The small sample size and the narrowness of their selection criteria may be seen as limitation of this study. The fact that certain participants were
previously known to the researcher could view as being invalid; a non-random purposive group of interviewees may be seen by other researchers as a bias sample. That said, the findings are comparable with other studies conducted in different target niches then, based on this particular situation, one can speculate that similar occurrences might be found in other parts of the sampling frame.
Therefore, on that basis, the current study contributes to the growing corpus data on the subject of reverse culture shock, and how to manage its effects on professional sojourners returning to Finland (Frey et al., 2000; Robson, 2002;
Krippendorff, 2013).
MANAGING THE REPATRIATION OF PROFESSIONALS 79 There were a number of obstacles which impeded the smooth conduct of this research project. For example, one of these was identifying a sufficient number of professionals for the sample group to take part. If the number of participants is too small, a study will undoubted not have the desired credibility. As stated earlier, conducting a similar number of interviews that have been included in other studies within the same discipline ought to have alleviated this problem. Furthermore, the current study intended to investigate how the phenomenon of reverse culture shock was perceived by returnees and the number of times a particular manifestation arose, it did not require reaching data saturation and therefore a relatively small sample was seen to be
sufficient. In reality, identifying the requisite number of individuals was not the main issue. Finding the time and the opportunity to conduct the interviews was.
Those identified were mostly employed in very active, highly responsible roles and therefore their time was at a premium. On occasions it took several
attempts over periods exceeding six months and a number of frustrating last-minute cancellations, to find a mutually acceptable time for an interview.
Allied to this was the ethical consideration of trust and anonymity. A number of participants expressly stated they did not wish their identity to be revealed as it may compromise their position within their professional environment.
Therefore, all interviewees were assured that all reasonable steps would be taken to ensure their anonymity through the use of pseudonyms but at the same time, were reminded that their data was likely to be published as part of this dissertation and therefore, it might be possible that colleagues could be able to identify them due to the nature of the subject under research and the relatively small working communities they belonged to. Despite all the best efforts in
MANAGING THE REPATRIATION OF PROFESSIONALS 80 striving to maintain their anonymity, a research cannot be expected to avoid revelations under such latter circumstances.
The English language skills of some of the interviewees also caused some difficulty. Therefore, there was potential that they will either not understand the questions, or misunderstand them. In the absence of any formal guidance on this matter, common-sense prevailed and that meant providing them with a copy of the discussion frame a number of days prior to the interview in order to help them prepare. An alternative solution could have been to employ the services of an interpreter. This was discounted on economic grounds.
There was also the Hawthorne effect and the matter of false reporting to consider. Peoples’ behaviour can sometimes be influenced by their awareness of the intentions of the researcher and that may lead to a change in their behaviour (Frey et al., 2000). Such changes are called the Hawthorne effect derived from a 1939 study about the relationship between perceived light levels and productivity in a factory in the U.S.A. An example of such behavioural change could be false reporting; participants reporting what they think the researcher wanted to hear in the misguided belief that they were being helpful. This may have been the case in this study due to the provision of a copy of the discussion frame to the interviewees as described in the previous paragraph. Such a threat to validity was hopefully combated by each interviewee being told of such concerns before the interview begins and reminded that only their own honest opinions and feelings were being sought.
Furthermore, taking inferences from the data is subject to interpretation on behalf of the researcher. If a sampling plan is too rigid it could be suggested
MANAGING THE REPATRIATION OF PROFESSIONALS 81 that the researcher found only what they were looking for or expected to see.
"Where texts and images are involved, or, more generally, where the
phenomena of interest to analysts are social in nature, mechanical instruments have serious shortcomings that only culturally competent humans can
overcome" (Krippendorf, 2013, p. 127). Yet even such cultural competence is susceptible to the researchers own influences brought about by their own life experiences and education. My own perspective as an immigrant, albeit one of many years' experience working in the geographical area, of the findings is etic and therefore may not be seen as being a "best fit" solution to the situation in Central Finland.