Parallel Prototyping Leads to Better Design Results, More Divergence, and Increased Self-Efficacy
Steven P. Dow, Alana Glassco, Jonathan Kass, Melissa Schwarz, Daniel L. Schwartz, and Scott R. Klemmer. 2010. Parallel prototyping leads to better design results, more divergence, and increased self-efficacy. ACM Trans. Comput.-Hum. Interact. 17, 4, Article 18 (December 2010), 24 pages. DOI=http://dx.doi.org/10.1145/1879831.1879836
Summary:
Based on the several advantages of iteration, the author proposed the question that whether creating and receiving feedback on multiple prototypes in parallel, as opposed to serially, could affect learning, self-efficacy, and design exploration? This is a great question because those two methods of prototyping are wildly used in researches, industry, companies, and educations.
The author manipulated an experiment to let participants create a web advertisement in two different conditions, parallel and serial. In parallel condition, participants created multiple prototypes before receiving feedback, while in serial condition, participants received feedback at once after creating each prototype. To validate the results, the author considered several dependent measures, the performance by click-through-rate and Google Analytics, divergence by pairwise comparison, self-efficacy by participants’ self-rating, and feedback from professionals and magazine editors. After all this measurement, the results show us something interesting. Parallel condition significantly outperformed those from the serial condition. And independent raters found parallel prototypes to be more diverse and divergence. Also, parallel participants reported a larger increase in task-specific self-confidence and self-efficacy.
Reflection:
In this paper, I think that choosing a web advertisement design is a great idea. This task is not difficult and complex to finish in a short period of time. Creating an advertisement could also be easy for most participants because we all browse many advertisements around ourselves every day. And the measure of creativity for these advertisements is directly and objectively. The combination of these different dependent measures is also a good idea, instead of just counting the number of clicks or likes.
The results are in the contemplation of myself. I think for most people, receiving critiques and feedbacks instantly after finishing some works is terrible. These activities could definitely blow people’s confidence. And repeating working on one specific stuff could actually fixate one’s mind and imagination. While in the parallel condition, participants could make multiple ideas at the same time, which means they don’t need to think about one thing deeply. And receiving feedbacks a few minutes later could give them time to reflect those works.
Discussion:
The author has chosen several different dependent measures to validate the results. I think those are all great measures. But giving them different weight may be better than considering them equally. For example, this experiment is about advertisement designing. People’s clicks and likes should more important than professionals.
Also, I think maybe in future work, the author could think about how many ideas are being designed simultaneously in one period of time could get a better design.
Or how could we improve serial condition to get a good result? Because in many cases, the parallel is not working. And many people may prefer serial condition than the parallel condition. Like extend the time between finishing works and receiving feedbacks?
Also, after the experiment, I think the feeling of those participants is really important.