The article “UpStream: Motivating Water Conservation with Low-Cost Water Flow Sensing and Persuasive Displays” by Stacey Kuznetsov and Eric Paulos presents the results of research in a number of approaches to water conservation encouragement in both public and private environments. The goal of the project was not only to decrease water consumption at the point of deployment of simple, affordable ecofeedback tools, but also to raise awareness about water consumption, such that people might also consider this issue at other points in their day. Often what ecofeedback strives to do is a bit of a reversal of automation for sustainability (e.g. sensor that turn off lights when there is no movement in a room), preferring instead to have the decision to consume less coming from within us. I would argue that how we can do this with some consistency is the main focus of research in this field today.
Kuznetsov and Paulos present both ambient and numeric modalities for the sink and shower in this paper, using simple prototypes. Interestingly, to keep the cost of the prototypes down, they opted to use sound to sense the volume of water consumed, rather than with a more conventional flow monitors (as seen with ShowMe).
The pilot study, conducted in two public bathrooms, and on one of several shared showers in a female dormitory on a college campus, showed that the 11 survey participants where not previously very knowledgeable about water conservation issues, but had previously sought to reduce waste for a variety of reasons. Interestingly, at the sink, the study demonstrated a significant increase in the amount of water used during the study period. Not only did more people use the faucet, but they deployed it for a longer periods of time. Conversely, the were able to show a decrease in time spent in the shower after the introduction of the feedback device. However, given that there was so little use of the shower once the ambient device was installed, these findings are not significant.
In the prototype, the ambient faucet tool provided something of a stop-light analogy to hand washers. The light was green so long as they were consuming less than average amounts of water, yellow when on par, and red when consuming above average. When one consumes well above average, the red light flashes angrily.
In the shower, users were presented with an LED bar graph which depicted both cumulative and individual water usage. Low, average, and high individual usage was again depicted with colours akin to those found on a traffic light.
The surveys conducted with participants showed that they were very interested in the pieces such that they wanted to play with them – which is what likely lead to the measured increases in water use over the period of study. This is likely due to novelty, and would likely decrease over time. Some awareness appears to have been raised through the project – however, it’s not clear how much this can be tied to the pre-test survey or the prototype itself.
With the findings of the prototype launch, the researchers were able to make some design changes and deploy the devices in a more private setting – the bathrooms of 3 households. The six participants were able to experience both ambient and numeric feedback, each for about a week. Most surveyed initially believed they would prefer the numeric feedback tool, but it was the ambient displays that seemed to be more effective in producing a significant decrease in water used in the shower (good news for the Waterpebble!). The post-test survey supports this as well, in that people did enjoy the ambient display more than the numeric, despite their initial beliefs – they found it less stressful and more aesthetic. Participants also reported having talked about the piece with their roommates and also suggested that it had encouraged them to undertake other pro-environmental behaviors, such as recycling. However, no one was able to recall their average amount of water consumed in the shower, suggesting that the numeric display didn’t leave a lasting impression (other then the sense of guilt a few participants felt while watching the numbers climb).
This study shows, that even with a small sample size you can document a wide range of experiences with a few simple devices. It would have been interesting to see how committed these participants were/are to the environment, and to see if this might have an impact on their interaction with such tools. How much might a participant’s “stage of change” affect their interaction with such tools… and how can we design for such variability in willingness to act?
Source:Kuznetsov, S., & Paulos, E. (2010, April). UpStream: motivating water conservation with low-cost water flow sensing and persuasive displays. InProceedings of the SIGCHI Conference on Human Factors in Computing Systems (pp. 1851-1860). ACM