A Shortcut to Rejection: How Not to Write the Results Section of a Paper

Abstract

This article discusses common errors in writing up the results of papers. It covers the following: (1) giving details about irrelevant topics, such as what program was used to enter the data, while ignoring important ones (for example, which options were chosen for various statistical tests); (2) reporting P levels of 0.0000 and negative values for t tests; (3) giving the P levels but not the actual values of the statistical tests; (4) not including confidence intervals and measures of the magnitude of an effect; (5) testing irrelevant hypotheses, such as whether reliability or validity coefficients are significantly different from zero; (6) attributing reliability and validity to tests rather than to the circumstances under which they are given; and (7) reporting levels of accuracy that cannot be supported by the data. Suggestions are made regarding the proper way to report findings.

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 — 31 citations exist upstream.

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