Retraction is a mechanism for removing flawed or fabricated results from the research literature. We used the Retraction Watch Database (http://retractiondatabase.org/) to explore trends in retractions in psychology (analysis code is available at https://osf.io/w6fev). The Retraction Watch database contains records of more than 15,000 retractions of scientific journal articles and conference papers from 1927 to today. Due to variation in retraction notices, there are a few inconsistencies in the database. The date of a retraction is often the date that the article’s retraction notice was published, but is sometimes the original publication date. The subject area categorization does not always align with our conceptualization of psychological science as inclusive of behavioral psychology, clinical psychology, and neuroscience. Consequently, our analyses - which use the ”(SOC) Psychology” database tag - may undercount retractions in the psychological sciences (e.g., a retraction in the journal Psychological Science related to hand sanitizer use in the workplace is tagged as “(HSC) Public Health and Safety”). Before conducting the analyses below, we removed two mega-retraction events from the database: 1) the simultaneous retraction of 7,500 articles and conference abstracts originally published by the Institute of Electrical and Electronics Engineering and 2) the simultaneous retractions of 434 conference abstracts originally published by the Journal of Fundamental and Applied Sciences. Prior to their removal, these two events accounted for 36% of all retractions with the tag “(SOC) Psychology”. We assessed 14,966 retractions in psychology (i.e., retractions tagged ”(SOC) Psychology”). The number of retractions in psychology is more than other social sciences (communication, education, political science, or sociology). In the years 2002-2010, the mean number of retractions per year was 3.4 (SD = 3.1), and the highest number of retractions in a single year was 11 in 2009. In 2011, the fraud of Diederik Stapel was discovered; since that time, psychology journals have retracted an average of 35.7 articles per year (SD = 11.8). Outside of major investigations of specific researchers, retraction for falsification or fabrication are relatively rare in psychology; more common are retractions citing concerns about the reliability of data and results. The Retraction Watch database lists the non-exclusive reasons for retraction of each article as described in the article’s retraction notice. 14.2% have no listed reason. The most commonly cited reason for retraction was errors in analyses, data, methods, results, or conclusions (21.1%). Other common reasons were concerns regarding the accuracy or validity of the data or results (20.6%), falsification or fabrication (20.4%), plagiarism or other issues regarding referencing (14.8%), and duplication of articles (i.e., “self-plagiarism,” 12.2%). The median lag time between publication and retraction in psychology is 1.9 years. The average lag time is considerably longer, 3.6 years. The slowest retractions have just appeared in 2020, retracting the work of Hans Eysenck and Ronald Grossarth-Matticek after more than twenty years. Researchers have argued that the increased rate of retractions in psychology over time is caused by improved journal oversight rather than increased incidence of fraud, citing more journals issuing retractions but not more retractions per journal (Brainard & You, 2018). The increase of retractions over time in psychology are likely not due to increasing misconduct, but instead due to an increasing political will and ability to detect research misconduct and impossible statistical results using tools like GRIM and SPRITE (Brown & Heathers, 2017; Heathers et al., 2018).