What TI-Z's "Integrity Index" Actually Measures, and What It Doesn’t

Transparency International Zambia has moved to correct what it says is a widespread misreading of one of its own election monitoring tools, clarifying that its Integrity Index measures the process behind Zambia's vote count, not an endorsement of the results themselves or of how the Electoral Commission of Zambia has managed the count as a whole.

TI-Z said it has deployed monitors nationwide to collect information on electoral incidents throughout the process, alongside an index comparing results announced by the ECZ at national level against those announced by returning officers at constituency totalling centres, a tool the organisation also used in 2021.

"A notable misunderstanding of the score is that TI-Z has assessed all the presidential results announced thus far by the ECZ," the organisation said. "Another misinterpretation has been that TI-Z is endorsing the entirety of the results management process that the ECZ is following. In view of this, we wish to not only clarify the methodology of the integrity index, but to also call upon stakeholders, including the media, to accurately interpret the index and to engage us directly where any clarification is sought."

WHY THE METHODOLOGY CHANGED MID-COUNT

The index's original design assumed that ECZ's national totalling centre would only receive results from constituency totalling centres after Gen 20 forms had been shared with stakeholders and results had already been announced locally. TI-Z said its monitors found that assumption did not hold in every case.

"Contrary to this assumption, our process monitoring subsequently revealed that in some constituencies, returning officers were instructed to send the Gen 20 forms to ECZ officials in Lusaka for verification before signing and before the announcing of the results was done at the totalling centres in the constituencies," the organisation said, describing it as "a clear departure from the process and procedure that was followed in the 2021 general elections." TI-Z said it wrote to the ECZ to seek clarification on the anomaly but had not received a response as of its statement.

In response, TI-Z said it expanded the index into a weighted measure: 50 percent of the score reflects whether the results TI-Z monitors collected match those ECZ announced, and the other 50 percent reflects whether the counting process itself was free of incidents that could compromise its credibility, such as unsealed ballot boxes in transit, delays pending sign-off from Lusaka, or a failure to display Gen 20 forms publicly at totalling centres.

THE NUMBERS SO FAR

Applying the weighted index to 53 of the 105 constituency presidential results the ECZ had announced at the time, TI-Z recorded a 96.9 percent Integrity Index.

All 53 results matched what TI-Z's own monitors had independently collected, but the organisation identified process-related incidents in seven of those 53 constituencies, the gap responsible for the score falling short of 100 percent. TI-Z said it would continue applying the index as ECZ announces further results, and would issue a final weighted score once all 226 constituencies have been declared.

WHAT THE SCORE DOES, AND DOES NOT, TELL VOTERS

The distinction TI-Z is drawing matters for how the figure should be read. A 96.9 percent Integrity Index does not mean 96.9 percent of Zambia's results are correct, nor does it mean TI-Z has verified or endorsed the entire results management process.

It means that among the specific results TI-Z has been able to independently check so far, the numbers match what ECZ announced, while flagging that in roughly one in eight of those same constituencies, something about how the count was handled fell short of the standard TI-Z is measuring against. The organisation is urging media houses and other stakeholders to reflect that distinction accurately, rather than treating the headline percentage as a verdict on the election as a whole.