What methodology did the 2015 Veale et al study use for measuring penis length?

1. Study design and inclusion/exclusion filters

Veale and colleagues conducted a formal systematic review with predefined eligibility rules: included studies had to report penis size measured by a health professional using a standard procedure and generally required at least 50 participants per sample, while studies of men with congenital or acquired penile abnormalities, prior penile surgery, self‑selected small‑penis complaints or erectile dysfunction were excluded to avoid biased estimates ^{[1] [2] [4]}.

2. What states of the penis were pooled and why that matters

The review distinguished three measurement states commonly reported in the literature — flaccid, stretched flaccid (stretched length), and erect — and constructed nomograms for each state because those conditions produce systematically different values and clinical relevance; Veale et al. noted that erect and stretched measurements correlate but that variability (especially in stretched/flaccid measures) was substantial across studies ^{[3] [1] [5]}.

3. How length was measured (operational definition)

Across the included studies the operational definition of length was consistent with clinical anthropometry: measurements were taken along the dorsal surface from the penopubic junction or pubic bone to the tip of the glans, using a rigid ruler or straight edge pressed to the pubic ramus to standardize the proximal landmark; stretched length was obtained by applying traction to the glans until a point of increased resistance was felt (an “easily appreciated endpoint”) and erect length was measured at full erection ^{[6] [7] [8]}.

4. Tools, environment and who did the measuring

Most component studies reported use of a semi‑rigid ruler for length and a flexible tape measure for circumference/girth; Veale limited inclusion to measurements taken by health professionals in clinical settings when possible, and many studies controlled environmental factors such as room temperature or used private periods for self‑stimulation before erect measures — although methods varied between studies, contributing to heterogeneity ^{[9] [10] [6]}.

5. How erect measurements were obtained and their limits

Erect measurements in the pooled data were often obtained without pharmacological induction — typically after voluntary erection via self‑stimulation or in the presence of a partner — but Veale flagged that relatively few erect measurements came from controlled clinical settings and that those limitations reduce confidence in pooled erect estimates compared with flaccid or stretched values ^{[3] [2] [11]}.

6. Data synthesis, nomograms and acknowledged sources of bias

The authors pooled eligible datasets (reporting up to 15,521 men across included studies) to construct percentile nomograms for length and circumference intended for clinical counselling and research, but they explicitly acknowledged substantial between‑study heterogeneity, interobserver variability in measurement technique and the disproportionate contribution of flaccid/stretched data to variance; they recommended cautious interpretation and highlighted the need for standardized prospective measurement protocols ^{[2] [3] [5]}.

7. Takeaway and what the paper does not claim

Veale et al. provide a clinical tool built from aggregated, clinician‑measured data and do not claim to remove all methodological uncertainty: their nomograms are a synthesis within the limits of variable measurement techniques, sample selection and sparse standardized erect measures, and they call for more standardized, prospectively collected data to reduce dispersion and observer error ^{[3] [9] [7]}.