What do I use
Government, parastatals and regulators
Policy analysis, service delivery monitoring, regulatory compliance, statistical reporting.
Your question, and what answers it
Monthly figures move constantly. A chart shows which movements mean something, which stops policy being made about randomness.
In ZStat: open your study, go to Analysis, and choose control chart.
Time series, and a forecast that is validated on later data rather than on a shuffled sample.
In ZStat: open your study, go to Analysis, and choose arima.
With denominators checked first, because most disputes about a rate are disputes about the denominator.
In ZStat: open your study, go to Analysis, and choose one way anova.
Whether the pattern across the map is real clustering or coincidence, before any resource is targeted on it.
In ZStat: open your study, go to Analysis, and choose morans i.
With the Hausman test choosing the specification rather than an assertion in the methodology section.
In ZStat: open your study, go to Analysis, and choose panel regression.
Where the policy applied to some and not others, and randomisation was never possible.
In ZStat: open your study, go to Analysis, and choose propensity matching.
Whether performance can meet a target consistently, not whether it did last quarter.
In ZStat: open your study, go to Analysis, and choose process capability.
What goes wrong in government, parastatals and regulators
- Publishing a figure without the method attached. Every result here carries the digest of the data behind it, which is what turns a contested statistic into a checkable one.
- Comparing rates across areas with denominators defined differently.
- Inferring what individuals do from what districts average. The relationship can reverse entirely.