Methodology
We believe that the work of researchers should be as transparent as possible. This contributes both to the credibility of the research and to the advancement of science. We encourage the use of data for research and public purposes. We publish anonymised data sets, questionnaires and reports for each piece of research on our GitHub page. All materials on this site are available under Creative Commons Attribution 4.0 International license. Citation of The Chronicles Project and website https://chronicles.report is required when citing.
The polls are conducted by Computer Assisted Telephone Interviewing (CATI). To generate the sample, ranges of Rossvyaz mobile phone numbers were used and random phone numbers (Random Digit Dialing) were generated according to the following algorithm.
From the first to the eighth chapter of the Chronicles (February‒October 2022), stratification was conducted by five enlarged federal districts: the Centre (Central Federal District, Volga Federal District, Southern Federal District, North Caucasian Federal District, North-Western Federal District), the Urals (Urals Federal District), Siberia (Siberian Federal District), the East (Far Eastern Federal District), and the Capital Cities (Moscow, Moscow Region, Saint Petersburg, Leningrad Region). In each stratum, the numbers were selected at random. The sample was quoted by strata: Center — 45%, Urals — 19%, Siberia — 12%, East — 4%, Capitals — 20%.
East
Center
Urals
Capitals
Siberia
In the ninth chapter of the Chronicles (February 2023), stratification was carried out in eight federal districts and two federal cities. In each stratum, numbers were selected at random. The sample was quoted by strata: Central federal district (excluding Moscow and the Moscow region) — 14%, Volga federal district — 20%, Southern federal district — 11%, North Caucasus federal district — 6%, Northwestern federal district (excluding St. Petersburg and the Leningrad region) — 4%, Urals federal district — 8%, Siberian federal district — 11%, Far Eastern federal district — 5%, Moscow and the Moscow region — 15%, St. Petersburg and the Leningrad region — 6%.
FEFD
CFD
SibFD
UrFD
Spb + reg
Moscow +reg
VFD
NFD
SFD
NCFD
At the beginning of the interview, the citizenship and age of the respondents was requested. Only Russian citizens aged 18+ were interviewed. As presented in Table 1, a total of 9 chapters were conducted, samples of each chapter include 1600‒1800 respondents. In two express interviews, 1000 and 800 respondents were conducted.
Table 1 — Dates of the poll and sample size
In each chapter, we observe a shift in the realized sample relative to the general population distribution from Rosstat (see Table 2), so the sample was weighted by gender, age and type of locality (urban/rural) according to official statistics. It should be noted that there was a sharp drop in the proportion of young men in the October survey, when mobilization was in full swing - this may be due to the reluctance of young men to answer calls from unfamiliar numbers.
Throughout all measurements we observe a decline of response rate. The proportion of refusals from potential respondents increased in the September express measurements, which may be related both to the mobilization announced the day before and to the use of a different dialing system (DEX) in these surveys. The highest rejection rate was at the beginning of February 2023, but the use of DEX must also be taken into account here in contrast to the other chapters where we used CATI.center.
Proportion of completed questionnaires from potential respondents (Response Rate, RR) decreased. We assume that this may be due, firstly, to the length of the questionnaire because with each successive chapter the number of questions in the questionnaire was increased. Operators warned respondents about how long the poll would take, and in the case of longer questionnaires respondents were more likely to refuse at the very start of the interview. Secondly, the decrease in RR can be attributed to an increase in the number of dialed numbers. The proportion of potential respondents increases with it, and as the sample size does not change (1600‒1800), the proportion of completed questionnaires in the total number drops. The number of dialed numbers is increasing due to the fact that mobile operators are increasingly putting spam filters on calls from unknown numbers.
The number of successfully completed questionnaires relative to the number of conversations ranged from 4% (February 2023) to 18% (February 2022). Unsuccessful conversations include those interrupted at the initiative of the respondent (refusal to participate in the poll immediately or interruption in the course of the poll) or by the operator due to the respondent not meeting the selection criteria (for example, when the respondent is under 18, is not a Russian citizen or has poor Russian language skills).
Gender, sex, location for weight (before weighing)
Table 2 — Distribution of gender and age structure by urban and rural areas in surveys before weighing the array and Rosstat data for the beginning and middle of 2022.
Table 3 — Dynamics of response rate
* We consider the following call statuses as potential respondents: Addressee is busy, Addressee does not answer, Questionnaire not accepted, On redial, Not suitable for the task, Rejected, Call rejected, Interrupted, Complete questionnaires.