Medical Student Anki Statistics: Step 1 Scores and Research
In the largest multi-school survey published so far, 68.3% of US medical students used Anki, averaging 17.8 hours a week on it (Halperin et al., 2024). The most quoted performance number comes from Deng, Gluckstein and Larsen (2015): every 1,700 unique Anki cards seen was associated with one extra Step 1 point. Every study below is observational. Students choose whether to use Anki, so none of them can prove Anki causes higher scores.
Key statistics
68.3% of US medical students who answered a resources question used Anki, in a 2020 survey of 560 students from 102 schools (Halperin et al., 2024).
Anki users in that survey reported 17.8 hours a week on Anki (SD 14.9), close to the 18.5 hours spent on lectures (Halperin et al., 2024).
Each additional 1,700 unique Anki cards seen was associated with one extra point on USMLE Step 1 among 72 students at one school (Deng, Gluckstein and Larsen, 2015).
Anki users at UNC averaged 241.1 on Step 1 against 235.5 for non-users, an effect size of d = 0.38 (Lu, Farhat and Beck Dallaghan, 2021).
Daily Anki users at the University of Minnesota had a median Step 1 score of 238 against 233.5 for other students, but no difference on Step 2 CK (250 vs 251) (Wothe et al., 2023).
A 2026 systematic review found 11 studies of Anki and exam results in medical students, 1,135 students in total, and called the evidence "largely observational" (Frappa et al., 2026).
94% of first-year medical students at UCF used Anki in 2025, and 97.6% of them relied on pre-made cards (Nour and Harris, 2025).
34.5% of those UCF Anki users said they had lost sleep because of Anki, and 82.1% found it overwhelming to some extent (Nour and Harris, 2025).
93% of first-time Step 1 takers from US MD schools passed in 2025, down from 98% in 2020, when Step 1 still reported three-digit scores (USMLE performance data).
The Step 2 CK mean for first-time takers from LCME-accredited schools rose from 244 in 2018 to 2019 to 251 in 2025 to 2026 (USMLE Score Interpretation Guidelines).
The average MCAT total score is 500.6, based on 305,494 scores from the 2023 to 2025 testing years (AAMC, 2026).
The AnKing Step Deck contains over 30,000 cards covering Step 1, 2 and 3 (AnkiHub).
How many medical students use Anki?
Somewhere between half and nine in ten, depending on the school, the year and how "use" is defined. Adoption has climbed since the mid-2010s, when commercial flashcard tools still competed with Anki.
68.3% (285 of 417 respondents to the resources question) used Anki in a national survey that collected 560 responses from 102 US medical schools between September and December 2020. Only online resources such as Boards and Beyond or Pathoma (83.2%) and lectures (82.5%) were used more often (Halperin et al., 2024).
Anki users in the same survey spent 17.8 hours a week on it on average (SD 14.9). The wide spread means some students used it lightly and others treated it as their main resource.
Of students whose schools recorded lectures, only 54% watched lectures live. Those watching recordings did so at an average of 1.75x speed, with 2x the most common setting (Halperin et al., 2024).
In 2015, only 31% of 72 surveyed students used Anki, while 49% used the commercial Firecracker platform and 10% tried both (Deng et al., 2015).
At the University of North Carolina, 132 of 201 analyzed respondents (66%) had used Anki significantly before Step 1, surveyed in March 2020. Among users of pre-made decks, 72.7% used the Zanki Step decks, a predecessor of AnKing (Lu et al., 2021).
About 70% of 60 first-year students responding at one school used Anki in a structure and function course (Harris and Chiang, 2022).
56% of 165 respondents at the University of Minnesota Medical School described themselves as daily Anki users (Wothe et al., 2023).
78 of 130 first-year students (60%) at Boonshoft School of Medicine used Anki for at least one exam after an Anki training program in 2021 (Gilbert et al., 2023).
94% of first-year respondents at the University of Central Florida were Anki users. The survey reached 74% of the class (89 of 120) (Nour and Harris, 2025).
Outside the US, 93.8% of 306 students at Alfaisal University in Saudi Arabia knew about Anki and 51% used it regularly (Alnaser-Almusa et al., 2026).
How med students use Anki
Almost everyone uses pre-made decks, and most use them most days.
97.6% of UCF Anki users relied on pre-made cards rather than writing their own (Nour and Harris, 2025).
67.9% used Anki at least five days a week, and 78% preferred fill-in-the-blank (cloze) cards to other formats.
87.8% believed Anki significantly contributed to their success in modules, and 59% thought it boosted productivity compared with other study tools.
Anki users tend to score higher on Step 1, typically by about 4 to 11 points, but the studies measure association, not cause.
In a multivariate model, unique Anki cards seen, practice questions completed, second-year honors and MCAT score all independently predicted Step 1 score. The model explained 67.2% of the variance (R² = 0.672) (Deng et al., 2015).
Each additional 445 practice questions or 1,700 unique Anki cards was associated with one more Step 1 point. Unique Firecracker flashcards seen did not predict the score.
Respondents in Deng et al. reported a mean Step 1 score of 253.6 (SD 10.8), well above the national mean, so the sample skewed toward high performers. The authors called their results "exploratory" and said causal claims would need a randomized trial.
Anki users at UNC averaged 241.10 (SD 13.17) on Step 1, against 235.51 (SD 17.68) for non-users (p = 0.012). The effect size, d = 0.38, is small to medium (Lu et al., 2021).
Consistency mattered more than use alone. Students who completed their full reviews "most of the time" averaged 246.94, against 236.32 for those who did so "sometimes" (p = 0.010).
39% of high Anki users felt they had to relearn less than a quarter of the material during dedicated Step 1 study, compared with 7% of low users. That is self-reported, not measured, retention.
Daily Anki users at Minnesota had a median Step 1 score of 238 against 233.5 for non-daily users (p = .039). On Step 2 CK the medians were 250 and 251, with no significant difference (p = .440) (Wothe et al., 2023).
The 560-student national survey found no correlation between third-party resource use and Step 1 score (Halperin et al., 2024). Not every study finds an Anki advantage.
Reported score gaps favor Anki users on Step 1, but the one study that also checked Step 2 CK found no difference.
Beyond Step 1: course exams, GPA and the systematic review
At Boonshoft, after adjusting for MCAT score, Anki users scored 6.4, 6.2 and 7.0 percentage points higher on three first-year NBME course exams (Gilbert et al., 2023).
On the Comprehensive Basic Science Exam, users averaged 72.2% against 59.3% for non-users, a 10.7 point gap after adjusting for MCAT score. Only 10 Anki users in that group took the CBSE, so the estimate is imprecise (95% CI 2.6 to 18.7).
At Alfaisal University, Anki use was not associated with cumulative GPA: 3.59 for users against 3.55 for non-users (p = 0.52) (Alnaser-Almusa et al., 2026).
A 2026 systematic review identified 11 eligible studies published from 2015 to 2025, covering 1,135 medical students. Three showed a consistent positive association with Step 1, with high-frequency users outscoring minimal users by 4 to 13 points. The only study to assess Step 2 CK found no significant benefit (Frappa et al., 2026).
Correlation, not causation. In none of these studies was Anki use randomly assigned. The review authors note that the apparent dose-response pattern "may partly reflect self-selection," since stronger, more motivated students with more time are more likely to keep up daily reviews. Wothe et al. point out that daily Anki users may simply have studied more hours overall, which they did not measure.
Response rates were also low in several studies: 18.6% at Minnesota and 36.9% at UNC. Students who like Anki are more likely to answer an Anki survey.
The "1,700 cards per point" number, read correctly
This is the most repeated Anki statistic in medicine, and it is often quoted as if 1,700 cards will buy you a point. The study does not support that reading.
The figure comes from a regression coefficient of 5.9 × 10⁻⁴ points per unique card (p = 0.024) in a sample of 72 students at one school with a 65% response rate (Deng et al., 2015).
A published erratum corrected Table 3, where the coefficient had been printed as 5.89 × 10⁻³, ten times too large. The body text was already right (Erratum, 2016). Anyone deriving "170 cards per point" from the table is using the typo.
Practice questions carried a larger coefficient per item: one point per 445 questions. Per unit of effort, question banks looked at least as valuable as flashcards in this data.
The data is mixed: daily users report better sleep quality in one study, while a majority of users in another describe Anki as overwhelming.
Daily Anki use was associated with better perceived sleep quality (p = .01) but not with more hours of sleep. Stress, burnout and leisure time showed no significant association with Anki use (Wothe et al., 2023).
82.1% of UCF Anki users found Anki overwhelming to some extent, and 67.9% said it made them feel anxious to some extent (Nour and Harris, 2025).
34.5% had lost sleep because of Anki, 36.9% had forgotten to eat, and 52.5% had skipped exercise to do cards.
If your review count has turned into that, our Anki burnout guide covers how to cut the daily load without losing what you have learned.
USMLE pass rates since Step 1 went pass/fail
Step 1 pass rates for US MD first-time takers were 96% to 98% from 2019 to 2021 and 91% to 93% from 2022 to 2025, after pass/fail reporting began.
Step 1 has been reported as pass/fail only for exams taken on or after January 26, 2022 (USMLE).
The same day, the Step 1 passing standard rose from 194 to 196 on the three-digit scale (USMLE announcement). The pass rate drop coincides with both changes, so it cannot be pinned on pass/fail alone.
The Step 2 CK passing score is 218 for exams from July 1, 2025, and the Step 3 passing score is 200 from January 1, 2024.
The last published Step 1 means before pass/fail were 232 (2019), 235 (2020) and 231 (2021) for first-time takers from LCME-accredited schools, each with an SD of 18 to 19 (USMLE Score Interpretation Guidelines, 2022).
With Step 1 unscored, the Step 2 CK mean has crept up: 244 (2018 to 2019), 246 (2020 to 2021), 249 (2023 to 2024), 250 (2024 to 2025) and 251 (2025 to 2026), SD 15 in recent years (USMLE Score Interpretation Guidelines, updated August 2026).
The Step 3 mean has held at 227 (SD 15) in 2023, 2024 and 2025.
Step 1 first-time pass rates by year
Percent of first-time takers passing USMLE Step 1. Source: USMLE performance data. Canadian schools are grouped with US schools through 2024 and with non-US schools from 2025.
Year
US MD first-time
US DO first-time
Non-US school first-time
Score reporting
2019
97% (21,308)
96% (4,794)
82% (14,046)
Three-digit score
2020
98% (19,772)
96% (5,235)
87% (11,742)
Three-digit score
2021
96% (22,280)
94% (5,309)
82% (16,952)
Three-digit score
2022
93% (22,828)
89% (4,659)
74% (22,030)
Pass/fail from Jan 26
2023
92% (23,100)
87% (4,798)
72% (22,611)
Pass/fail
2024
91% (23,280)
86% (4,873)
73% (22,713)
Pass/fail
2025
93% (23,028)
89% (4,623)
75% (22,066)
Pass/fail
Numbers in brackets are first-time examinees tested. Through 2024, the MD and DO columns include Canadian schools.
First-time Step 1 pass rates dropped for every group in 2022, when pass/fail reporting and a higher passing standard started on the same day.
Step 2 CK and Step 3 pass rates
Percent of first-time takers passing. Step 2 CK years run July 1 to June 30; Step 3 years are calendar years. Source: USMLE performance data.
Exam and year
US MD first-time
US DO first-time
Non-US school first-time
Step 2 CK, 2023 to 2024
98% (22,439)
96% (5,318)
89% (15,064)
Step 2 CK, 2024 to 2025
98% (22,249)
96% (5,025)
90% (16,186)
Step 3, 2024
97% (21,213)
93% (110)
89% (12,134)
Step 3, 2025
96% (21,589)
97% (129)
88% (12,480)
Repeat takers pass far less often. In 2025, 71% of Step 1 attempts by repeaters from US MD schools passed, and 54% of attempts by repeaters from non-US schools (USMLE performance data).
Most med students study from pre-made decks, but lecture-specific material still needs cards. StudyCards AI turns your lecture PDFs into Anki-ready flashcards, including cloze cards, so your school content gets the same spaced review as AnKing.
MCAT score data
The average MCAT total score is 500.6, while the average student admitted to a US MD school scores 512.1.
The mean MCAT total is 500.6 (SD 11.2), based on 305,494 scores from the 2023, 2024 and 2025 testing years combined. These are the percentile ranks in effect May 1, 2026 to April 30, 2027 (AAMC percentile ranks).
A score of 500 sits at the 48th percentile, 508 at the 74th, 510 at the 79th, 515 at the 91st and 520 at the 97th. A 522 or above rounds to the 99th percentile or higher.
Applicants to US MD schools for the 2025 to 2026 cycle averaged 506.3 (SD 10.0), across 54,699 applicants (AAMC FACTS Table A-16).
Matriculants that cycle averaged 512.1 (SD 6.5), with a mean total GPA of 3.81, across 23,440 matriculants. The matriculant mean was 511.5 in 2020 to 2021.
On "dual coding" and the MCAT: the term is widely used in MCAT prep, but the AAMC content outline for the Psychological, Social, and Biological Foundations of Behavior section does not name it. Memory content sits in Content Category 6B, which lists encoding, storage, retrieval and forgetting, including "processes that aid in encoding memories" (AAMC, Content Category 6B). Dual coding is best studied as one of those encoding aids, not as a named test objective.
AnKing deck size
The AnKing Step Deck has over 30,000 cards, far more than most students finish before Step 1.
The official page lists "over 30k flashcards for Step 1, 2, & 3," says the deck is trusted by over 100,000 medical students, and reports over 300,000 downloads (AnkiHub, AnKing Step Deck).
The deck descends from Zanki and Lolnotacop. In the 2020 UNC survey, 72.7% of pre-made deck users used Zanki Step decks and 21.2% used Lolnotacop (Lu et al., 2021).
Put against Deng et al., 30,000 unique cards would correspond to roughly 18 Step 1 points (30,000 × 5.9 × 10⁻⁴). That is an extrapolation far beyond the 2015 sample and should not be read as a prediction.
Our AnKing deck guide explains how to unsuspend cards by lecture rather than trying to see all of them. For the wider evidence on spacing itself, see our spaced repetition statistics.
Methodology
We included peer-reviewed studies that measured Anki use among medical students, official USMLE and NBME performance data, and official AAMC score tables. Every number was checked against the primary source, the journal article, the USMLE or AAMC page, or the official AnkiHub page, in September 2026. We did not use secondary statistics roundups. Where a study reported medians rather than means, or self-reported rather than measured outcomes, we say so. Survey response rates are given where published because low response rates inflate apparent Anki use.
Sources
Halperin SJ, Zhu JR, Francis JS, Grauer JN. (2024). Are medical school curricula adapting with their students? A survey on how medical students study and how it relates to USMLE Step 1 scores. Journal of Medical Education and Curricular Development. PMC10807313
Deng F, Gluckstein JA, Larsen DP. (2015). Student-directed retrieval practice is a predictor of medical licensing examination performance. Perspectives on Medical Education, 4(6), 308 to 313. PMC4673073
Erratum to: Student-directed retrieval practice is a predictor of medical licensing examination performance. (2016). Perspectives on Medical Education. PMC5122521
Lu M, Farhat JH, Beck Dallaghan GL. (2021). Enhanced learning and retention of medical knowledge using the mobile flash card application Anki. Medical Science Educator, 31(6), 1975 to 1981. PMC8651966
Wothe JK, Wanberg LJ, Hohle RD, Sakher AA, Bosacker LE, Khan F, Olson APJ, Satin DJ. (2023). Academic and wellness outcomes associated with use of Anki spaced repetition software in medical school. Journal of Medical Education and Curricular Development. PMC10176558
Gilbert MM, Frommeyer TC, Brittain GV, Stewart NA, Turner TM, Stolfi A, Parmelee D. (2023). A cohort study assessing the impact of Anki as a spaced repetition tool on academic performance in medical school. Medical Science Educator, 33(4), 955 onward. PMC10403443
Harris DM, Chiang M. (2022). An analysis of Anki usage and strategy of first-year medical students in a structure and function course. Cureus, 14(3), e23530. PMC9045547
Nour H, Harris DM. (2025). Utilization patterns and perceptions of a spaced repetition flashcard program, Anki, among first-year medical students. Cureus, 17(10), e95674. PMC12662189
Alnaser-Almusa O, Sajid MR, El-Sibai A, Ammar O, Aldahoul S, El Hayek T, Tamim H. (2026). Exploring the effectiveness and awareness of Anki flashcard application on medical student's academic performance: a cross-sectional study from Saudi Arabia. Frontiers in Medicine, 13, 1896043. PMC13474006
Frappa N, Chernov D, Dillon M, Alben MG. (2026). Anki use and academic performance in medical education: a systematic review of evidence and learning theory. Medical Science Educator, 36(2), 1015 to 1025. PMC13197492
United States Medical Licensing Examination. Examination results and scoring. usmle.org
United States Medical Licensing Examination. Change to Step 1 passing standard begins January 26, 2022. usmle.org
United States Medical Licensing Examination. USMLE Score Interpretation Guidelines, updated August 17, 2026. PDF
United States Medical Licensing Examination. USMLE Score Interpretation Guidelines, updated May 19, 2022 (pass/fail transition). PDF
Association of American Medical Colleges. (2026). Summary of MCAT total and section scores, percentile ranks in effect May 1, 2026 to April 30, 2027. PDF
Association of American Medical Colleges. FACTS Table A-16: MCAT scores and GPAs for applicants and matriculants to U.S. MD-granting medical schools, 2020 to 2021 through 2025 to 2026. XLSX
Association of American Medical Colleges. Psychological, Social, and Biological Foundations of Behavior section: Content Category 6B. students-residents.aamc.org
You are welcome to use these statistics in your own writing. Please credit StudyCards AI with a link to this page.
APA
Groves, C. (2026). Medical Student Anki Statistics: Step 1 Scores and Research. StudyCards AI. https://studycardsai.com/blog/medical-student-anki-statistics
HTML link
<a href="https://studycardsai.com/blog/medical-student-anki-statistics">Medical Student Anki Statistics: Step 1 Scores and Research</a>
Frequently Asked Questions
What percentage of medical students use Anki?
Most published surveys put it between about half and nine in ten. The largest multi-school study, 560 students from 102 US medical schools surveyed in 2020, found 68.3% of those who answered the resources question used Anki (Halperin et al., 2024). Single-school surveys range from 31% in 2015 (Deng et al.) to 94% of first-years at UCF in 2025 (Nour and Harris).
Does Anki raise your Step 1 score?
Anki users score higher on average, but no study proves Anki causes the gain. At UNC, Anki users averaged 241.1 on Step 1 against 235.5 for non-users, a small to medium effect (d = 0.38; Lu et al., 2021). Every study is observational, so motivation and total study hours could explain part of the gap.
How many Anki cards equal one Step 1 point?
In Deng, Gluckstein and Larsen (2015), each additional 1,700 unique Anki cards seen was associated with one extra Step 1 point, after controlling for MCAT score, grades and question bank use. The same model tied one point to every 445 practice questions. The sample was 72 students at one school, so treat it as exploratory.
What is the Step 1 pass rate now that it is pass/fail?
In 2025, 93% of first-time takers from US MD schools passed Step 1, as did 89% of US DO first-timers and 75% of first-timers from non-US schools (USMLE performance data). The US MD rate was 96% to 98% from 2019 to 2021, before pass/fail reporting and a higher passing standard of 196 began on January 26, 2022.
What is the average MCAT score?
The average MCAT total score is 500.6 (SD 11.2), based on 305,494 scores from the 2023, 2024 and 2025 testing years (AAMC percentile ranks in effect May 2026). People admitted to US MD schools score much higher: matriculants in the 2025 to 2026 cycle averaged 512.1, and applicants averaged 506.3 (AAMC FACTS Table A-16).
How many cards are in the AnKing deck?
The AnKing Step Deck has over 30,000 cards covering USMLE Step 1, Step 2 and Step 3, according to its official AnkiHub page, which also says it is trusted by over 100,000 medical students. At 1,700 cards per point from Deng et al. (2015), finishing it is not a promise of a score, since that estimate is correlational.