TUT 08/06/2012 - 08/10/2012

From OpenWetWare

TUT Weekly report – 08/10/2012

When the first participant was shown the calendar, she was confused why usage information was not displayed for all the days instead up to the end of the billing period. The black marker, around the 21st, was not a clear enough indicator to the participant that the calendar signified a hypothetical situation—where today was supposed to be the "21st." This lead me to set up a scenario for the remaining participants after the introduction text. I explained to them that the calendar represented a hypothetical situation and today was not the 10th of August. This confusion most likely occurred because we are currently in the month of August, whereas it was easier for participants to assume a hypothetical situation during the month of July. While I was setting up the scenario, I thought it might be best if we come up with an introduction similar to the one we have to the Smith's house in the IHD simulation. One question that came to mind was if we were setting up the situation for the participant's own usage or for a hypothetical family like the Smith's.

The information conveyed by the calendar was well received by all the participants. One of the ideas that was brought up was to be able to choose whether you see price, kWh, or a combination of the two on the calendar. This idea might be good, however, seeing the cost and kWh consumption will make the relationship between cost and use more salient to the customer. It also might prove to be an effective way for customers to learn about electricity use. I recognize that we are still testing whether that assumption is true or not. One of the participants expressed that she really liked the color scheme and that it helped her differentiate usage rates and their relation to high and low usage days. However, one common question that keeps on reoccurring is how the high, normal, and low are calculated. Prior to any explanation, participants have guessed that it was a comparison to last month or social comparison to their neighbors; but as they walk through their thoughts, they conclude that they have no idea. While adding excessive amounts of information is a line that we continually tread, I think it would be important to explain how the thresholds are chosen. I tried using the summary table last time, but it was not intuitive that the numbers on the table were associated with the colors saw on the calendar or the speedometer. I have thought about color coding the numbers on the table to help facilitate the association.

Feedback on the speedometer, in general, was better than our previous attempts. Contrary to our previous interviews, two out of three of the participants found that having no numbers on the speedometer to be confusing. While they liked the simplicity and spaciousness of the design, they proposed that a number for the threshold between normal and high use be added. One of them even suggested that we add numbers to the beginning and end of the speedometer. I propose that we test this by having two speedometer images side by side so that they can actively compare the two designs. The participants thought of the speedometer in a favorable manner and were able to associate the colors with the usage shown on the calendar. One concern that I had was that all three participants associated the dial on the thermometer with the average rate of usage number below. The relationship between the average rate of usage and daily consumption is the same, so it might be best to advertise the speedometer as the average usage rate rather than the daily cumulative consumption.

I added a line graph that displayed the daily average usage rate. Participants found that it was easier to recognize patterns of usage on this display compared to the calendar. Despite that fact, most participants have been able to recognize that there was a pattern just by looking at the calendar. Instead of having a horizontal line representing median usage, I made the line represent the ceiling usage rate between high and normal usage. The title of the graph mislead the participants to believe that the line was the average usage rate for the month, and only one of the participants were able to see that the line was too high on the graph to be the average. Even after being told that it was the ceiling, the participants were not able to inference the relationship between the maximum daily average usage rate and the high usage color schemes they had seen. Some of the participants were in favor of the line graph because it helped them identify the patterns in their usage. They even proposed that you can toggle between hourly, daily, monthly, and yearly usage rates on the line graph. One suggested that we super impose the graphs on each other. The same participant was a fervent advocate of including more granular feedback on a daily level. He said that being able to see rates of consumption at different times of the day will help people associate behavior and high consumption rates. It's interesting because at the beginning of the interview, he said that his family received a notice from the utility company that they were one of the highest users in their neighborhood. He was unable to answer whether the utility voluntarily sent him the notice or whether it was something his household signed up for. He was very keen on the usage information being displayed and also the types of tips that were presented.

I had altered the tip—raising your thermostat from 68F to 72F—that was shown in the slide to include dollar information. The dollar feedback associated with the tip was calculated through the IHD simulation that we currently have in Qualtrics. One of the suggestions about the tips were that we included electricity usage facts, in conjunction with the tips being displayed. She said that adding "cool" facts about usage (or generation) might make the tips more interesting to look at. She also recommended that the tips were the first thing you saw when toggling between pictures and usage information. She said the idea behind the "cool" facts were from the smartphone application called Fruit Ninja. After losing, they would present the player with a fact about a fruit, which supposedly made the experience more pleasing after losing a round.

There were other suggestions to add the temperature information to either the speedometer or the calendar. Adding this information to the speedometer might make it more confusing than necessary; however, more than one participant had voiced their preference to see the relationship between temperature and usage.