Interview Method

The interview was conducted using the Delphi method. The name Delphi comes from Greek mythology; the oracle Delphi helped people by giving political instructions. The main purpose of the method was to make decisions before a major course of action. The method was launched in the 1950’s and early 1960’s by Kaplan, who thought that the method should be published.

After publishing, the method came into wide use in medicine and sociology, and later on it became popular in other fields. (Franklin & Hart, 2007, pp.

237-238; Iñaki, Landín & Fa, 2006, pp. 813-816; Loo, 2002, pp. 762; Steurer, 2011, pp. 959-960.)

The Delphi method can be divided into three different sub-methods by Franklin and Hart (2007, p. 238): classical Delphi is based on facts from a specific situation or topic; decision-making Delphi, as the name implies, is a method for making collaborative decisions; and policy Delphi is based on experts. First, the topic of the problem is created and then opinions are asked from the experts. After this, the collected information is analyzed and

summarized and the feedback is given to each person who took part in the interview. The main purpose is to collect opinions and make sure that the interviewer has understood the answers correctly. (Franklin & Hart, 2007, pp.

238-239.)

Although the classical Delphi method is mainly used to handle technical issues and topics, policy Delphi was chosen for this thesis. This sub-method seems to give significantly opposing views related to the issued topic. The first difference to other sub-methods is that the participants do not have to be the best experts in the chosen field; they can be people who are familiar with the topic. The second difference is that the researcher will concentrate on the analysis of the issue, not on the mechanism of how the decisions are made.

Mostly, policy Delphi deals with statements, comments and discussion;

therefore, it is important to organize a meeting where the participants can talk freely around the research problem. Still, there have to be certain questions to which the researcher wants specific answers. (Turoff, 2002, pp. 80-83.) The method follows always a certain route (Iñaki et al., 2006, pp. 814-816;

Loo, 2002, pp. 764-768; Steurer, 2011, pp. 959-960). First, the researcher defines the problem to be investigated (Steurer, 2011, p. 960). When the problem is established, the researcher prepares questions which are, of course, related to the main topic; the researcher determines what kind of information they would like to have. The questions should be simple and clear so the respondent can easily answer them; normally, the questions are open-ended questions and no right answer exists. (Iñaki et al., 2006, pp. 814-815; Loo, 2002, p. 765-766.)

Secondly, an expert panel is selected, which is a requirement for the Delphi method. It is crucial to select qualified experts; if the researcher does not know enough experts, the experts who have been invited to the panel can suggest other suitable candidates. This can lead to the situation where the panel consists of experts that are connected to each other somehow or another (for example colleagues and friends). This may be a problem in case if the answers of the panelists are affecting to each other’s. (Steurer, 2011, p. 960.) In addition, the important thing is that the panelists remain the same through the whole study; all the panelists participate anonymously and the issues discussed are confidential (Loo, 2002, p. 766; Steurer, 2011, p. 960). Also, the selection criteria are freely chosen: criteria can be based on, for example, personal experience or the selection can be done randomly (Iñaki et al., 2006, p. 814; Steurer, 2011, p. 960).

The number of the panelists is not specified; the Delphi method is only based on the gathered research information (Loo, 2002, p. 765; Iñaki et al., 2006, p.

814). When choosing the panel size, the research problem should be clear because this clarifies the number of panelists (Loo, 2002, p. 764). Depending on the topic to be discussed, the panel’s size may vary from one member to thirty members and because of the careful choice of panelists; the results can still be reliable (Iñaki et al., 2006, pp. 814-816). The main thing is to collect opinions, beliefs and judgments; and also to give opportunity for anonymous individual responses (Steurer, 2011, pp. 959-960).

After the previous steps are done, the actual Delphi method takes place as a questionnaire. The goals of the researcher’s study have to be clear and the literature review should be done, as these are the tools for generating reasonable and effective questions for the panel (Loo, 2002, pp. 764-765;

Steurer, 2011, p. 960). The questionnaire can be performed as a face-to-face interview, through mail/e-mail or on the Web (Loo, 2002, pp. 766).

When the researchers have the responses, the actual analyzing of the results begins (Iñaki et al., 2006, p. 815). The results may be listed in a graph or as calculated medians: this depends on the main question which was set in the beginning of the survey. (Steurer, 2011, p. 960.)

In the second round of the questionnaire, the analyzed answers from the previous round are shown to the participants At this point, the participants can check the answers and make corrections or additions. (Iñaki et al., 2006, p. 815; Loo, 2002, p. 766; Steurer, 2011, p. 960.) If the researcher feels more information is needed, a new questionnaire may be set and a third round begins. In most cases, the final analysis is done according to the results from the first and second rounds. (Steurer, 2011, p. 960.)

As with other methods, the Delphi method also has critical properties. It is true that the answers to the questions are based on the respondent’s personal beliefs and opinions but this can also be the strength of the method, depending on the main research question. When choosing different panels with different experts, the answers will lead to different results. (Steurer, 2011, p. 960.) However, as Steurer (2011, p. 960) wrote: “In general, the Delphi method is used only when scientific evidence is either absent or contradictory, and judgmental information is necessary”. (Loo, 2002, pp.

767-768; Steurer, 2011, pp. 959-960.)

The greatest benefit is that the Delphi method is based on the researcher’s questions. The conversation is focused on a certain problem and the

discussion stays on the right tracks. Still, because it is not a quantitative method, it gets considerable criticism from academia. Nevertheless, the benefits of this method outweigh this criticism (Franklin et al., 2007, pp. 241-242; Iñaki et al., 2006, p. 816):

1. The members in the panel are the experts and they will mutually complement each other.

2. Members’ opinions can be compared with other opinions.

3. Each comment in the research comes from an expert and expresses their point of view.

4. All the information collected during the panels is anonymous, so the participants are free to share their opinions or beliefs. As Steurer (2011, p. 960) stated, the anonymous aspect will protect the participants; they can say their real opinion and they will not lose face.