Economic Evaluation of an Investment in Medical Websites and Medical Web-Based Services

22.2Survey Formation

In order to have an estimation of the users’/patients’ value gained from the use of medical web-based services in the context of WTP, the formation of an online survey/ questioner was decided. The online survey setting will not affect the survey results or consist the survey biased because of the fact that the participants will be only people that have access to a personal computer and know how to use Internet and search engines, since the main objective of this survey/study is to measure the value of the medical web-based services and these participants’ skills are considered a perquisite assumption. The survey is addressed to people living in UK was circulated via email to various mailing lists of organizations like universities and /or non-profit health organizations, and via social networks such as Facebook and Twitter. Due to the nature of the online setting, the crowd-sourcing methodology that was followed and the worldwide availability of the medical websites through the World Wide Web it is difficult to guarantee that all the participants were people living in UK for quite a long time. Instead the reader can be assured that every possible attempt was made to restrict the sample in people living or lived in UK for a long period and the sample is considered to be consisted of this kind of participants in a percentage greater than 85%. This percentage was specified by calculating the probability that people from other countries had to access the survey link through the crowd-sourcing methodology that was applied and will be analysed in the following sections.

In order to justify the choice of this kind of survey format concerning the estimation of public WTP, a thorough study of a well-established body of literature was followed the initial thoughts. Comparing the online WTP survey setting with the more popular Face-Face interview WTP survey setting the comprehensive study of the extant literature led to the choice of the online survey setting as the more suitable for achieving the objectives of the current study since does not present any significant differences compared to the Face to Face interviews.

Nielsen (2011) in his work examined if there is any difference between estimating WTP from a web survey or a face to face interview concluding that although the National Oceanographic and Atmospheric Administration (NOAA) guidelines (1993) suggest that face to face or via phone interviews should be used, these guidelines were established during a period of time where the use of Internet and online surveys was not a popular and frequently used method and even the people were not so familiar with the use of Internet and its capabilities. By examining the two approaches for the Dillman’s (2009; 2000) four types of survey errors (coverage and sampling error, no response bias and measurement error) his results have shown that the response rate in the web bases survey was significantly low but this approach had the same potential non respond bias with the face-to face interviews implying that the low response rate did not influence significantly the survey results. Moreover, a bias concerning what is socially desirable seem to influence more the face-to face approach’s results. In addition to these, the web survey’s sample presented greater variance concerning the survey’s scope sensitivity. In general his results can be considered promising for the use of Internet web surveys to estimate the WTP but some more caution is needed mainly to avoid survey mode effects and biases.

Furthermore, Iraguen and De Dios Ortúzar, J. (2004), state that the WTP estimation through an Internet survey setting presents many advantages since it enables the analyst to get the answers immediately after the respondent answers them and also reduces the errors and biases from coding the data from the answers. In addition, they state that it allows the automatic randomisation in case of a bidding game setting survey, and in general the randomization of the scenarios the participant is engaged to avoiding this way the sequence effect. Moreover, this survey setting allows for complete anonymity and make participants keener to answer unbiased the questions. Finally they state that this type of survey setting can have the major disadvantage of been vulnerable to biases related to sample selection since the use of Internet is only widespread among specific groups of people.

The standard approach to estimate the WTP was chosen instead of the marginal since according to the research of Shackley and Donaldson (2002), produced results more accurate for the scope of this survey. In the standard approach, the questionnaires are formed in such way so the participants to be asked about their WTP for one healthcare program ignoring the other programs while in the marginal approach the questionnaires are formed in such a way so the participants to be asked first to rank the healthcare programs and then they were asked to express their WTP first for the lowest ranked healthcare program. Shackley and Donaldson (2002) compared the standard and the marginal WTP approach in the healthcare field by contacting a survey using two questioners one according the standard approach and one according the marginal approach. The results have shown that the marginal approach is better in making participants express a higher WTP since the mean WTP of the marginal approach questionnaire was much higher than the same measure from the standard approach questionnaire. On the other hand marginal method didn’t manage to solve the problem of the inconsistency between the participant’s preferred program ranking and the allocation of their higher WTP.

The questions that were chosen to be part of the online survey questionnaire were divided into 3 sections. In the beginning of the survey a text was informing the participants for the purpose of this research and its voluntary character providing them full details for the scope of the survey and the sections that were consisting the survey questionnaire. In addition to the aforementioned details, the first screen of the survey was informing the ethical approval that the survey received from the school’s Research Ethics Committee and provided also the email address and telephone number of the president of the committee in case the participants wanted to authenticate this approval (Appendix B).

The first section of the survey’s questionnaire included general demographic questions asking the participants to provide information about their age, their ethnicity, their levels of income and education as well as the their occupation. These demographic data were used to produce some descriptive statistics about our sample and at a later stage to extract some more useful information/conclusions and to support the findings of this research concerning the value of the medical web-based services and the medical websites.

In the second section of the online survey’s questionnaire there were included questions about the willingness to use or the current use of medical web-based services by the participants, questions about which medical site have they use/have used, about the reasons that make them reluctant to use such online services as well as questions about their maximum WTP or potential maximum WTP (depending if they have used or they are willing to use such services). The questions referring to their maximum WTP concerning the current or potential use of the medical web-based services/applications, were trying to engage the participants in a specific scenario, which main purpose was to make them think the benefits of the use of these services/applications. The participants had to think about these aforementioned benefits by comparing the use of these medical services/applications to the time that they potentially can save from going to and waiting in consultancy rooms, the better results that can may have compared to the traditional telephone communication with their personal doctor, the anonymity that the use of the Internet offers as well as the ability to have a number of alternative diagnoses in just few seconds and after few clicks. Moreover, in section 2 there were questions regarding the level of certainty of the participants concerning their choice of their maximum Willingness to Pay as well as a question asking them to provide feedback about the potential difficulties that they had in interpreting or answering any questions of the online questionnaire.

The third section offers to the participants the option to take part in a context that was offering them the chance to win a top of the range tablet pc, used as incentive to attract more participants as will be described further down in the crowd-sourcing methodology description for this study.

Finally, as Frew et al. (2004) pinpointed the fact that one of the first problems that the researchers have to face when they are dealing with the WTP is to ensure that the participants understand how the method works and the nature of the problem they are asked to solve. A pilot study was formed in order to solve this issue and to be sure that the survey will produce valid and useful results. The initial questionnaire was spread to a small number of participants (50 participants) in order to receive feedback about any difficulties the participants faced while interpreting or answering any question. The pilot-participant, corrections and recommendations as well as the changes implied to be necessary after the statistical analysis of these first results, were taken into serious consideration for the formation of the final version of the questionnaire.