MPG/Gallup International Poll of September 25, 2026: Examining an Anomalous Jump in the Numbers

1. What Was Published on September 25

Aram Navasardyan, head of the Armenian branch of the Gallup International Association (the company MPG), presented the results of a telephone survey conducted September 22–24, 2026 (sample size 1,001). Approval of the government’s work stood at 55.9% of respondents (25.1% “fully approve” + 30.8% “somewhat approve”), approval of parliament’s work at 41.5%, and personal approval of the prime minister at 53% (31% “fully approve” + 22% “somewhat approve”), against 29.3% who fully disapproved. On the question of whether the June parliamentary elections were fair, 52.7% said they were, while 39.4% believe there was fraud. Sources: Arminfo, EADaily. The same day, MPG also published data on Russian military base No. 102: 64.4% favored keeping it, under 30% favored its withdrawal (Izvestia, Caucasian Knot).

The official accredited source — MPG’s own website — describes itself as “Armenia’s most frequently quoted research company by media” and clarifies that the Gallup International Association is a Swiss non-profit association of national institutes, not a division of the American Gallup Inc. (gallup.am). The use of the recognizable “Gallup” brand in publication headlines, however, contributes to the perception of the figures as more authoritative than the company’s formal status warrants.

2. Timeline of Previous Measurements — Starting with Independent Observers

February 2026, IRI (International Republican Institute, USA). The survey was conducted February 3–13 by Breavis on behalf of IRI (sample size 1,506, margin of error ±2.5%). Among “likely voters,” support for Civil Contract stood at roughly 29–30%, and personal trust in Pashinyan was 20% (18%+2%); 47% said the country was headed in the right direction. Sources: IRI — official report (PDF), Public Radio of Armenia.

May 19–21, 2026, MPG/Gallup. Among decided respondents, support for Civil Contract stood at 28.8%, with 8.6% refusing to answer and 3.4% undecided. Source: Panorama.am.

May 23, 2026, IRI. Personal trust in Pashinyan measured 29% (27%+2%). Source: ARMENPRESS.

June 7, 2026 — Elections. The Central Election Commission recorded 49.81–49.89% for Civil Contract at 59% turnout. Sources: ARMENPRESS, Wikipedia.

September 22–24, 2026, MPG/Gallup. Government approval at 55.9%, personal approval of Pashinyan at 53%, elections deemed fair by 52.7% (see Section 1).

Date	Source	Metric	Value
Feb. 2026	IRI (independent)	Civil Contract support among likely voters	~29–30%
Feb. 2026	IRI	Personal trust in Pashinyan	20%
May 2026	MPG/Gallup	Civil Contract support among decided voters	28.8%
May 2026	IRI	Personal trust in Pashinyan	29%
June 2026	CEC (official)	Votes for Civil Contract	49.81–49.89%
Sept. 2026	MPG/Gallup	Approval of Pashinyan’s work	53%
Sept. 2026	MPG/Gallup	Elections deemed fair	52.7%

Over the same four-month period during which independent observer IRI recorded a rise in trust in Pashinyan from only 20% to 29%, MPG’s figures rose from 28.8% (May party support) to 53% (September personal approval of the prime minister) — a jump roughly double the pace recorded by the independent source. MPG does not publicly disclose the methodology behind how the base was calculated (the share of refusals and undecided respondents) in the September survey.

3. The Border Demarcation Issue — A Factor That Should Have Lowered, Not Raised, Approval

At the same time the September poll was published, Armenia’s border delimitation and demarcation process with Azerbaijan continues, including its most sensitive point — the fate of the village of Tigranashen. Former parliament speaker and ex-prime minister Khosrov Harutyunyan has publicly warned that ceding Tigranashen to Azerbaijan would create serious risks for transport links to the country’s southern regions and could have long-term consequences for its security; he expressed similar concerns about the village of Bardarluh. Sources: Sputnik Armenia, yerkramas.org, Voice of Armenia. In September 2026 the two sides agreed to begin the next, more difficult phase of demarcation — the northern section starting from the point where the borders of Georgia, Armenia, and Azerbaijan meet (EADaily). Territorial concessions are among the most sensitive issues for Armenian society, and logically this ongoing process — unfolding in parallel with the poll — should have been pushing approval figures downward rather than upward.

4. The Conflict with the Armenian Apostolic Church — Another Factor That Should Have Reduced, Not Increased, Trust

Since summer 2025, tensions have been escalating between Pashinyan’s government and the leadership of the Armenian Apostolic Church (AAC). In January 2026 the prime minister, together with ten bishops, signed a church “renewal” plan calling for the removal of Catholicos Garegin II, and on February 14 a criminal case was opened against him. On August 28, 2026, an unprecedented trial for the post-Soviet space began against the Catholicos and six bishops on charges of failing to comply with a civil court ruling — carrying a potential sentence of up to two years’ imprisonment; the church itself called the charges an illegal interference in its internal affairs the following day. Sources: RBC — chronology of the conflict, mlyn.by, EADaily, Vedomosti. The church has historically remained one of the most trusted institutions in Armenia — earlier IRI polling consistently found the armed forces and the church among the most respected institutions in the country (jam-news.net). Direct confrontation between the government and this institution — the criminal prosecution of the Catholicos and several archbishops — represents another factor that would objectively be expected to undermine, not strengthen, trust in the government during the very weeks when MPG recorded rising approval.

5. Media Coverage: Russian Outlets Reacted Far More Widely Than Armenian Ones

The results of the September poll almost immediately appeared prominently across major Russian outlets and news agencies: Lenta.ru, Izvestia (citing the “MIC Izvestia” agency), EADaily (citing RIA Novosti), 5-tv.ru, OSN, Mail.ru News. Most of these publications cite one another and the original source “Sputnik Armenia” — a Russian state-affiliated media outlet — rather than a direct MPG presentation. By contrast, notably fewer Armenian outlets covered the same poll — chiefly Arminfo itself and the portal yerkramas.org (the latter with a critical commentary on the figures themselves). Such a disparity in coverage — for what is, in essence, an internal Armenian political poll — is itself a significant fact: it suggests the publication resonated more within the Russian information space than the Armenian one, a point worth weighing when assessing whose interests this particular wave of coverage may have served.

6. Conclusion

The timeline of independent and domestic measurements shows: the rise in government approval figures recorded by MPG/Gallup in September represents a sharp jump — roughly double the pace recorded by independent observer IRI over the same period. This jump occurred during months when the government was simultaneously pursuing an unpopular process of territorial concessions (Tigranashen, demarcation with Azerbaijan) and criminally prosecuting the leadership of the Armenian Apostolic Church — two factors that, by the logic of public opinion, should have reduced rather than increased trust in the government. Additionally striking is the disproportionate media resonance: the publication received substantially wider coverage in Russian media than in Armenian media. Taken together, these circumstances are not direct proof that the data was falsified, but they are sufficient grounds to demand that MPG disclose its full sampling methodology before the September figures are treated as an accurate reflection of genuine Armenian public sentiment.

Artatsolum