In 2015, the Open Science Collaboration attempted to replicate 100 published psychology studies. Only 39 out of 100 produced the same results.
That’s a 61% failure rate in peer-reviewed, published research.
This is the Replication Crisis, and it doesn’t just affect psychology. Similar problems have been found in medicine, economics, neuroscience, and nutrition science.
Why does this happen?
🍂 Publication bias – Journals prefer exciting, positive results. Studies that find “no effect” rarely get published.
🍂 P-hacking – Researchers tweak their data analysis until they get a statistically significant p-value (p < 0.05).
🍂 Small sample sizes – A study of 30 university students cannot reliably represent all humans, yet thousands of such studies have shaped mainstream psychology.
🍂 HARKing – Hypothesizing After Results are Known. Presenting a post-hoc explanation as if it was the original hypothesis.
🍂 Funding bias – Industry-funded studies are significantly more likely to produce results favorable to the funder.
How to read research critically:
Before trusting any study, ask:
🍂 What was the sample size, and who was in the sample?
🍂 Has this been independently replicated?
🍂 Who funded the research?
🍂 Was it pre-registered? (Did they publish their hypothesis before collecting data?)
🍂 Is it a single study, or part of a meta-analysis?
A single study proves almost nothing. Converging evidence across multiple independent studies is what builds reliable knowledge.
🌷Why this matters for your own research:
Understanding these flaws makes you a better researcher. Pre-registering your hypotheses, being transparent about methodology, and publishing null results are all things the scientific community is increasingly demanding.