1 of 7

Behavioral Economics��Estimating the reproducibility of psychological science��Open Science Collaboration�� Corresponding Author: Brian Nosek (270 co-authors)��A Popperian Test of Modern Research Integrity��Prof. Yuval Heller��Yonatan Cahan���June 2026

1

2 of 7

Why Replicate? The Popperian Foundation

"Non-reproducible single occurrences are of no significance to science." -Karl Popper, 1935

2

The Falsification Rule

Scientific claims must be testable and repeatable. A finding is only "scientific" if another researcher can reproduce it under the same conditions.

The 2015 Question: Does modern psychology stand up to this foundational requirement?

3 of 7

Research Methodology: Putting Theory to the Test

Massive Scale

270+ researchers collaborated to replicate 100 studies from three top-tier journals (2008).

3

High Fidelity

Replications used original materials, protocols, and high statistical power.

​

 

4 of 7

Key Findings: The Falsification Shock

  •  

4

 

36%

Replication studies succeeded in finding significant results.

5 of 7

Visualizing Figure 1: Effect Size Comparison

  • The diagonal line represents a perfect world where original and replication effects are identical.
  • The vast majority of data points sit below the diagonal, showing a systematic overestimation in published research.
  • Points near or below zero represent studies where the effect completely disappeared or reversed.

5

6 of 7

Underlying Causes: Systemic Bias

Publication Bias

Journals prioritize "positive" and exciting results, while null findings stay hidden in drawers.

Low Power

Small samples in original studies led to many "false positives" that could not be replicated.

6

P-Hacking

Selective reporting and flexible data analysis to force p-values below the 0.05 threshold.

7 of 7

"No matter how many instances of white swans we may have observed, this does not justify the conclusion that all swans are black."��-Karl Popper (Logic of Scientific Discovery, 1935)

The Core Lesson: Even successful replication is not absolute truth. We must shift from chasing confirmation to embracing transparency and falsification.

7