Taylor Wells and Dr. Paul Benjamin Lowry, Information Systems
As e-business is becoming an increasingly important part of the world economy, businesses are looking for methods and strategies to increase visitation to their websites and more importantly positively affect purchase intentions. One strategy for increasing purchase intention that has been found effective is to increase consumer trust with websites; websites that generate higher levels of consumer trust tend to have higher hit rates and purchase. Although trust is a complex and multidimensional construct, one study has suggested that color can have an effect on trust with websites (Kim & Moon, 1998). To further understand this phenomenon, the relationship between color and trust, we first examined the existing literature in the psychology, marketing, and human-computer interaction literature to determine the potential reasons for the interaction. Next we formulated hypotheses based on existing theory and conducted an experiment to test these hypotheses. This paper will describe the research process, the results discovered, and outline directions for future research.
In reviewing the literature, we found that although color has been researched in the human-computer interaction and information systems literature, very little research is empirical or based on theory. One of the main discoveries in the literature was the application of Apter’s theory of psychological reversals to color (Walters, Apter, and Svebak, 1982). This theory states that a person can be in one of two stable psychological mood-states at a time, yet can reverse to the other stable state at any time. In the telic-paratelic mood-state pair, pleasure is a function of which mood-state a person is in and their level of psychological arousal. As color can influence arousal, its affect on pleasure has been found in office workers. Based on this theoretical basis, we created hypothesis that integrated reversal theory with the interaction between color and trust on the web.
The hypotheses created were based on the way that color affects psychological arousal. Warm colors increase arousal while cool shades decrease arousal. Thus, a person in a telic (arousal avoidance) state should be more satisfied and experience higher levels of trust with a website that contains cool colors as opposed to warm colors while a person in a paratelic (arousal seeking) state should be more satisfied and experience higher levels of trust with a website that contains warm colors as opposed to cool colors.
To test these hypotheses, we conducted a laboratory experiment in which subjects experienced two websites that had specifically designed color treatments. Their mood-state and satisfaction levels were measured. Additionally, some of the task instructions were designed to help induce particular mood-states. During this process, I designed the websites and helped to manage the subjects. It was a great learning experience to go though the procedures to obtain IRB approval for human subject use. Additionally, before the experiment was conducted, my mentor had me meet with the statistical consultation center to help make sure that the experimental design was reasonable and thought through.
After we collected the data, we met again with the statistical consultation center for assistance with the statistical analysis. The results of the experimental treatment were inconclusive. There was no statistical difference between the groups, nor was there a difference between the groups that had instructions designed to induce a particular mood-state. None of the hypotheses were supported by the experimental results. As this was not the results that were expected, we more closely examined the experimental design, treatments, and previous literature to determine the cause of our lack of significant results. This process of reexamination was perhaps the most instructive of the entire process.
After further examination, it was determined that the treatments may not have been strong enough operationalizations of the hypotheses and that the color component may have been operationalized incorrectly. Color is a complex construct that can be mathematically represented in several ways. One of the most common is the hue saturation value (HSV) model. Some research has suggested that hue may be the most important factor in operationalizations of color, yet other studies claim that saturation or value is much more important that hue. The treatments in the experiment were based on the research that claims that hue is most important. Warm and cool colors were defined by their hue. By looking at the experiment from the perspective of saturation and value, the treatments may have been confounding. Even though the results of the experiment did not support our original hypotheses, we believe that the results were not a function of incorrect theory or inappropriate hypotheses, but rather a slightly incorrect operationalization of the hypotheses.
As previously mentioned, after finding that the experimental results did not support our hypotheses, we went back to the literature to determine if we had done anything incorrectly. While thoroughly investigating the literature, we found that there was no extensive published literature review on the effects of color in relation to websites. Thus, we drafted an interdisciplinary literature review investigating the effects of color. This paper has recently been submitted to the journal Communications of the Association for Information Systems under the title, “Building a Theoretical Framework to Guide Research on Color Use in Websites: A Review and Critique of Interdisciplinary Research on Color.”
This research experience has been invaluable in enriching my undergraduate education and influencing my career choice. I am currently applying to Ph.D. programs in information systems with hopes of pursuing a career in academia that includes research. We plan or reworking our treatments and experimental design so that a new experiment can be performed and the paper can be submitted to an academic conference or journal. The opportunity to conduct mentored research has been a wonderful learning experience.
References
- Kim, J. & Moon, J. Y. (1998). Designing towards emotional usability in customer interfaces–trustworthiness of cyber banking system interfaces. Interacting with Computers, 10(1), 1-29.
- Walters, J., Apter, M. J., & Svebak, S. (1982). Color preference, arousal, and theory of psychological reversals. Motivation and Emotion, 6(3), 193-215.