Replication Crisis in Psychology

When a study's results cannot be duplicated or validated using the original data or methods.

Reproducing a result but needing to change the analysis method from the original study to get that result.

A study's reported significant results do not appear when reanalyzing the same data.

An effort to reproduce the results of 100 scientific studies across different laboratories.

The percentage of studies that successfully replicated the original findings.

Cognitive psychology studies have a 50% replication success rate.

Social psychology studies have a 26% replication success rate.

A common reason for reproducibility failure, insufficient information is given about data collection methods and the analysis performed.

A detailed plan for research methods filed online before data collection, including hypotheses, methodologies, and analyses. This plan is set in stone and cannot be changed.

A detailed research plan submitted for peer review before data collection. High-quality protocols are provisionally accepted for publication if the authors follow the registered methodology.

Pre-registration reduces bias by preventing researchers from changing their methods after seeing the data, thus improving the reliability of research findings.

A unique and permanent string of letters and numbers assigned to articles filed online, allowing easy access and identification.

Pre-registration can be applied to studies using existing or secondary data, ensuring transparency and reproducibility of analysis methods.

A statistical method used to determine if there's a significant difference between groups or conditions. It assumes there is no difference.

The probability of obtaining the observed results if the null hypothesis is true. A small p-value suggests evidence against the null hypothesis.

Unethical practice of manipulating data or analysis to achieve statistically significant results.

A result that indicates an effect when there is none. Occurs when p-hacking leads to accepting a false hypothesis.

Stopping data collection when p < .05 to ensure significance.

Analyzing many measures but only reporting those with p < .05.

Adding variables to the analysis until p < .05.

Removing participants until p < .05.

THARKing can be justified in science because it can lead to more efficient and effective research by focusing on areas that show unexpected or interesting results.

HARKing is unethical because it deceives readers into thinking the hypothesis was a priori, which can lead to misleading scientific conclusions.

A researcher created long questionnaires and found significant correlations. These findings were presented as a priori hypotheses, but the hypothesis actually came after analyzing the data.

Researchers found a small correlation between a drug and survival rate, but they noticed a difference between male and female patients. Some researchers may try to present this difference as an a priori hypothesis.

Formulate hypotheses based on existing research and theory before collecting any data. This ensures that the research process proceeds in a logical, scientific manner.

A detailed plan for a research study, outlining methods, hypotheses, and analyses, filed online before data collection.

A template for preregistering quantitative psychology research, emphasizing transparency and reproducibility.

A section in a preregistration plan that outlines how and where the study data will be made available to other researchers.

The date and time when the research plan was preregistered, providing a timestamp for the protocol.

A section in a preregistration plan that indicates whether the research protocol has been approved by the Institutional Review Board (IRB).

The initial review of a Registered Report's proposed methodology and research plan.

The approval of a Registered Report's methodology and research plan, guaranteeing publication if the research is conducted as planned.

The final review of a Registered Report after the research is completed, verifying if the methods were followed accurately.

A research plan that includes methods and hypotheses but doesn't undergo a formal Stage 1 review.

The traditional peer review system that occurs after a research study has been completed.

Pre-registration only plans the study beforehand. Registered Reports also have a Stage 1 review and 'In-Principle' acceptance, guaranteeing publication if the study is conducted according to the plan.

The tendency for studies with statistically significant results to be published more often, skewing the scientific literature.