BIBLIOTEKA REGIONALISTY REGIONAL JOURNAL

: A dilemma for qualitative research is to achieve adequate data saturation and to reduce the time required for data collection. The aim of the article is to provide an answer to the question: to what extent should an in-depth interview be moderated in order to ensure an adequate level of saturation and time required for the analysis of several topics? The research methods include a literature analysis and an analysis of the author’s dissertation research reports. The article may help to improve the collection of high-quality data during in-depth interviews and to contribute to developing the methodological aspects of qualitative research.


Introduction
Qualitative research is becoming an increasingly common method of discovering 'reality', allowing researchers to explore the causes of various phenomena, which is usually a tedious and lengthy process.Over the past few years, there has been an increase in researchers' interest in qualitative research, which is reflected in the growing number of publications released in leading scientific journals and websites (Arino et al., 2016).Following the growing popularity of qualitative research and its share in the total number of all publications published (Bluhm et al., 2011), there is also a kind of 'warming of the image' -qualitative research does not seem to be a 'worse form of science' in relation to much more popular and commonly used quantitative research.Moreover, researchers are increasingly combining quantitative and qualitative methods (Czakon & Glinka, 2021) to obtain a more complete picture of complex phenomena, with particular emphasis on their sense or meaning, which is most often obtained through qualitative research (VanMaanen, 1997).Their primary role is therefore a reliable or even -as Geertz noted -"dense description" of the studied phenomena (Geertz, 1973), which means that the researcher should not separate the studied phenomenon from the context in which the phenomenon occurs.Czakon and Glinka emphasised that in qualitative research one should pay special attention not only to co-current processes of the studied phenomenon, but also to various nuances (e.g., cultural codes, ironies).For this purpose, observations are usually enriched with in-depth interviews, which are probably the most used technique of obtaining information (Czakon & Glinka, 2021), thanks to which it is possible to document the experiences, impressions and thoughts of the participants (Alvesson, 2003).Although subjective, interviews can be an effective method to explore the respondents' point of view, their emotions, attitudes or circumstances that affect their decisions;subjectivity is not a disadvantage in this case.Perhaps that is why one of the good practices used by qualitative researchers is to allow respondents to speak freely during interviews, regardless of previously prepared scenario questions (Czakon & Glinka, 2021).It is worth mentioning that both the observations and the accompanying interviews should lead to achieving an appropriate level of theoretical saturation.Suri defined this state as a willingness to complete the data collection process when no new information or issues arise in subsequent attempts (Suri, 2011).The level of saturation is usually defined by the researcher -it is the researcher who determines whether the collected data are sufficient to determine the so-called 'saturation limit' and conclude that the collected data are sufficient.Obtaining the right level of saturation while reducing the time consumption associated with the process of accumulation is one of the dilemmas faced by qualitative researchers.Mason observed that in the absence of a strict theoretical framework or definition, which is relatively common in qualitative research, it is difficult to determine the criterion of sufficiency (Mason, 2010).It seems reasonable to ask: to what extent an in-depth interview should be moderated, bearing in mind the achievement of a satisfactory level of saturation for the researcher and the time-consuming nature associated with the potential analysis of many threads, remarks or digressions?In other words -how extensive and how detailed answers to research questions should be provided in order that issues of interest to the researcher meet the criterion of sufficiency?

Types of in-depth interviews and the saturation criterion
The interview, being the most common method of collecting qualitative data, usually consists of direct conversation with respondents aimed at getting to know the current feelings and emotions related to the studied phenomenon (Mwita, 2022).It is assumed that the method, which has its roots in anthropology and sociology, began to be used at the beginning of the twentieth century.Its current development based largely on attempts to understand the relation between the researcher and the research is the result of the continuous evolution of philosophical research paradigms (Adhabi & Anozie, 2017) and experiences gathered over decades.
The nature of this interview was aptly defined by Kvale as "attempts to understand the world from the subject's point of view, to unfold the meaning of peoples' experiences, to uncover their lived world before scientific explanations" (Kvale, 1996).It creates a form of bond between researcher and subject of research (Adhabi & Anozie, 2017) regulated by ethical norms, according to which the interviewee is a subject, not an object.Edwards and Holland called this relation an exchange in which the context is taken into account as well as the subjective interviewee's view of reality, not only one's objective perception (Edwards & Holland, 2013).
From a technical point of view, there are three types of interviews that differ in the strength of influence and the extent of the researcher's interference (Mwita, 2022).

Structured interview
Structured interviews are usually fully controlled by the researcher, giving less space for free speech or interviewee's behavior (Stuckey, 2013).It is usually based on a questionnaire that contains a sequence of questions usually asked in the same way and in the same order, allowing the researcher to determine the rhythm of the interview.It is typically used in situations where the researcher is trying to obtain comparable data or information from a large number of respondents (Edwards & Holland, 2013).DiCicco-Bloom and Crabtree noted that structured interviews promote the acquisition of quantitative data (DiCicco-Bloom & Crabtree, 2006), which may provide a basis for designing further research hypotheses or separating issues for quantitative research.Although the simplest, this type of interview is also associated with a certain risk -interviewees may negatively perceive its too formal style and do not feel comfortable in a situation where important contexts and issues that do not fit into the area of research questions constructed by the researcher may be omitted.

Unstructured interview
This interview technique derives directly from the tradition of ethnographic research (DiCicco-Bloom & Crabtree, 2006) and allows interviewees greater autonomy and flexibility, which seem to be the key to conducting a "conversation with a purpose" (Burgess, 1984).In other words, this gives unstructured interviews a form of active listening rather than following a list and schedule.While the researcher's goals and research issues are clearly defined, the way of formulating thoughts or the directions of conversation, are not imposed by means of research questions (Edwards & Holland, 2013).The interview in this case takes the form of a story developed by the interviewee, focusing on the narrative description of reality (Muylaert et al., 2014), but still focused on goals relevant to the researcher (Jamshed, 2014).It is in a sense not a moderated interview but less formal conversation between the researcher and the respondent (Thompson, 2016), during which the researcher can gain a deeper insight into the reality described by the interviewee and capture various nuances that could be omitted during structured interviews due to the limited number of questions, or time.
It is also important to mention the limitations worth taking into consideration: due to their less formal style they are difficult or impossible to repeat and usually require high interpersonal skills to encourage the interviewee to share confidential or personal information.A too loose form of the interview may also imply certain ethical restrictions related to the risk of disclosure (conscious or unconscious) of certain information that may be seen as sensitive or confidential by the interviewee.The validation and reliability of the research itself may also be questioned, as it may be difficult to document the research process and meet the comparability criterion.However, the biggest challenge is time -free forms of expression, numerous digressions and too many threads appearing during the conversation can transform an interview into an overlong story (Thompson, 2016).

Semi-structured interview
Semi-structured interviews are very popular among qualitative researchers.This is probably due to the fact they combine the features of strictly planned structured interviews embedded in the time frame and the narrative freedom of an unstructured approach.In a typical semi-structured interview, the researcher typically uses a list of research questions or a series of topics to raise but takes a flexible approach to how and when the questions are asked, giving the interviewee freedom of expression and form.The language is usually open-minded, and the aim is to take the form of a dialogue rather than interrogation.Importantly, the researcher should be interested not only in the content of the message itself, but also in the context that should be considered, such as the interviewee's understanding of the issue raised by the researcher (Edwards & Holland, 2013).Mason came to similar conclusions by identifying the main features of the semi-structured approach (Mason, 2006): 1. Interaction between the researcher and the subject (or subjects) occurring most often in the form of a direct dialogue.
2. Consideration and understanding of the broader perspective by focusing on contexts (e.g., situational) whose fuller understanding occurs during the interaction between the researcher and the subject (or subjects).
3. A flexible but focused on the topic of the study structure that favors a narrative approach and at the same time allows the researcher to raise issues of interest.
Semi-structured interviews are also characterized by a certain structure, thanks to which it is possible to compare data obtained from many respondents (e.g., repeated regularities or observations).The researcher, having a list of questions or topics, sets the framework of the interview which allows to return to the research questions even after digressions or additional threads from the interviewee appear.There is also no rigor associated with the need to raise issues strictly according to the scenario assumed at the beginning by the researcher (Edwards & Holland, 2013).
The limitations include the time consumed, which can be influenced by a high level of flexibility.In the case of unstructured interviews, there is a risk that many digressions and threads may appear, significantly extending the duration of the interview and thus the need to encode a large amount of data in subsequent stages of the research process.The role of the researcher as a moderator and the adopted strategy of conducting interviews are also important.Limiting the number of threads and using a more structured form or, on the contrary, enabling a free narrative, can significantly extend the interview and the next stages of the research process.

Saturation criteria
The concept of saturation, although often mentioned in qualitative research papers, is sometimes unclear and insufficiently specified in the context of some studies, which may be particularly visible in the work of younger scientists (Bowen, 2008).This concept derives from naturalistic inquiry which is characterized by, among others, conducting research in natural conditions, purposive sampling and an approach based on grounded theory and inductive analysis, with an extensive reporting system that meets the trustworthiness criteria.A researcher studies phenomena occurring in reality in their natural environment without manipulating them (Lincoln & Guba, 1985).It is worth noting that naturalistic inquiry is associated with grounded theory, which is based on the systematic collection and analysis of data leading to the understanding of phenomena and areas that have been only scarely studied (Strauss & Corbin, 1990).
Strauss and Corbin noted that the researcher's grounded theory-based assumptions may evolve during the course of the research due to the various interactions that occur during qualitative data collection and data analysis (Strauss & Corbin, 1994).This process often implies the need to introduce new threads or participants, which allows to obtain a complete picture of the studied phenomenon.It is assumed that the appropriate level of saturation is achieved when the data set is complete, and the introduction of new threads, categories, or information changes little or has no significant impact on the study (Strauss & Corbin, 1990).This also applies if no new data can be found or new topics extracted.The issue of saturation is not unambiguous and easy to define.What is considered to be completed for one researcher can be unsatisfied for another.Although there is no universal and replicable method to achieve a satisfactory level of saturation in studies (Fusch & Ness, 2015), it is worth bearing in mind the following principle with which the level of saturation can be diagnosed: no new data, no new coding, no new themes and ability to replicate the study (Guest et al., 2006).When these assumptions are met, theoretical saturation can be considered as appropriate.
It is worth stressing that achieving an appropriate level of saturation does not depend on the number of sources (e.g., the number of interviews conducted), but on the quality and 'depth' of the obtained data (Burmeister & Aitken, 2012).For instance, one hundred in-depth interviews may be conducted, and an inadequate level of theoretical saturation may be reached due to overly general questions.Fifteen interviews may also be conducted and a satisfactory level of saturation may be achieved through appropriately selected questions and deeper interaction with the respondent.
In order to increase the chances of obtaining an appropriate level of saturation, various techniques and research are usually diversified.This activity, which usually involves document analysis, various interview techniques (e.g., in-depth interview and focus group interview) and observations, is called triangulation.Triangulation is therefore a method of achieving an appropriate level of theoretical saturation (saturation), as it allows for the simultaneous exploration of multiple perspectives on an issue taking into account the relations between people, place and time (Denzin, 2012).Triangulation techniques can be used simultaneously -observations can be conducted during the in-depth interview (Mwita, 2022), so that the researcher can verify, for example, the congruence between non-verbal (body language) and verbal (speech) communication and documentation can be discussed with the interviewee at the same time.It is therefore worth bearing in mind that the more data the researcher manages to obtain during a single interview, the greater the efficiency and benefit for the entire research process.

Achieving the right level of saturation during in-depth interviews and the time criterion -a case study
There is no doubt that overlong sessions aimed at collecting as much data as possible in a relatively short time carry some risks.This is particularly important during indepth interviews where the researcher faces a dilemma: limit the number of questions and introduce time rigor or on the contrary -allow the interviewee to speak freely, treating the time frame as a secondary matter?Loosveldt, Beullens and Johnson noted that very long interview sessions may have a negative impact on the quality of the data obtained -the interviewee may become tired, lose interest in meeting with the researcher or provide irrelevant information due to their discouragement (Johnson et al., 2019;Loosveldt & Beullens, 2013), or give too laconic answers trying to end the meeting as soon as possible.Thus, it is most beneficial for the researcher if the interviewee is willing not only to provide comprehensive information, but also to share emotions and impressions, and the interview turns into the form of a dialogue.
Edwards and Holland noted that the ability to listen attentively with interest should be a key characteristic of the qualitative researcher.Interviewees should not be treated as passive respondents providing only answers to the questions asked.The ideal situation is when the interview becomes a conversation and there is a 'natural flow' between the respondent and the researcher (Edwards & Holland, 2013).Bernard pointed to several good practices to help an interviewee 'open up' and encourage conversation to obtain high-quality qualitative data (Bernard, 2000): 1.When the interlocutor has finished answering the question, it is good practice to wait a moment for a possible reflection, continuation of the speech or completion of the thought.
2. Repetition of the last sentence or thought spoken by the interlocutor.This is especially useful when the interlocutor describes a situation or event.According to Bernard, this signals to the interlocutor that everything is understandable and encourages them to continue or develop the thread.
3. Saying: yes, I see, right, I understand, really?and so on when the interviewee is talking, affirms what the interviewee has said.It is rather like paying attention to what has been said, a signal that the interviewer is truly focused on the interviewee's story.
4. After the interviewee has answered a question, a good practice is to encourage him/her to elaborate and go further through follow-on questions: "Why do you feel like that about it?""Can you tell me more about that?" "What did you mean when you…?" 5. Long questions may help at the beginning of interviews.Bernard gives the example: "tell me about diving into really deep water.What do you do to get ready, and how do you ascend and descend?What is it like down there?" The author of this article, when conducting qualitative research on the phenomenon of waste in the activities of foundations and associations in Lower Silesia, noted that the key moment is the beginning of the meeting.The further course of the conversation depends on it as well as its depth and quality.A less formal beginning, a casual conversation, and an open and welcoming disposition of the researcher are factors that encourage the interviewee to interact and dialogue more openly, which in turn results in a higher level of theoretical saturation.Such an atmosphere is conducive to a free exchange of thoughts and direct conversation, in which the researcher can gain insight into situational contexts, emotions or processes occurring in the organization, and sometimes even in documentation.It is also an excellent opportunity to conduct observation and focus attention on emerging threads when the interviewee behaves naturally and reciprocates the interaction with the researcher.It is also noteworthy that in some cases, thanks to the response of the interviewee, it is also possible to analyze or verify the information contained in the documentation (e.g.reports, information from social media, etc.).
During twenty interviews, it was observed that there are factors worth paying attention to achieve a high level of theoretical saturation.Due to the extent of the issue, several of the most important ones have been selected that may be relevant to the interviewee: 1. Clothing and first impression.It is necessary to adapt in this respect to the studied environment, but in most cases, casual prevails rather than official.In some situations, too formal or too casual clothing can build a distance between the interviewee and the researcher, which negatively affects the process of collecting qualitative data.
2. Behavior of the researcher -open, friendly, not fake.Too formal behavior (e.g.overly restrained) or too casual, can be negatively perceived by the interviewee.In turn, the researcher's natural behavior (reactions, showing emotions, compatibility of verbal and non-verbal communication) and open attitude favors more effective interaction.
3. Communication.The compatibility of verbal communication with non-verbal communication is particularly important if the researcher wants to create a partner relationship with the interviewee, leading to obtaining high-quality qualitative data.Scientific jargon or too formal vocabulary, as well as too casual communication are usually avoided.However, there are exceptions -a good example is an interview conducted with the manager of an association promoting punk rock music, during which words commonly considered to be censurable were spoken.The researcher had to adapt the way of communication to be understood and accepted by the environment, using unconventional methods, such as asking questions in a less formal way, using words and codes used in the environment of people associated with punk rock music.It was a way of entering this environment and gaining favor to obtain information about the cooperation of the local government and non-governmental organizations operating in the area.4. A common area of interest.Familiarity with the interviewee's environment or issue may be an asset (although it is not a requirement or condition of success).This increases the likelihood that the researcher will be treated as the interviewee's partner rather than an ‚investigative journalist'.It may be good practice to mention this before the formal part of the interview begins.
5. Moderating the interview process.Long interviews are not a prerequisite for obtaining high quality information but allowing the interviewee to digress, make additional comments or insights, even if not directly related to the research questions, may add value to the study or shed light on noteworthy themes.However, it should be pointed out that extending the time of the in-depth interview results in longer coding and thus prolongs the research process.

Conclusions and summary
The dilemmas faced by researchers using in-depth interviews as a method of obtaining information are not simple and unambiguous to resolve.It is the researcher who ultimately decides what type of interview is used and the limit determining the appropriate level of theoretical saturation.Due to the triangulation of analytical methods, it is possible to create a certain theoretical framework defining the range by which the saturation level can be described as satisfactory.However, this is still a subjective assessment depending on many factors, including the time spent on research, financial resources, availability of information sources and their quality, and finally -the researcher's decision to recognize that the collected research material is complete and new information will not change much.Undoubtedly, the time the researcher spends on interviews should be well used and planned, although, when exploring relatively new or unexplored phenomena, it is advisable to be open to threads and information that seemingly may not be related to the issue under investigation.
Regardless of whether the researcher uses a structured, semi-structured or unstructured approach, none of the methods mentioned are free of drawbacks; neither are they universal means of solving a research problem in qualitative research.It is ultimately up to the researcher, based on his or her knowledge and experience, to decide on the choice of method and when theoretical saturation reaches an appropriate level.
In most cases, a semi-structured approach represents a reasonable trade-off between a flexible, structured interview and the overly casual narrative characteristic of unstructured interviews.When choosing this type of approach, however, it should be borne in mind that less controlled communication may result in longer interview times and therefore more time-consuming analysis of the data obtained in subsequent stages of the research process.Yet, despite the existing risk of prolonging the time of the research process, it may be worth considering whether to give the interviewee more freedom of expression to gather as much information as possible in one meeting.Such a procedure may reduce the necessity of spending extra time on repeating meetings with interviewees or additional interviews.

Future research
This article, which is a rather general overview of the reflections of selected researchers over the years, is merely an attempt to highlight important methodological and organizational aspects related to the process of qualitative data collection by means of in-depth interviews.Due to the complex nature of the various technical and organizational conditions, it is necessary to analyze them in an application perspective considering, among other things, aspects of logistics (planning, organization, and implementation), interpersonal communication (building partnerships with interviewees, which result, among other things, in obtaining information relevant to the research) and good practices.It is also worth noting the practice, becoming increasingly common among qualitative researchers, of outsourcing interviewing and coding -and the impact of these practices on the quality and trustworthiness criterion.