Bland-Altman Agreement Analysis
Bland–Altman analysis assesses agreement between two measurement methods by estimating the mean difference (bias) and the 95% limits of agreement within which most differences fall.
Bland-Altman 분석은 두 측정 방법의 차이 평균(편향)과 대부분의 차이가 들어가는 95% 일치 한계를 추정해 두 방법의 일치도를 평가합니다.
Does a new wearable heart-rate monitor agree closely enough with an ECG to replace it?
새 웨어러블 심박계가 심전도를 대체할 수 있을 만큼 측정값이 잘 일치하나요?
Bland–Altman analysis plots the difference between two paired measurements against their mean and summarises agreement with the mean difference and the limits bias ± 1.96 SD of the differences. Confidence intervals for the bias and limits, and a regression of differences on means, show whether disagreement is systematic or grows with magnitude.
When to use Bland-Altman Agreement Analysis
- Two methods measure the same continuous quantity on the same subjects
- You need to know how far individual measurements may differ
- Checking for fixed or proportional bias between instruments
When to use something else
- You only need a single index of agreement, where Lin's concordance correlation coefficient summarises it
- Ratings are categorical, where Cohen's or weighted kappa applies
Assumptions and requirements
- Differences are approximately normally distributed
- Bias and spread of differences are constant across the measurement range
- Paired measurements on independent subjects
What the analysis reports
- Mean difference (bias) with 95% CI
- Lower and upper 95% limits of agreement with CIs
- Proportional bias regression slope and p value
- Normality check of the differences
- Difference-versus-mean plot
How to report it
The mean difference was [value], 95% CI [[value], [value]], with 95% limits of agreement from [value] to [value] (N = [value]); the regression of differences on means gave a slope of [value], p = [value].
Related methods
Run Bland-Altman Agreement Analysis on Quorant
- Data you need
- a table (CSV or Excel) with item columns (e.g. questionnaire items)
- Price
- 250 credits for one analysis up to 10,000 rows (complexity level 2); larger or wider data costs more. Optional methodologist review: 1200 credits.
- Checking
- Checked by property tests in every deployment
- Before you pay
- A free pre-analysis diagnosis checks your data and gives an exact quote; nothing is deducted until you approve it.