Why Muscovites Avoid Drinking Tap Water: An Analysis of Determining Factors

I run an online survey of Moscow residents. Participants are recruited through a professional survey platform ensuring that the sample reflects the population of Moscow in terms of basic demographic characteristics. Respondents are randomly assigned either to a control group or to an information treatment related to tap water quality in the city. To be eligible, participants must be adults (age 18+) currently living in Moscow.

Participants are recruited through an online survey platform and consist of adult residents of Moscow (age 18+). Respondents first answer questions about their socio-economic background (such as gender, age, education, income, household composition, and district of residence). They are then asked questions about their drinking water consumption habits, including how frequently they drink different types of water at home and how much bottled water their household purchases. The survey also includes questions about perceptions of changes in tap water quality over time and an open-ended question asking respondents what comes to mind when they think about drinking tap water in Moscow.
Respondents are then asked blocks of questions designed to measure potential mechanisms behind tap water avoidance. These include questions about trust in public services, personal experiences with water quality problems in the past 12 months (such as taste, smell, color, or sediment), perceived sources of water quality problems (for example water treatment plants or pipes), and questions about when respondents stopped regularly drinking unfiltered tap water.
After these sections, respondents are randomly assigned either to a control group or to a treatment group. In the treatment group, participants receive an informational message explaining how Moscow’s tap water is purified and describing improvements in water quality. The control group does not receive this information. Afterward, respondents answer questions measuring perceived safety of tap water and their willingness to drink or recommend tap water in the future.

2026-03-18

2026-03-19

The primary outcomes are respondents’ perceived safety of tap water in Moscow, their stated willingness to increase consumption of unfiltered tap water, and their willingness to recommend drinking tap water to others.

The main outcomes are based on respondents’ answers to post-treatment questions about tap water safety and intended behavior. Perceived safety of tap water is measured using a categorical survey question asking how safe respondents consider tap water in Moscow. Intended behavior is measured using questions about the likelihood of increasing tap water consumption without additional treatment and the likelihood of recommending tap water to others. These outcomes will be analyzed both as categorical variables and as binary indicators capturing whether respondents express positive perceptions or willingness to increase tap water use.

I am interested in heterogeneity in responses to the informational treatment across several mechanisms that may explain tap water avoidance. In particular, I examine whether treatment effects differ depending on respondents’ trust in public services, their personal experiences with water quality problems, and indicators of habit formation related to tap water consumption.

Trust in public services is measured using respondents’ ratings of the quality and reliability of several public services in Moscow. Local infrastructure experiences are measured using questions about the frequency of water quality problems observed at home in the past 12 months and respondents’ perceptions of where such problems originate (for example water treatment plants or pipes). Habit formation is measured using questions about when respondents stopped regularly drinking unfiltered tap water. These variables will be used to examine heterogeneous responses to the informational treatment.

I randomize the information provided to respondents in an online survey. Participants are assigned either to a treatment group receiving information about tap water quality in Moscow or to a control group that does not receive this information.

The study is implemented as an online survey experiment with two groups: a treatment group and a control group. Respondents assigned to the treatment group receive an informational message about the purification process and quality of tap water in Moscow, while respondents in the control group do not receive this information.
The task of collecting responses is outsourced to Yandex Vzglyad, an online survey platform operated by Yandex, a large technology company that provides access to a broad pool of registered survey participants. Registered users have accounts on Yandex or affiliated services and have provided basic demographic information such as age, gender, and location. Participation in surveys is voluntary. To be eligible for this study, participants must confirm that they are at least 18 years old and currently reside in Moscow.
The survey platform allows targeting respondents based on demographic characteristics. The sample is collected to broadly reflect the demographic structure of Moscow residents. During the fielding of the survey, the platform monitors demographic characteristics and response rates in order to maintain balance across key characteristics.
Participants are assigned to either the treatment or the control survey version, ensuring that exposure to the informational treatment varies across respondents.

Respondents are assigned to either the treatment or control survey version through the survey platform at the moment of participation.

Randomization is at the unit of individuals.

No

0

400 individuals

200 people in treatment, 200 in control group