutinizing evidence that supports undesired claims. They may fall prey to excessive scientism, mistakenly conflating a finding being published with the finding being an established fact. They may also fall victim to status quo bias or status biases, such that they err on the side of maintaining the scientific consensus, or use prestige associated with a researcher, rather than strength of underlying evidence, as a heuristic when evaluating that researcher’s work. These factors may explain why some “scientific breakthroughs” fail to hold up when subjected to scientific scrutiny. The chapter concludes with recommendations for limiting scientific gullibility.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 8 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.