Methodology
We publish how every figure on this site was obtained, so that readers can judge for themselves how far to rely on it.
Survey mode
UPOT (Ukrainian Public Opinion Tracking) is conducted monthly using computer-assisted telephone interviewing (CATI). Mobile numbers are generated at random, so every subscriber of a Ukrainian mobile operator has a chance of being selected.
The target population is residents of Ukraine aged 18 and over in all regions except the temporarily occupied territories and areas without stable Ukrainian mobile coverage. In June 2026 the average interview lasted about 18 minutes.
Sample and margin of error
| Wave | Respondents |
|---|---|
| February 2026 | 1,200 |
| March 2026 | 1,503 |
| April 2026 | 1,503 |
| May 2026 | 1,503 |
| June 2026 | 1,500 |
| July 2026 | 1,502 |
| August 2026 | 1,505 |
| September 2026 | 1,503 |
For a sample of 1,500, the margin of error does not exceed ±2.5% at the 95% confidence level (±2.8% for the February wave). This estimate does not account for the weighting design effect, so the actual error is somewhat larger. For subgroups, such as the residents of a single region, the error is considerably larger.
Weighting
As in any telephone survey, some groups respond more readily than others: in the unweighted sample young people are under-represented and older people over-represented. To correct for this, the data are weighted by age and settlement type to the structure of the adult population according to the State Statistics Service at the beginning of 2022. The pre-war structure is used for comparability with earlier research; no up-to-date population data exist for the period after 2022.
The eastern regions are under-represented compared with before the war, as much of their territory is occupied or lacks stable mobile coverage.
How the indicators are calculated
- Trust in institutions. Respondents rate trust on a scale from −3 to +3. Trust means answers from +1 to +3; distrust, from −3 to −1. Net trust is the difference between the two. “Don't know” answers are excluded.
- Consumer confidence index. The mean of four assessments on a −3 to +3 scale, rescaled to −100…+100: change in household finances over the past 12 months, expected change over the next 12 months, whether now is a good time for major purchases, and expected spending on major purchases. Calculated for respondents who answered all four questions. Measured since April 2026.
- Statistical significance. We call a month-on-month change significant if it exceeds random fluctuation at the 0.05 level. When many indicators are compared at once, we apply a correction for multiple comparisons.
Limitations
The survey does not cover people without a mobile phone, residents of occupied territories or Ukrainians abroad. Telephone survey response rates are low (6.2% in June 2026), which is common for the method but may bias results towards people more willing to talk. Answers to sensitive questions may be affected by social desirability.
Fieldwork
Interviews are conducted by Rating Group on behalf of ITCCS. The questionnaire, analysis and conclusions are prepared by ITCCS Ukraine.
Methodological research
Before launching the monitoring, we experimentally compared two scale formats. See “How we chose our scale for measuring trust”.