Article “Progressive Levels of Interactivity and Vividness in Web Marketing Sites” published in the Journal of Advertising contains research on the importance of such factors as vividness and interactivity affecting the attitude of the commercial web site visitor. Researchers Coyle and Thorson use the conclusions driven by Smith and Swinyard (1982,1983) who claimed that direct experiences better predict consumer behavior then indirect experience, to form the hypothesis of telepresence affecting consumer behavior in a similar manner. Telepresence is defined by Steuer (1992, p 76-77) as “the mediated perception of an environment”. Steuer also defined interactivity and vividness as two of the most important determinants of telepresence (p 76-77). Coyle and Thorson created 4 websites with 16 versions of each. These websites versions provided for different levels of interactivity and vividness using different amounts of audio materials, animation, links and clickable parts to research how telepresence level affects consumers. They asked the sample of 84 people to view their websites and then questioned them to find out what impact had level of interaction had on them. Conclusions were driven using analysis of variance applied to the results of questionnaires. This research had an objective to help web designers, marketers, and advertisers develop more efficient web sites. Successful branding on the web is based on understanding of consumer needs and influencing factors, and Coyle and Thorson helped recognizing these factors.
Coyle and Thorson used empirical research approach in their work. The existing solution to a question was that telepresence can influence significantly the positive attitude towards online resource being a virtual equivalent to direct experience of a consumer. They tried to provide empirical evidence to the problem and conducted experiment. Their experiment was set to support of falsify several hypothesis as follows:
H1: Perceived telepresence will increase as levels of interactivity in a web site increase
H2: Attitude toward web sites that are more interactive will be more positive than those toward less interactive sites.
H3: Attitude-behavior consistency related to web sites that are more interactive will be higher than attitude-behavior consistency related to web sites that are less interactive.
H4: Perceived telepresence will increase as levels of vividness in a web site increase.
H5: Attitudes toward sites more vivid will be more positive than those toward less vivid sites.
H6: Attitude-behavior consistency toward sites more vivid will be higher than that toward less vivid sites.
H7: Attitudes toward web sites that are more vivid will be more enduring than those toward less vivid sites. (Coyle, Thorson, 2001)
In order to test these hypotheses ANOVA was used.
To measure participants’ attitude toward web sites semantic differentials were used (like-dislike, good-bad, favourable-unfavorable). This measuring method is most appropriate to the subject of measuring as it is easy for participant’s understanding and for further reliability coefficient calculation.
To measure intention of return to the site seven-point Likert scale was used. It allowed participants easily express their agreement and disagreement with the statement “It is very likely that I will return to this site” and “I will return to this site next time I need a product”. The same method was used to measure purchase intention.
Dependent variables were measured with the Likert scale using statements designed by Kim and Biocca (1997) in their research on telepresence. This method has shown to be ineffective for Kim and Biocca research, however Coyle and Thorson did not develop their own statements to measure telepresence. This allowed to sufficiently minimize time resources for performing the measuring however information received for the research is not enough reliable.
Manipulation check and control variables were measured for research using Likert scales. Manipulation check variable is an important part of research which allows understanding whether the manipulated attributes were noticed by participant. In case they were not – all the data collected on the manipulated attributes would have been omitted. In this case manipulation check variable showed that participants paid attention to the manipulations, however it could have not been the case. To drive such conclusion a series of repeated measure ANOVAs was conducted.
Researched did the ANOVA to examine effects of order in which different sites were exposed to participants and none of such were noticed. However the sample of participants and order variances were not large enough to drive statistically proven conclusions. This part of research seems to have some kind of inaccuracy and for further research increasing participants sample and more efficient order variance scheme should be recommended.
Repeated measure ANOVA was conducted to test all the hypotheses. H1, H5 and H7 were supported, H2, H3, H6 were not supported, H4 received partial support. Testing hypotheses for his research was conducted thoroughly and provided enough material to drive necessary conclusions. Carefully examining each of hypotheses it becomes apparent that vividness of the web site has a significant effect on developing strong attitudes toward site, however interactivity did not show to have the same effect. Research has also shown that there is no relationship between levels of vividness and interactivity and attitude-behavior consistency. However, high levels of vividness create more enduring attitudes. These results may have been driven from the ANOVA allowing to develop some guidelines for marketers and web designers on how to develop a successful commercial web site.
Some strengths of the described research would be structured presentation of the research to the audience. The topic of the research, its objectives, data-gathering, analysis of the data and conclusions are all well presented and argumentative. Scales used for questionnaires were appropriate for this type of research and facilitated both filling out the questionnaires and analyzing them. Researches used scientific approach and number of secondary data to form their hypotheses therefore fairly hoping for objective results.
Nevertheless this research has some weaknesses as well. First of all the major weakness of the research would be the sample of participants. It was too small and not random, which might have caused statistical error.
In forming hypotheses number of assumptions was used. Many theories and researches that authors cite can be applied to the offline marketing and business, however online business should be oriented on the average Internet user. Coyle and Thorson mostly referred to the studies on web-marketing, however some of the works they cite can not be 100% applicable to the online business.
The conclusion part of the research may also need some revision. It clearly identifies the results of the research however in the form it is presented in the article it is not explaining well enough the application of this research for web-marketing.
When developing different web-sites with different levels of vividness and interaction, researchers did not develop the web site containing excessive vividness and interaction. In the research not enough attention was paid to the role of overly animated and interactive sites causing the aversion of a user toward such web site. Different means to provide vividness and interaction were also not discussed in the research. It could have been that all the websites contained the same amount of animation, however the quality and appeal of animation was different and this fact was not considered when conducting the study.
There are several suggestions on how this research can be improved. First of all more attention should be given to population sample. 86 people is just not enough to conduct a serious study. It is also important that population can be picked randomly. It was not possible to use appropriate sample in the University town with a population mainly consisting of students, however for results improvement it is necessary to pick the sample in an average city/village with an average population to provide more sufficient results of a study.
More attention should be paid to the sources used to form the hypotheses. Sources that do not suppose to be applied in online business should be omitted from the study.
Research article could be improved providing some hints for web designers and marketers on how the results of the study can be applied in their work. For example some tips should claim that colorful animation on the website develops positive attitude of an average visitor to this site. Such statements would be helpful in applying gained knowledge and more understandable for an average journal reader.
When creating different types of we sites some attention should be paid to developing web site with an excessive vividness and interaction to test how excessive use of such can influence negative attitude or decline of a positive attitude of a visitor.
The research conducted by Coyle and Thorson is definitely efficient and helpful in understanding the role of telepresence in developing visitors’ attitude and behavior regarding certain web sites, however there are some ways to improve the research and develop it providing even more correct data for interested in this subject public.
COYLE, J.R., THORSON, E., 2001. The Effects of Progressive Levels of Interactivity and Vividness in Web Marketing Sites. Journal of Advertizing, 30(3), 65-73.
KIM, T. and BIOCCA, F., 1997, "Telepresence via Television: Two Dimensions of Telepresence May Have Different Connections to Memory and Persuasion," Journal of Computer-Mediated Communication, 3 (September)
SMITH, R. E. and SWINYARD W.R.,1982. "Information Response Models: An Integrated Approach," Journal of Marketing, 46 (Winter), 81-93.
SMITH, R. E. and SWINYARD W.R.,1983. "Attitude-Behavior Consistency: The Impact of Product Trial Versus Advertising," Journal of Marketing Research, 20 (August), 257-67.
STEUER, J., 1992. "Defining Virtual Reality: Dimensions Determining Telepresence," Journal of Communication, 42(4), 73-76.