TC-02 SPRITE (Sample Parameter Reconstruction via Iterative TEchniques)
Reconstructs possible datasets that match reported descriptive statistics and produces plausible samples showing the range of possible item distributions.
What it does
Reconstructs possible datasets that match reported descriptive statistics. Artefact: plausible samples. Read the range of possible distributions of item values.
When to use it
Use when a report gives only a mean, standard deviation and sample size and you want to know what data could have produced them.
When it does not fit
The abstract does not state the inputs; mean, standard deviation, sample size and scale range are assumed. Code exists in MATLAB, R and Python.
Result format
The result gives the reported descriptive statistics, and one or more plausible samples that match them, each listing the item values. It ends with the range of possible distributions of item values across the samples.
Requirements on the material
| Property | Requirement type | Requirement |
|---|---|---|
| Content type | Always applies | requires=text; requires=tabular |
| Unit of analysis | Always applies | requires=dataset |
| Variable types | Raises a warning | prefers=ordinal; prefers=counts or fractions |
| Level of detail available | Always applies | requires=aggregate tables |
| Appraises source credibility | Applies in some cases | no |
Not restricted on: Structure, Time structure, How the sample was chosen, Number of sources, Whether sources can be joined, Amount of material, Known data problems, Producer interest, Source dependence, Origin, Document genre, Language, Personal information, Security classification, Licence.
| Minimum material | Reported mean, standard deviation, sample size and scale range for each reconstructed measure. |
|---|---|
| Additional sources needed | none |
| Not for (rules) | none |
| Not for (text) | Wrong choice when any of the four inputs is missing or when the measure is not a bounded scale. A reconstructed sample shows only that data are possible and is not the real data. |
Choosing and ordering
| Kind of technique | Verification technique, runs in step 6 |
|---|---|
| Phase tag | validate |
| Other phase tags | verify and prepare |
| Kind of finding | contradiction; risk |
| Question answered | what |
| Strongest statement it supports | describe |
| Needs as input |
|
| Makes as output |
|
| Must exist first | None |
| Loop target | proceed with limit stated |
Library checks by stakes
Each stakes level includes the checks of the level below it. Checks that need a person apply at high stakes only, so the process leaves them out at lower stakes (decision 38). Repeatability: repeatable with fixed settings.
| Stakes | Check | Person involved | Counts as a different method |
|---|---|---|---|
| exploratory | input validation | none | no |
| decision-informing | input validation | none | no |
| decision-informing | code execution for arithmetic | none | no |
| decision-informing | spot check | optional | no |
| high-stakes | input validation | none | no |
| high-stakes | code execution for arithmetic | none | no |
| high-stakes | spot check | optional | no |
| high-stakes | independent review | needed | no |
Failure modes and limits
| Model suitability | model plus code |
|---|---|
| Typical model failure | Presents one reconstructed sample as the real data, or accepts a sample that breaks the scale limits. |
| Error sources | input data; statistical assumptions |
| Certainty limits | Many samples fit the same statistics, so a reconstruction shows possibility and not what happened. Certainty drops when rounding is unknown or the scale range is unstated. |
Related techniques
- should follow TC-01 GRIM test (granularity-related inconsistency of means): the mean must first be shown to be possible
- pairs with TC-03 statcheck (recomputation of reported p-values): statcheck tests the reported test statistics that depend on the same descriptives
Evidence and tag basis
Documented. Description only, from a preprint. (Heathers et al., PeerJ preprint 26968)
Tag basis: input profile inference; selection inference; sequencing inference; robustness inference. All tag values are inference unless a source is named.