Lab QC · All departments

Total Error Allowed Calculator

Easy to use — just follow the 4 steps.
Step 1: Pick your department.  Step 2: Pick your test.  Step 3: Type in your numbers.  Step 4: See if you passed or failed.
Example: the same sample reads 122 on Method A and 124 on Method B — this tells you if that gap is small enough to be OK.

Disclaimer — For laboratory professionals only

This calculator is for education and quality-control checking only. It is not medical advice, and it does not diagnose or treat patients. It does not replace your laboratory's procedures, manufacturer instructions, CLIA/CAP requirements, or your laboratory director's judgment. Always follow your lab's validated procedures.

Words in plain English

You only need these 4 words. They show up everywhere below, with this same plain meaning next to them.

TEa — the limitThe biggest mistake allowed for this test. If your total error is at or under this number, you pass.
Bias — always off in one directionHow far your average result sits from the true answer — like a scale that always reads 2 pounds heavy.
CV — how much results jump aroundRun the same sample 10 times — CV tells you how much those answers scatter. Smaller is steadier.
Sigma (σ) — how much room you haveHow much safety room is left before you start failing. Higher is safer — 6 is excellent, under 3 is risky.

Step 1 — Pick a department

Tap the part of the lab you work in. Big buttons — just tap one.

Step 2 — Pick a test

Now pick the test and where its limit comes from — CLIA or CAP. The limit fills in for you. You can always type over it.

Using CLIA. Limits below are the CLIA proficiency-testing limits for the test you pick.

CLIA
%

Source: CLIA — loaded from the test. Type over it if your lab uses a different limit.

Add your own test

Step 3 — Type in your numbers

Quick check — compare two results. Same sample, two machines. Type the first result (the one you trust) and the second result. Your answer shows right below — that is Step 4 for this quick check.

Reference
units
Compare
units
Yes — within 5%
Method B 124 is 1.64% higher than Method A 122.
1.64%
difference
0%marker = your limit
Difference
+2
Method B minus Method A
% difference
+1.64%
vs. the reference result
Allowed at limit
±6.10
5% of 122
Room left
3.36%
of the limit still unused
Acceptable for Method B: 115.9 to 128.1 — 124 is inside.
(124 − 122) ÷ 122 × 100 = +1.64%

Step 3 — Or do the full check

Full check — bias & CV for two methods. Bias = how far off you are. CV = how much results jump around. Type one set of numbers for each method. It is already filled with an example — just type over it. Your answer is in Step 4, right below.

%
%
%
%
Advanced — coverage factor (Z is 1.65 standard)

Step 4 — See your result

For 5% goal — both methods vs. the same limit, TEa 5.0%.

Enter both methods
Fill in bias and CV for each method above and the head-to-head verdict appears here.
Method A —
—
total error
Sigma σ — safety room
—
Room left
—
Budget used
—

Enter this method's bias and CV to see its breakdown.

Method B —
—
total error
Sigma σ — safety room
—
Room left
—
Budget used
—

Enter this method's bias and CV to see its breakdown.

0%scale: 5%
Bias Imprecision (Z × CV) Marker = TEa limit

Once both methods have numbers, this spells out the difference — how far apart their total errors and sigmas are, and where the gap comes from (bias or imprecision).

What answers are OK? — acceptable range

With the limit at 5% (from Step 2), a result on a known target has to land inside this window. For an exact-match test, the window is the target itself.

units
Lowest OK
—
to
Highest OK
—
units

How the math works — in plain words

The standard total-error model used in lab QC.

% difference = (compared − reference) ÷ reference × 100
Total Error = |Bias %| + Z × CV %
Sigma σ = (TEa % − |Bias %|) ÷ CV %
  • Bias is the steady shift away from the true value. CV is the random scatter, as a percent of the mean.
  • The method passes when total error is at or under TEa. Z = 1.65 is the usual one-sided 95% factor.
  • Sigma says how much room the method has before errors start crossing the limit — higher is safer.
  • In a method comparison, the lower total error wins — and if both pass, the higher sigma is the safer method to run.
  • Not every test is graded by %. Sodium, potassium, and calcium use a fixed amount (like ±4 mmol/L for sodium). Urine dipstick tests are graded on agreeing with the target result. Blood Bank tests — ABO, Rh, antibody screen — must be 100% correct: any mismatch is a fail, so the two-result check asks for an exact match there.
SigmaWhat it means
6.0 +World class — very hard to fail
5.0 – 5.9Excellent
4.0 – 4.9Good — routine QC holds it
3.0 – 3.9Marginal — needs tighter QC
2.0 – 2.9Poor — failures likely
Under 2.0Unacceptable — fix the method
TEa Pro — for labs that need records, not just a result

Basic calculator stays free: all 44 tests, CLIA / CAP limits, bias & CV, sigma and method comparison. Pro adds saved results, PDF reports, history, QC charts, custom limits and the team workspace below.

Pro — 3-day free trial, then $2.99/mo · $29.99/yr · Team $29.99/mo or $49.99/yr per lab

Subscription terms (App Store version): TEa Pro starts with a 3-day free trial. When the trial ends, the plan you picked starts — TEa Pro Monthly is $2.99/month or TEa Pro Annual is $29.99/year — and is billed to your Apple ID unless you cancel before the trial ends. Team is $29.99/month or $49.99/year per lab. Subscriptions renew automatically unless cancelled at least 24 hours before the end of the current period, and can be cancelled any time in the App Store account settings. This web page takes no payment and sells nothing — every Pro feature below is unlocked here so you can try it before subscribing in the app. Prices are in USD and may vary by region. Full details are in Privacy, Data & Terms below.

Save & PDF Report PRO

Save the current head-to-head result, then download a clean PDF report for your records or to share with a supervisor. The PDF is generated on this device — nothing is uploaded. Saved in this session only — close the tab and the list clears.

No saved results yet. Run a comparison above, then save it.

History — is your method getting worse over time? PRO

Every saved result for the test you have picked in Step 2 is plotted here. A falling line means the method is losing room before it starts failing.

Save at least 2 results for the same test to see the sigma trend.

Demo points are labelled “Demo” in the table above when you load them — they are examples, not your lab's data.

Westgard & Charts — daily QC check PRO

Paste recent QC results for one level. The chart plots each point against the mean ±1, 2 and 3 SD, and the rules below flag patterns to review. The flags are a checking aid only — whether a run is accepted or rejected is decided by your lab's own QC procedure.

    Custom Limits — use your lab's own limit PRO

    Your lab's own limits — from biological variation, manufacturer claims, or your director's goals — saved alongside CLIA / CAP. Apply one and it loads into Step 2.

    No custom limits yet — add your first one above.

    Team workspace — on this device PRO · TEAM

    This is a single-device preview of the Team workspace: the shared saved-results table above and the audit trail below, with your role and a director sign-off line for the report. Nothing here leaves this device — there is no login on this page, no account is created, and no result is sent to or shared with anyone. Cross-device sharing between lab members is part of the Team plan ($29.99/mo or $49.99/yr per lab) in the App Store version, where each member signs in; it does not happen on this web page.

    Audit trail

    • —Workspace opened. Actions you take — saves, exports, custom limits, sign-offs — are logged here with time and operator.