Presenting percentages from tiny samples falsely implies representativeness.

Evidence

The small size of the sample, 171 subjects, meant that the percentage figures given in the tables sometimes concealed absurdly small real numbers. But when quantitative data were being presented it was impossible to avoid giving the impression that they were representative, and in the spring of 1959 Friedeburg therefore undertook a supplementary study. — Rolf Wiggershaus, The Frankfurt School, 581

Categorization

• DEFINE: Clarifies why a follow-up study with 550 students was run—the original 171-subject sample could not support percentage claims.
• XREF: connects to: Very low response rates make popular sex surveys misleading; Aggregate scoring percentages cannot demonstrate per-category rater agreement.; Small student samples made percentage figures misleadingly representative — Rolf Wiggershaus, The Frankfurt School: Its History,…, 2351 (+5 more)
• yield-rescued: below the book’s yield-target cutoff, included anyway for connecting to: Very low response rates make popular sex surveys misleading; Aggregate scoring percentages cannot demonstrate per-category rater agreement.; Small student samples made percentage figures misleadingly representative — Rolf Wiggershaus, The Frankfurt School: Its History,…, 2351 (+5 more)

Broader Connections

• Small student samples made percentage figures misleadingly representative #expands (0.95) — Rolf Wiggershaus, The Frankfurt School, 2351