Validation

This bundle takes a harder line on sourcing than most, because the subject demands it. Nearly all popular writing on planning and goal setting is produced by companies selling goal-setting software or planning consultancy. Several of its most repeated statistics turn out to have no underlying study.

The three grades

These are the categories the pages sort practices into. They are stated in the prose beside each claim rather than rendered as labels. Read the sentence around a claim to see which applies. Where a practice is convention rather than evidence, the page says so in the same sentence that recommends it.

Evidenced. A meta-analysis, a randomised experiment, or a peer-reviewed study with a disclosed method and sample. Where an effect size exists it is given, and where the study population differs from a workplace team that is stated rather than glossed.

Convention. The practice traces to a named person's judgment, or to one organization's internal habit, and has not been tested. This is not a criticism. Some conventions are well reasoned and worth adopting. It is a statement about what kind of confidence is available.

Folklore. Widely circulated, and traceable either to a source that does not support it or to nothing at all. These are named explicitly rather than quietly omitted. A reader is likely to meet them elsewhere and deserves to know why they are absent here.

Sourcing rules specific to this bundle

The originating text, not its populariser. Several practices in this field are attributed to a founder and prescribed by a later author who changed them. Where a claim is attributed to an original management text, it has been checked against that text.

Named arguments stay named arguments. Some of the most useful material here comes from practitioners making an argument rather than reporting a study. Those arguments are cited as arguments, and labelled as unvalidated, however influential the author.

Vendor material is labelled every time it appears. A company that sells planning software is a legitimate source for its own product's documentation and a poor one for whether the practice works.

A statistic with no traceable method is left out rather than softened. Where a widely quoted figure has been traced and found empty, the trace itself is reported. Knowing a number is hollow is more useful than never having heard it.

What this does not guarantee

Grading is not the same as being right. A claim graded Evidenced can still be poorly transferred. Several of the strongest studies here measured individuals rather than teams, and that gap is an inference every time it is crossed. The grade describes the quality of the underlying source, not the safety of applying it to your situation.