Information Nudges and Consumer Willingness to Invest in Microplastic-Free Water

Treatment groups will receive detailed information about microplastics in drinking water while the control group will receive a basic definition of the term.

This field will remain hidden until the completion of the study, and will then become available to help people searching the registry understand what you did in more detail. Describe your study in more detail here.

Treatment and control groups will receive information about microplastics in varying degrees of specificity. The control group will receive a simple definition of microplastics to provide context for the decisions they will be asked to make for the remainder of the survey. The treatment group will be shown a tailored information nudge (an infographic) with the headline “You might be drinking plastic” containing a definition of microplastics as well as information about their potential health effects and a prompt detailing how much microplastic the respondent could be consuming depending on their primary water source, based on information from Senathirajah et al., 2021. The infographic also contains an image of microplastics to further contextualize the information provided.

2025-07-21

2025-08-21

The primary outcome we are measuring is willingness-to-pay (WTP) for water filters that remove microplastics/are certified to remove microplastics.

To gauge respondents’ WTP for water filters for microplastic removal, we will use a Discrete Choice Experiment (DCE). The DCE was designed using Stata’s “dcreate” package to generate choice sets with a d-efficient design for maximum efficiency. Respondents will be given a series of choices between two water filters, which vary quasi-randomly by price, type, whether or not they remove microplastics, and whether or not they are certified by the National Sanitation Foundation to remove microplastics. Using a mixed logit model, we will convert the choices respondents made to a willingness-to-pay value. By combining variables for whether respondents received the treatment or control and filters with microplastics removal, we can determine if there is a correlation between receiving the information treatment and WTP for microplastics-removing water filters.

A secondary outcome observed in this study is respondents’ knowledge of and attitudes toward microplastics and tap water quality.

Attitude and knowledge information will be collected in both parts of the survey for different populations. The first survey will ask respondents who already have a water filter that removes microplastics a series of Likert-scale questions about their perceptions of the quality and safety of their home water source. The second survey will ask the rest of respondents the same questions, allowing us to compare public perceptions of water quality across regions, among demographic groups, and between those who own microplastic-removing water filters and those who do not.

The objective of the study is to measure the effect of a personalized information nudge on the willingness to pay for water filters that remove microplastics in drinking water. Respondents will be randomly assigned to receive the nudge or not, and willingness-to-pay will be measured using a discrete choice experiment.

The objective of the study is to measure the effect of a personalized information nudge on the willingness to pay for water filters that remove microplastics in drinking water.

We will launch two surveys on Qualtrics for this study. Respondents will be recruited using survey recruiting platform Prolific. The first survey will identify respondents' primary water source—bottled, filtered, or tap—and identify the segment of participants who already have a water filter that removes microplastics. We will ask these respondents a series of questions about their current perceptions of their tap water safety at home, then remove them from the sample that receives the second survey for screening purposes, in hopes of eliminating bias from the disproportionate disinterest this population may have in purchasing another filter. The population receiving the second survey will include all individuals who consume bottled and tap water as well as individuals without a microplastics-removing water filter at home.

The second survey will collect information on attitudes and knowledge about microplastics, and respondents’ primary water sources. Individuals classified by their primary water sources—tap, bottled, and filtered water—will be randomly assigned to the control and treatment groups proportionally. The control group will receive a basic definition of microplastics, while the treatment group will receive an information nudge with microplastics' health effects and a statement about the potential quantity of microplastics people can be exposed to depending on their water source. After the information nudge intervention, the survey will introduce a series of questions asking participants to select their preferred product based on different water filter's characteristics—type of water filter, range of prices, certification status, and whether it removes microplastics— to estimate individuals' preferences and their willingness to pay for each of them. Finally, the survey will prompt participants for some demographic information. Participants will be compensated with $1 and $3 upon completion of the first and second survey, respectively.

Eligible subjects will be randomized into the treatment groups using Qualtrics’ randomization tool in equal proportions.

Individual

No

N/A

600 individuals

The sample size by treatment arms cannot be determined because an unknown portion of the initial sample will be screened out before being assigned to treatment or control groups. Because these individuals will be screened out before assignment, the remaining individuals will still be evenly distributed between treatment and control groups.

The chosen sample size was determined using Orme’s rule of thumb for discrete choice experiments: N ≥ 500 * (Lmax/JS) Where Lmax is the highest number of levels for an attribute, J is the number of alternatives per choice task, and S is the number of choice tasks per respondent. Given Lmax=6 (the number of levels for the “price” attribute, J = 2 (each respondent receives 2 alternatives per choice task), and S = 5 (each respondent receives 5 choice tasks), the minimum sample size is 600. The final sample size may be recalculated following survey pretests.