There more like a conversation, the content

There is no
unstructured interview in social sciences, because the study requires at least
some prior knowledge about the topic. Even if there is a perception that semi-
structured interviews are more like a conversation, the content of the
conversation is controlled by the interviewer and his or her research questions
(DiCicco- Bloom& Crabtree, 2006). Also, because of the conversation-
guidance position, interviewer is in a power position (Qu& Dumay, 2011)

In face- to face semi-
structured interviews closed questions are avoided as much as possible, topics
usually are roughly described before, because the aim is to get an interview
closer to the conversation than questioning. Interviews usually are not fixed,
there is a large room for improvisation. Also, interviewer’s role is huge in
the interview; interviewer is not just influencing and controlling the pace of
the interview, but also interviewer’s demographic characteristics might
influence the answers. Because of the interviewer- role to the study, it is not
possible to replicate results (another interviewer might get different
responses). Interviewer bias might be avoided by gaining interviewers
experience in the interview guidance, also if the gained data do respond to the
theoretical concepts and the aim of the study, it is considered as a
confirmation of the validity (Heyink& Tymstra, 1993; Flick, 2009; Qu&
Dumay, 2011).

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!


order now

For this respective
study, high validity was important to achieve. Before conducting all
interviews, a careful pre- study of the national identity and group identity
theory and other studies of the Latvian- Russian issues were conducted. Before
the interviews, a table of operationalization (Appendix 3)was created, to
achieve the close link between the questions and theoretical concepts. However,
the interviewer is not very experienced in interview making; before the study
there was an experience of 8 semi- structured interviews. Also, the
demographics of the interviewer should be noted- it is a white, ethnic Latvian
female in her mid 20s.

All interviews are
different from each other not just because of the responses, but also because
of the fact that different people experiences lead to different following up
questions. Some questions are “missing” in some interviews. Such an approach is
typical for semi- structured interviews, because there should be an open
discussion and also the interview itself should be adopted to the respective
person (Flick, 2009).

The interview guide
(interview guide is seen in Appendix 2) was planned by using the guidance of
DiCicco- Bloom and Crabtree (2006), which lead to the interview structure as
follows- starting with a broad and open-ended question (first associations with
Latvia or home) and then depending from the answers, interview continued with
other related topics. The more sensitive questions (opinion about 9th
of May) were eft for the latter part of the interview, when some trust was
gained. Interviewees were encouraged to be as open and elaborative as possible
both before the interview and during the interview.

It is important that
the interviewee feel relaxed and unassessed during the interview (Qu&
Dumay, 2011; Doody & Noonan, 2013; Dempsey, Dowling, Larkin& Murphy,
2016); all interviews took place in several cafes in Riga or near area. When
organizing interviews, interviewees were asked whether they have any
preferences of the place. According to their preferences, interview place was
set.

Interviewees are not supposed to know everything about
the study, but they have to be informed about the interview process and how the
data will be used (Qu& Dumay, 2011). This rule was also respected-
interviewees were informed about the aim of the study and the general idea of
it beforehand and information about the interview process was given just before
the interview. Interviewees were encouraged to ask any questions if they have.

When using qualitative interviews, there is a high
risk that the sample and interviewee selection will be biased, which, however,
would affect the very validity of the study (Olszewski, Macey& Lindstrom,
2006). The recruiting process of the interviewees is described in the following
section.

 

 

 

3.1.1.     Interviewee recruiting process

Recruiting
interviewees was a complicated process because there are no clear boundaries of
the ethnic group. Also, the responsiveness was not very high (28 were asked to
participate, 16 responded).

Even if there is no
aim to get a statistically representative sample in the qualitative study,
usually at least some characteristics are considered as major- there should be
comparability on some levels (Flick, 2009; DiCicco- Bloom& Crabtree, 2006).
The target group of the study is Latvian- Russian millennials, thus there are
two important characters- age and ethnic background.

Since it is not a
good praxis to interview person, you are very familiar with, it was found
appropriate to ask friends or familiar persons, whether they know someone, who
would be interested in to participate.  The
network of interviewees is summarized in the graph (Graph 1) below. Two of the
interviewees were known before, but not closely. The snowball sampling method
was used, however, as it could be seen in the graph, it did not work as planned,
because of the low responsiveness. Also, one person (Anastasia) was recruited
through her official email (she was recruited because of her political
activity).

The main demographics
of the interviewees are summarized in the table below. Some names are changed,
some are not due to the willingness of the interviewee. In order to provide the
autonomy, the question of which name the respective person would like to have
in this study, is excluded. It is also not shown in the table whether it is a
real or changed name of the person. There are eight females and seven male
respondents included in the study. Average age for the interviewees- 25,2.
However, average age for females- 23,75 and males- 26,86. It is hard to
determine how representative the sample is, because there are no valid data,
for instace, of the education level of Latvian- Russian millennials.