Sample handling matters: if a sample is contaminated or degraded before testing, the test gives you the result for the degraded sample, not for fresh material. That is not a failure of the test; it is the test working correctly.
What a simple presence test is genuinely useful for — that is the question, and I have not found it answered plainly anywhere I have looked.
Trying to work out what would count as evidence on simple presence test, before collecting any. This is the part I usually skip and it is the part that makes the rest useful.
If two explanations predict the same observation, observing it does not help. So: what observation would separate them?
False positives and false negatives: no test has 100% sensitivity and 100% specificity. A negative result does not guarantee absence; a positive result does not guarantee presence. The predictive value depends on prevalence.
That is the shape of it. The detail is where I would expect to be corrected.
Sample handling matters: if a sample is contaminated or degraded before testing, the test gives you the result for the degraded sample, not for fresh material. That is not a failure of the test; it is the test working correctly.
| Topic | Participants | Replies | Views | Activity |
|---|---|---|---|---|
|
Cost versus information: making the trade-off explicitly
Cost versus information: making the trade-off explicitly Writing it up because I had to work it out twice and would rather nobody else did. Asking about cost versus information directly, because I have read…
|
+2 | 6 | 12k | 11h |
|
A comparison of home results against a laboratory result on the same vial
Posting this under the heading it deserves: A comparison of home results against a laboratory result on the same vial Everything below is what sits behind that. Comparison of home results against: setting out…
|
+61 | 65 | 35k | 8mo |
|
A comparison of home results against a laboratory result on the same vial — does this still hold?
A comparison of home results against a laboratory result on the same vial — does this still hold? I have a specific reason for asking rather than idle curiosity, and the context is below. Comparison of home…
|
+2 | 6 | 15k | 11h |
|
What I would want from a home test that does not exist yet — what changed since
Asking directly, because I could not find a straight answer: What I would want from a home test that does not exist yet — what changed since Two reports on the same lot, and I would like help deciding whether…
|
+101 | 111 | 1.9k | 21h |
|
Reading a home test result without over-claiming — what changed since
On the subject in the title: Reading a home test result without over-claiming — what changed since Working notes rather than a conclusion. What changes if the standard account of Reading a home test result is…
|
2 | 12k | 6mo |
| Topic | Participants | Replies | Views | Activity |
|---|---|---|---|---|
|
About the COA interpretation category
What a certificate of analysis commits its issuer to, which fields matter, and how to spot one that says almost nothing. This post is a community wiki: any member at trust level 3 or above can edit it, and…
|
+8 | 12 | 8.5k | 2mo |
|
Revisiting: Spike-and-recover accuracy for a peptide assay
Revisiting: Spike-and-recover accuracy for a peptide assay — setting out what I have, and where I think it stops being reliable. I was wrong about Spike-and-recover accuracy in a thread last spring and I…
|
2 | 6.8k | 21h | |
|
Repeatability versus intermediate precision
Repeatability versus intermediate precision — setting out what I have, and where I think it stops being reliable. Trying to establish the boundary conditions on repeatability versus intermediate precision.…
|
+119 | 134 | 34k | 13mo |
|
Splitting a vial for two laboratories: how to do it without ruining both
Splitting a vial for two laboratories: how to do it without ruining both — setting out what I have, and where I think it stops being reliable. Something about splitting a vial does not reconcile and I would…
|
+64 | 70 | 4.4k | 13mo |
|
What a simple presence test is genuinely useful for — does this still hold?
What a simple presence test is genuinely useful for — does this still hold? I have a specific reason for asking rather than idle curiosity, and the context is below. Trying to work out what would count as…
|
2 | 22k | 16mo |