The researchers used a focus group interview.  Focus groups are very efficient. Several interviews are carried out at the same time. In addition, since the goal was to carry out several sessions in order to see if there was a consistency of response to the questions, using focus groups really cut down on the amount of time and the number of researchers necessary to carry out the research. Secondly, focus groups help to prompt discussion. One-on-one interviews may feel less naturalistic, less like a free conversation about a topic. Focus groups are high in ecological validity and the participants may get ideas from each other that they would not have thought of in a one-to-one setting. This may help the researchers to obtain richer data.

The researchers used a stratified sampling technique. Participants were selected non-randomly according to some fixed percentage. In stratified sampling you want to represent the major characteristics of the population by sampling a proportional amount of each - that is, you want the characteristics of the group to reflect the proportion of those characteristics which exist in the target population. For instance, if you know that 20% of the population has an advanced degree, 50% has an undergraduate degree, 20% has a vocational degree and 10% has a high school education, then you want a sample that has that same distribution of education level.  The reason for using a stratified sample is so that the researcher can more easily generalize the findings to the target population. If the sample reflects Swiss society, then it is likely that the findings will reflect the general behaviour of the Swiss community.

To avoid bias, the researcher might use reflexivity.  Personal reflexivity asks the researcher to reflect on his own biases and how they may have affected the study. Whenever "issues" are studied, one has to question the motivation of the researcher. It is possible that the researchers are very pro-environmentalist and are very concerned about global warming.  This may mean, then, that they may be more critical of those who do not change their behaviour. They may read into comments about the "difficulty of changing behaviour" as a "reluctance to change behaviour." They may also mistake the expression of interest as being "alarmed." The researchers should disclose this information as the interpretation of the transcripts is highly subjective and their personal biases could have an effect on their conclusions.

Another way to control bias is to consider the method that is being used.  Could a different approach have brought about a different understanding of the topic? In this case, the researchers may reflect on the use of the models and the expert. It could be that the reactions of the group were less about their own feelings about climate change and more about the models that were presented. This, however, was accounted for by having five different meetings in which different models were presented.  It could be that the group was responding to the expert. Feelings that the expert was arrogant, overly confident or patronizing could lead the group to take a contrary opinion and be more argumentative. If the expert seemed to hesitate or express doubt, then this could account for the response from the participants. Many of these ideas could have been communicated on a non-verbal basis to the participants. Finally, the use of a group interview has its disadvantages. The researcher should consider how conformity effects or power hierarchies within the group may have affected individual responses. It is important that the researcher reflects on the extent to which these factors could have played a role in the findings.