Semi-structured dance research interviews

Rationale for Using Interview in the Dance Heritage Research

The ethnochoreological approach for this research required that I gathered three different pieces of information about raqs sharqi: its form, the culture and the experience.

The first two pieces of information have been gathered mainly via the analysis of online videos and of online and offline texts.

Regarding the experience, the interviews brought a more personal account of what it is for raqs sharqi practitioners to be involved in this dance form.

Indeed, as Mason states (2002, p. 63), I chose to include qualitative interviews in my research, because of my ontological position, which ‘suggests that that people’s knowledge, views, understandings, interpretations, experiences and interactions are meaningful properties of the social reality which your research questions are designed to explore’.

In-Depth Semi-Structured Interviews

In-depth interviews are generally recommended when the researcher holds a relativist ontology and constructionist epistemology, and when the aim is to understand how each participant views the world (Mason, 2002; Gratton and Jones, 2004; Cohen, Manion and Morrison, 2007; Marshall and Rossman, 2010).

Indeed, as Gratton and Jones (2004, p. 142) posit, ‘Interviews allow unexpected data to emerge’ because participants are allowed to express their thoughts in their own words.

Sampling and Recruiting Interviewees

Interviewees were recruited among those who I am in contact with online: some were drawn from my local dance community (in Cardiff and South Wales) and some were practitioners I had met during dance related trips in the past.

Being an insider in this field, helped me to contact participants and, as Roulston posits, (2009, p. 98) ‘relative intimacy and rapport with participants may enhance the generation of data in interview settings in ways not possible for ‘outsider’ researchers’.

The initial method of contact was by email or personal messages via social media.

Indeed, I would not have been able to be in contact with any of the participants, without the Internet.

Connecting with Interviewees

All of the interviewees were connected to me on Facebook. This medium helped facilitate contact and build rapport because the participants and I could see each other’s profiles and updates.

I agree with Baker (2013, p. 140), who states that ‘using Facebook facilitated communication between the researcher and participants across geospatial and temporal boundaries’.

Even with participants, I contacted by email, we were connected on Facebook, which helped us maintain familiarity with each other.

The sampling of participants was strategic, through a purposive sampling strategy, as defined by Sparkes and Smith (2014), to provide a relevant range of experiences from the Raqs Sharqi community.

However, since the participants were experts in Raqs sharqi (the field of interest of my research), the strategy could also be defined as expert sampling, as Kumar (2011) explains.

Because Egyptian Raqs sharqi is a transcultural phenomenon, sampling included international participants, although they had to be either English or Italian speakers, due to my own language limitations.

In order to gain an understanding of how cultural awareness about Raqs sharqi develops, participants were selected to include different levels of experience in relation to Raqs sharqi.

Also, I included people who have traveled to Egypt for dance training; some who live or lived there to dance professionally, and others who have never been there.

To include a diversity of approaches to the dance for comparison purposes, sampling was originally meant to include some participants who do not experience Raqs sharqi as a cultural dance or are breaking with its traditions.

However, such participants were too hard to involve, because their lack of interest in the cultural aspects of raqs sharqi, meant that they were not interested in this research.

Egyptian Nationals and Male Dancers

Two groups of people that I did not manage to recruit, were Egyptian nationals and male dancers.

Had I been able to travel to Egypt during the research, Egyptian practitioners would have been easier to contact, but online they do not seem as active as other nationalities and I faced a language barrier, as I cannot speak Arabic.

There are two Egyptian Raqs sharqi dancers living in the UK, whom I tried to contact but with no success.

One of them was recommended by another interviewee, so I contacted her but she never replied.

The second one is a man, who is in high demand as a teacher and performer worldwide, whom I had met in person during workshops.

I could not contact him though, because his agent answers his emails, acting as a gatekeeper, only agreeing to paid engagements.

The Issue of Gatekeepers in Research and Interviews

Dealing with gatekeepers is a common issue for researchers (Sampson and Thomas, 2003; Wanat, 2008; Reeves, 2010; Lund, Panda, and Dhal, 2016).

Gatekeepers, Cohen et al. (2007, p.123) explain, ‘may wish to avoid, contain, spread or control risk and therefore may bar access or make access conditional’.

As a result, gatekeepers can condition the choices that researchers make in terms of sampling.

Similarly, male Egyptian-style Raqs sharqi practitioners are few and far between, so I could not find any men to interview.

As a result of these difficulties, although my sampling strategy was mainly purposeful, there was also an element of convenience sampling, according to which (Cohen et al. 2007, p.114) ‘researchers simply choose the sample from those to whom they have easy access’. I interviewed 10 participants.

As I also employed other research methods, I reached saturation point (when no more new insights were generated from the data) at nine interviews, but I had already arranged the tenth at that point.

In qualitative research, as Cohen et al. (2007), Flick (2009) and Kumar (2011) agree, the focus is on the quality of the findings rather than the quantity and the researcher decides when saturation point has been reached.

Participant Responses

Out of the 10 participants: I proactively approached six; three volunteered by replying to posts I wrote on Facebook looking for participants and one volunteered after reading a post I wrote on my website about this research (Lo Iacono, 2015), following Kozinets’ advice (2009, p. 152).

Table 8 shows the 10 participants with their nationality, whether they have been to Egypt, their level of involvement, how they have been interviewed and how I contacted them.

They are all women, between the age of 30 and 72 (see footnotes), and all the interviews were conducted in English. In the section that follows, I will be discussing how I designed the interview and how I prepared the questions.


Table of interview participants.
Table 8 – Interview Participants

Design of the Interviews

I chose to conduct semi-structured interviews, writing what Maykut and Morehouse (1994, p. 83) define as an interview guide, ‘a series of topics or broad interview questions which the researcher is free to explore and probe with the interviewee’.

I used a guide to help me keep focused on issues relevant to my research while allowing for flexibility to let the data emerge.

The interview was as an open, informal conversation between the participant and me.

As Mason suggests (2002, p. 62), in qualitative interviews, ‘meanings and understandings are created in an interaction . . . involving researcher and interviewees’.

Cohen et al. (2007, p. 351) identify a range going from the most formal interview, with a standardized schedule, to non-directive interviews, in which ‘the interviewer takes on a subordinate role’.

The interviews for this study occupied the middle ground, whereby a guide provided the base for an informal conversation.

Gratton and Jones (2004) suggest that semi-structured interviews offer the researcher structure, with the flexibility to change the sequence of questions, or ask additional probing questions.

According to Maykut and Morehouse (1994, p. 78), ‘the skilled researcher will discover what is important to the interviewees, within the broad boundaries of the interview topics and questions, and pursue these new discoveries in the interview’, but they advise (ibid, p. 82) (particularly research beginners) to have some questions prepared in advance.

My questions were based on the research questions, following Cohen et al.’s (2007, p. 356) suggestion that doing so helps to keep the interview on track with what the researcher wants to find out.

The first question was a life story question, asking the participants to tell a short history of their involvement in Raqs sharqi, to have an overview of their background, and also as a warm-up.

This could be classified as an experience/behaviour question, about things that participants know, which, as Maykut and Morehouse (1994, p. 85) advise, are ‘useful to begin an interview’.

In a similar fashion, Gratton and Jones (2004, p. 144) suggest that an ‘easy’ question asked at the beginning should put the participant at ease.

The other questions revolved around transmission, authenticity, culture, uses of heritage, and embodiment trying to understand participants’ experiences and perceptions on these issues.

Where possible, I tried to group, as Gratton and Jones (ibid) advise, ‘questions about the same concept together’.

However, I did not worry about it too much because, due to the semi-structured nature of the interview, rather than keeping to a pre-set order, I chose the next question according to the flow of the conversation.

Appendix 1 contains my interview guide.

In what follows, I will explain the interview process, after recruiting the participants.

Interviews Process

In this section, I will outline the process I went through for the interviews, from the preparation stages to the rapport building phase, to the actual interviews.

Before the interviews took place, participants read an information sheet and filled a consent form. In the ethics section, I will elaborate more on this topic.

Building Rapport with Participants

For the interviews, I built rapport with participants over time. According to King and Horrocks (2010, p. 48), ‘rapport is . . . about trust – enabling the participant to feel comfortable in opening up to you’.

I had already met some of the participants before in person, long before starting this research, and I continued connecting with them online (mainly via Facebook) over the years.

I exchanged emails with participants before the interview and, following the interview, we kept being connected on Facebook as friends.

Verifying the Identity of Participants

In terms of identity verification, having an online presence, helped both me and my participants build trust in each other, as we knew we were who we said we were because we have more than one profile online such as Facebook, YouTube, websites, blogs, e-books, online classes.

Cross-referencing these profiles aided identity verification. As Sullivan (2012, p. 56) observes, drawing on Goffman’s presentation of self (1990 [1959]), ‘so much of our time is spent on the web that presentations of self online are potentially more accurate than they were 20 years ago’.

Using Skype Research Method

Interviews took place in person for participants based in/near Cardiff. For those located further afield, Skype® was used for interviews with the voice and video recorded by EVAER® software.

All participants agreed to be recorded and recording was deemed important ‘to obtain the best possible record of the interviewee’s words’ (Maykut and Morehouse, 1994, p. 93).

Out of the ten interviews carried out, three were done in person, one by email and six via Skype, as shown in Table 8.

In order to connect with practitioners globally, Skype was an invaluable tool that allowed me to reach distant participants in a time-efficient and financially affordable manner.

All the research participants were computer literate and there were no technical issues as they all had high-speed internet. The only exception was a call to Italy, during which the line was cut off a couple of times, but this was no obstacle and rapport was easily resumed after each interruption.

For Skype interviews, some researchers argue that building rapport is challenging (Cater, 2011; Rowley, 2012, Lo Iacono et al, 2016). Seitz (2015) points out that technical difficulties, such as when the call is interrupted, may harm rapport as it may be hard to start the conversation again.

Others though agree that it depends on the topic of the conversation, or that Skype even helps rapport.

Seitz (2015), for example, admits that, although she found it harder to receive answers on sensitive topics (such as online dating experiences) using Skype, this may be due to suspicions related to the video element of Skype, rather than lack of trust in the researcher.

Lo Iacono et al. (2016, para. 5.7) argue that ‘whether Skype or face to face interviews are better to build rapport, really depends on the topic of the research and on the personality of the participant and interviewer’.

Deakin and Wakefield (2013, p. 8) found that ‘Skype interviewees were more responsive and rapport was built quicker than in a number of face-to-face interviews’.

Other researchers argue that Skype is beneficial in building rapport, particularly if the participant or the interviewer is shy, because, as Hanna (2012, p. 241) states, ‘both the researcher and the researched are able to remain in a safe location without imposing on each other’s personal space’.

The fact that the participant is in a familiar environment may be, as Seitz (2015, p.4) suggests, ‘more beneficial to participants who are shy or introverted, allowing them to feel more comfortable opening up in front of a screen’.

Meho (2006) used email interviews in order to interview shy people and those who have difficulty to otherwise express themselves.

A notable difference between in person and Skype interviews is the fact that participants and researcher are not sharing the same space, so some of the non-verbal cues may be lost.

According to Hesse-Biber and Griffin (2012, p. 56) ‘tone of voice, and gestures, all provide a certain richness to qualitative data’.

Cohen et al. (2007, p. 153) and Novick (2008, p. 5) agree on the importance of nonverbal cues and on the fact that these can be lost in some forms of interview, such as when using telephone interviews.

According to Talja and McKenzie (2007, p. 102) ‘paralinguistic cues such as gesture, facial expression, and tone of voice can both convey emotion and provide the hearer with clues for interpreting the meaning of an utterance’.

With Skype at least, it is possible to not only hear each other’s voices, but also to see each other, so not all the non-verbal cues are lost.

The screen though, unless an external wide-angle camera is attached to the computer (as Petralia (2011) did during a dance cooperation project between Europe and the US), allows us to only see the head and shoulders of the other person.

This was not a problem for me, however, and, as Lo Iacono, Symonds and Brown point out (2016, para. 5.13) ‘in the context of Skype interviews . . . by focusing on the head and the shoulders, we can gather more details of these specific body parts, which can counterbalance not being able to see the rest of the body’.

Overall, I found Skype a very good ‘complementary data collection tool for qualitative researchers, which works well alongside other data collection methods as part of a broader research design and strategy’ (Lo Iacono, Symonds and Brown, 2016, para. 7.1).

With regards to the one email interview I did, I found that it allowed the participant time to reflect on the answers, thus generating rich data, and I was able to reach somebody who lives far away and who did not have access to Skype.

However, I found that rapport was weaker and the interaction lacked spontaneity, comparing to the in-person and Skype interviews.

I did not notice much difference in terms of rapport and richness of data generated, between Skype and in person.

I know this through watching the participants’ body language and because, at the end of each Skype interview, I asked them directly how they found being interviewed via Skype and all their reactions were positive.

Also, after comparing the richness of the data between Skype and in-person interviews, I could not see any differences

Because of the topic of the interview, which involved something that both the researcher and the participants were passionate about, establishing rapport during the interview and eliciting information was not hard. I shared Bourdieu’s (1996, p. 24) feeling that:

The person questioned took advantage of the opportunity we offered her to examine herself and . . . of the permission or prompting afforded by our questions or suggestions . . . to give vent . . . to experiences and thoughts long kept to herself.

Bourdieu (1996)

Indeed, my research participants were able to share thoughts and ideas on a topic that they are able to cover only with those who share their passion, which may not be very often.

Hence, their enthusiasm for sharing their thoughts was evident. The interviews lasted between one and two hours each.

As the topic of the research was not sensitive, there were no particular issues regarding ethics. However, specific guidelines still needed to be followed. What follows outlines the ethical procedures for each research method.


1 – Although there are people under 30 worldwide who practice belly dance, as my research focuses on the specific genre of Egyptian raqs sharqi and specifically on its cultural connotations, participants under 30 were not easy to find.

One reason may be that, when people start practicing belly dance, they are more interested in its leisure aspect rather than the culture. The more they learn about this dance, the more they become knowledgeable about and interested in its cultural significance and the different genres and styles, and this may take years of practice.

I had found one participant in her twenties who was initially interested in taking part to my research, but, in the end, she did not follow up.

Next Page >> Research ethics for this PhD in dance heritage.