Discovery Lab Answers

Suitbert Ertel — Experiments and Data

Suitbert Ertel (1932–2017) was an emeritus professor of psychology at the University of Göttingen, recognized in the German anomalistics community as a pioneering researcher in parapsychology and anomalistic science ^[1]. His empirical work covered several distinct lines of investigation; the sources retrieved for this question concentrate on two: his Ball Selection Test (BST) program and his reanalysis of existing astrological data.

The Ball Selection Test (BST)

The BST is Ertel’s central experimental paradigm. Participants draw ping-pong balls (or beads) blindly from a bag and attempt to guess — before each draw — which number and/or color they will select. The paradigm was designed to be simple enough for unsupervised home testing, with high-performing participants then brought into supervised laboratory conditions.

Design and replication logic

Ertel’s program followed a two-stage selection model:

1. Home phase — large unselected samples tested at home; those scoring significantly above chance identified as “gifted” participants.

2. Laboratory phase — selected participants retested under direct experimenter supervision to check whether the home results replicated under controlled conditions.

This sequence appears across multiple venues and replication sites described in the sources.

Key reported results

The sources contain intact statistical figures for several experimental runs. Where the surrounding text is clear, they are summarized below.

The APRU replication at Goldsmiths College, London, is notable in its design: it was conducted by two graduating students under the supervision of a skeptical professor at the Anomalistic Psychology Research Unit ^[3]. Ertel reports that the APRU participants’ lower hit rate (relative to GEMI participants) was predicted in advance, on the grounds that unselected samples would show lower effect sizes than the selected GEMI group ^[3].

Sensory leakage controls

A central methodological question throughout the BST program is whether results reflect psi or sensory/tactile/memory leakage. The 2013 paper ^[2] addresses this directly by requiring seven high-performing participants — including members of the Ukrainian family who had performed in earlier studies — to complete BST runs while wearing both blindfolds and gloves, designed to preclude visual, tactile, and temperature cues.

Ertel also tested a memory-leakage hypothesis: if participants were remembering ball placements from earlier draws, their hit rate should be higher on repeated-sequence trials (“11111..22222..”) than on varied trials (“12345..”). The sources report this control was included in the 2013 study ^[2], and that hit rates under the standard and blindfold/glove conditions were nearly identical.

Reanalysis of Carlson’s Astrology Data

A separate line of Ertel’s work involves reanalyzing published datasets from other researchers, most prominently **Shawn Carlson’s 1985 study published in Nature**, which Carlson presented as a decisive negative result for astrology ^[4].

Ertel’s reanalysis ^[4] argued that Carlson had applied an incorrect statistical test — using a normal approximation where a binomial test was appropriate. Recalculating using the binomial:

• Study 1 (CPI profile matching, 3-way forced choice): binomial z = 1.61, p = .054 (marginal, three-way forced choice); a 10-point rating method applied to the same data yielded τ = 0.088, z = 1.78, p = .037 (one-tailed), ES = 0.10 ^[4].
• First + second choice combined: observed proportion 86/116 = .741, against an expected proportion of .663 (77/116), with a 95% confidence interval of .57–.75 on the expected value ^[4].

Ertel’s conclusion was measured: the results were “insufficient to deem astrology as empirically verified, but sufficient to regard Carlson’s verdict as premature” ^[4]. He did not claim positive proof of astrological validity, but contested the statistical basis of Carlson’s negative conclusion.

Skeptical Critiques

What critics argue. The APRU replication context is directly relevant here. The Anomalistic Psychology Research Unit at Goldsmiths College was explicitly characterized as a skeptic-led environment ^[3]. Broader methodological concerns in this literature focus on whether BST procedures adequately exclude sensory leakage — particularly tactile and temperature differences between balls — and whether selection of “gifted” participants from home tests introduces regression-to-the-mean artifacts.

What the experimental data show. Ertel’s 2013 study ^[2] reports that hit rates under the standard BST procedure and under the blindfold-plus-glove modification were nearly identical across 3,840 trials, with both results significantly above the mean chance expectancy of 20%. The APRU result ^[3], conducted under skeptical supervision with unselected participants, was statistically significant by two independent tests. However, the APRU effect size was substantially lower than GEMI laboratory results, and Ertel explicitly predicted this gap — attributing it to the difference between unselected and selected participants rather than to an experimental artifact ^[3].

Analysis. The gap between effect sizes at GEMI (ES = 0.247) and IGPP (ES = 0.110) for the same Ukrainian family participants ^[5] is itself a datum requiring explanation — the same participants, the same basic procedure, different laboratories, a more than twofold difference in effect size. Ertel acknowledges the cross-site variability. Independent replication by investigators outside Ertel’s own group is represented by the APRU result ^[3], but the APRU study was itself analyzed and reported by Ertel, raising the question of whether a fully independent replication — with independent data analysis — has been published.

The Ertel (2010) study-audit page on ESP-Nexus covers the APRU replication study in detail, including document-integrity notes and a review of the paper’s own reported results.