1 Introduction

Digital transformation, globalisation, war, climate change, migration: We live in a time of upheaval and transformation that makes us increasingly aware of our own vulnerability. It is not surprising that the general interest in vulnerability has increased in recent years. While initially focused on medical sciences, vulnerability has also attracted increasing attention in other disciplines—not least in communication and media studies. This is evidenced by a number of studies and publications, including those focused on questions of health communication (e.g. Stans et al. 2018) climate justice and media (e.g. Crawford et al. 2024), visibility (e.g. Keul 2023), and on vulnerability in general (e.g. Maio 2024, Rostalski 2024). The forms in which people’s need for protection manifest themselves are as plural as society itself: physical, psychological and social vulnerability, including a lack of protection or forms of political participation. The concept of vulnerability refers to a wide range of phenomena, and yet, it is a fundamental issue of human life that cannot be overcome despite all social efforts in medicine and technology. Rather, such developments continue to highlight the fundamental vulnerability of human beings and even create new forms of vulnerability (Dederich and Zirfas 2022, p. 2).

If we consider the general significance of vulnerability for human existence outlined above, it becomes clear why this topic is important for communication and media studies, which deal with communicators, recipients and media content of all kind. From a media ethics perspective, it is therefore desirable to give greater visibility to vulnerability as an anthropological constant in the media world. Media-ethical actions in this context should aim to “enable the communicative participation of all social groups in society’s process of self-understanding.” (Röben 2013, p. 10Footnote 1) The task of communication and media studies therefore is to investigate the extent to which this demand can actually be met. The goal is to identify particularly vulnerable groups and analyse what constitutes a ‘good’ representation of these groups from an ethical standpoint and how this can be ensured. Media and communication studies should engage more closely with vulnerable groups to make their lives, needs and challenges more visible, foster understanding and thereby lay the groundwork for participation. However, it is crucial to acknowledge that research involving vulnerable groups presents unique challenges and demands heightened ethical sensitivity. Studies involving groups identified as particularly vulnerable (Biddle et al. 2013; Bluvstein et al. 2021; Clark 2017), such as children and adolescents, individuals with disabilities or refugees, require specific ethical considerations. In addition to technical and legal issues, empirical researchers increasingly need to consider the context of a pluralist and participatory society (Pittaway et al. 2010; von Köppen et al. 2020).

This article highlights the importance and necessity of addressing issues of human vulnerability and the multiple ethical challenges associated with communication research with vulnerable groups. It is precisely in such complex ethical questions of principle, for example, in dealing with vulnerable groups, that the importance of ethical authorities becomes clear. These include ethical guidelines, codes of conduct, ethics councils such as institutional review boards (IRB), advisory bodies, and self-regulation instances as external frameworks, and the ethical skills for justifying individual ethical decisions. How these two areas are related will therefore be discussed in the next chapter (Chap. 2.1). As highlighted in the introduction, human vulnerability is a concern for everyone. However, for research contexts, this broad concept needs further differentiation. Chapter 2.2 therefore discusses further approaches to the concept of vulnerability. It becomes clear that a mere division into assigned groups such as ‘people with a migration background’, ‘children and adolescents’, ‘delinquents’ and ‘people affected by poverty’ can be problematic from an ethical perspective because it conceals intersectionality and the individual extent of vulnerability. In addition, other research participants may also be vulnerable in certain constellations. For this reason, this article also looks at the researchers themselves and project staff, such as coders. Building on this, Chap. 2.3 provides a theoretical basis for the design of an ethically-reflective research process in the form of principles, among other things, to provide initial indications of how moral decisions can be made in ethically challenging situations.

The overarching goal of addressing why the discussion of research ethics with regard to vulnerability is specified for the entire research process is to improve the quality of research results. To this end, the article develops concrete recommendations for action that can promote ethical sensitisation. The further structure of the article therefore follows the classic sequence of research processes in a practice-oriented manner, starting with the planning of the study (Chap. 3), followed by the sampling and data collection phase (Chap. 4) and their evaluation and publication (Chap. 5). The focus is on the two methods commonly used in communication studies: survey and content analysis (Altmeppen et al. 2011). Thus, our contribution focuses directly on the interplay between research ethics and a specific methodology. Recommendations for practice are derived (Chap. 6). An important basis for both the practical examples of ethical challenges and the proposed solutions based on them is a World Café that took place in the context of the annual conference of the ‘Methods’ section of the German Communication Association (DGPuK) in Potsdam in September 2023.Footnote 2

2 Ethical perspectives

2.1 Research ethics: more than legal standards

Few would dispute the importance of doing research right in a moral sense. However, it is not always clear what is ‘right’ and ‘good’ in a particular research situation. Values and norms that are used to justify (ethical) decisions and patterns of action can lead to different outcomes depending on how they are interpreted and weighted. For example, the central values of ‘freedom’ and ‘truth’ build an important ethical basis for science to prevent restrictions on research activities. However, these reach limits when the rights and dignity of other human beings are compromised by science and research (Fenner 2022, p. 234). Thus, the freedom of science and research cannot be considered absolute. Rather, it is important to consider which needs and demands must be taken into account in the research process on the basis of the legal, professional, ethical and individual needs of those involved, along with how these can be reconciled with the expected gain in knowledge.

In the context of communication studies research projects, ethical considerations often largely overlap with classical normative research paradigms (Schlütz and Möhring 2016, p. 483). Keywords, such as transparency and fairness, can be justified both in terms of communication studies quality criteria and from an ethical perspective. At the same time, however, the requirements of research ethics go beyond research standards and legal requirements for research processes. Legal requirements and professional codes provide only a narrow framework for what is considered ‘permissible’. This does not mean that any action that is not addressed is automatically considered “morally good or desirable” (Paganini 2020, p. 65).

While legal requirements and codes (e.g. of the DGPuK or the German Sociological Association (DGS)) provide an external framework and journals require certificates from ethics committees, researchers must apply ethical considerations and guidelines. The ability to ethically justify one’s own research activities is therefore becoming increasingly important for all those involved in the research process. This is all the more the case when the field of research concerns an ethically sensitive area or a particularly vulnerable target group, such as in research about sexual violations, bullying, hate speech or war and crises. In these contexts, ethics can also support the research process itself: It can help build trust between researchers and the researched. By slowing down the research process, ethical reflection processes also ensure that space is created for error checking and that the researchers’ individual approach to the actual research question is investigated and discussed. In this way, sensitivity to possible conflicts of interest can be created, and a critical attitude towards externally imposed expectations of results and standards can be developed. An ethically reflective attitude during the entire research process is therefore also desirable with regard to the quality of the research. In dynamic and complex research projects in particular, it seems important to demonstrate a certain basic ethical sensitivity, i.e. to have already reflected on ethical issues to be able to deal with sudden and unexpected ethical challenges confidently. Corresponding skills are also relevant with regard to the concept of ‘vulnerability’. After all, vulnerability is not always recognisable at first glance. Ideally, personal perspectives on this should be critically scrutinised throughout the entire research process. The importance of this will be explained in more detail in the next step.

2.2 Vulnerability: approaches to a complex term

The concept of ‘vulnerability’ is complex and linked to a number of ethical questions. The following explanations are intended to provide a brief classification of the term and justify our approach to it. A comprehensive ethical classification of human vulnerability can be found, for example, in Maio (2024). We refrain from applying a rigid definition of vulnerability; however, we see the point that this decision makes vulnerability a problematic ground for deriving a set of ethically sound strategies. In the following, we will first take a look at central approaches and understandings of the term and explore what these can mean for an ethical classification. When people with a migration background, children and adolescents, delinquents and people affected by poverty were mentioned as examples of vulnerability previously in this article, we singled out specific groups that, due to certain characteristics, are generally regarded as particularly worthy of protection. It was assumed that certain characteristics, such as a migrant background, limit and/or restrict participation in society. In general terms, vulnerability thus describes “the degree to which people, groups of people, societies, infrastructures, systems and (living) spaces are susceptible to risk or damage” (Burghardt 2018, p. 105). For research ethics approaches, the level of susceptibility to harm or the associated restriction of participation at which one can speak of ‘vulnerability’ at all remains to be discussed. Corresponding considerations also depend on cultural and historical contexts. Rostalski (2024, p. 9) notes that vulnerability is becoming increasingly prominent in contemporary social discourse. The potential vulnerability of individuals has entered public awareness and is being taken seriously, largely due to recent crises, such as the coronavirus pandemic. This shift is evident in the heightened efforts to protect the rights of socially marginalised groups.

Recognising and taking societal vulnerabilities seriously is not a given. Human vulnerability contrasts with the long-promoted ideal of a strong, sovereign individual (Stöhr et al. 2019, p. 3). Historically, even within science and research, the primary focus on ethical issues concerning vulnerable groups has not always centred on protecting them from harm or improving their conditions. Instead, history reveals instances in which science has actively reinforced exclusionary processes and upheld existing hierarchies (Dittrich and Radtke 1990, 11–12). This has exacerbated the vulnerability of certain groups based on factors such as gender, origin and faith (see Kühl 2018, pp. 105–111 for discussions on racism).

Research ethics efforts and guidelines as well as research procedures for dealing with vulnerability and identifying particularly vulnerable groups were originally discussed intensely in the field of biological and medical research. The term ‘vulnerability’ first appeared in the Belmont Report (1978) on research involving human subjects. With the requirement of respecting human dignity, the special protection of persons whose autonomy of action is limited was also elaborated in the course of the statement that potential participants in research projects should be regarded as autonomous actors (National Commission for the Protection of Human Subjects of Biomedical and Behavioral and Research 1978). Most of the research ethics guidelines and documents that have emerged since then have a common understanding of vulnerability based on the inability of certain groups of people to represent their own interests to the outside world. For example, in the Declaration of Helsinki (2008), vulnerability is differentiated both in terms of specific groups of people, such as those living in poverty or in need of care, and in terms of specific criteria, such as the capacity to give consent (Swartz 2011, p. 572).

Research ethics documents typically use a narrow but operationalisable concept of vulnerability to identify groups in need of special protection. Such an approach seems plausible, especially with regard to pragmatic interests in the research process. However, given the reality and the anthropological dimension of human vulnerability in general, the limitations of such a conceptual approach become apparent. Based on the idea of the gradual nature of vulnerability, another aspect that complicates the conceptual classification becomes clear. If we understand ‘vulnerability’ in the sense of ‘weakness’ or ‘fragility’ and the resulting need for protection, it must be stated with regard to humans as fragile beings, and the following must be taken as an indispensable part of life: “Vulnerability is a basic phenomenon of human life, and not only of human life, but of all life” (Coors 2022, p. 1).

Vulnerability proves to be a multifaceted concept that can be examined from various perspectives. Stöhr et al. (2019) provide an initial overview of key works in vulnerability research, with a particular focus on pedagogical issues. This article primarily aims to offer practical guidance for conducting ethically responsible research with vulnerable groups. It predominantly draws from approaches in medical research ethics but also incorporates Butler’s concept of vulnerability due to its significant relevance, especially in media and communication studies. Dedrich (2020) describes Butler’s approach to vulnerability as a two-level model. Beyond the general vulnerability inherent to all humans, Butler identifies (2005) a socially constituted vulnerability that is unequally distributed. This perspective necessitates considering vulnerability within specific social, political and cultural contexts. Thus, what is deemed vulnerable is always influenced by social circumstances and linked to specific subject positions, which are often based on historical and stereotypical criteria. This raises pertinent questions. What is socially recognised as vulnerability? And what happens if individual experiences of harm are not socially acknowledged as such? Butler’s approach emphasises that social recognition of vulnerability alters its meaning and structure. This has significant implications for research, as an overly narrow definition of vulnerability can be problematic. For instance, recognising group-specific vulnerabilities can paradoxically reinforce the very characteristics that lead to systematic disadvantages and discrimination, as noted by Dedrich (2020). This highlights the importance of a nuanced understanding of vulnerability in research contexts. Furthermore, Butler’s concept of vulnerability draws attention to our mutual dependence. Recognising our own vulnerability makes us aware of our need for protection and our reliance on others, just as they rely on us. Viewing others as deserving of recognition despite or because of their inherent vulnerability presents an opportunity for fairer and more equitable coexistence (Pistrol 2016, 259).

It is clear that an overly narrow understanding of the term entails the risk of losing sight of the levels and conditions associated with vulnerability and, not least, the qualities it conceals (Schnell 2017, p. 21). For example, the certified need for protection is often primarily due to political and social circumstances rather than belonging to a specific group defined as vulnerable. This can be seen in the example of migration. Here, it is certainly questionable whether the limited opportunities for participation can really be attributed to belonging to the ‘migrant’ group or whether they are instead due to problems of social justice and other levels, such as lack of language of host country, poverty or social recognition. An overly narrow understanding of vulnerability can therefore lead to important interrelationships being overlooked and the need for protection, lack of agency and opportunities for participation being reduced to personality traits (Wild 2014, p. 298). This seems all the more problematic as a certain form of vulnerability can result in new forms of vulnerability: “In certain constellations, there seems to be something like a self-reinforcing cascade of vulnerability” (Coors 2022, p. 2).

Despite the corresponding challenges associated with the idea of an operationalisable concept of vulnerability, the need for a comprehensible and, above all, applicable understanding of the term for research practice is key. An overly broad concept of vulnerability could obscure the fact that—even if it can be assumed that (human) life is fundamentally linked to its vulnerability—there are certain groups that require special protection and consideration. Luna (2019, p. 88) summarises this as follows: “This view ultimately ‘naturalizes’ vulnerability: if we are all vulnerable and vulnerability is a ‘natural fact’ that we all share, we do not need to avoid it or protect some people from it.” There are different approaches to making the concept of vulnerability tangible and categorising it. For example, a rough distinction can be made by trying to identify primary and secondary vulnerability factors (Dederich and Zirfas 2022, p. 3). In this sense, primary factors are fixed, unchangeable factors that are, for example, genetically determined. In contrast, secondary factors can be variable, for example, due to life crises or problematic environmental factors. While such a categorisation may seem obvious at first glance, it comes with the central challenge that such a categorisation must always remain blurred. Biological sex, for example, can certainly be identified as a primary vulnerability factor, but in reality, it is usually only a characteristic that causes vulnerability due to social and cultural circumstances (Dederich and Zirfas 2022, p. 3). One perspective is to try to summarise possible vulnerability factors and categorise them into specific domains. An example of such an approach from a biomedical perspective can be found in Kipnis (2001). He explicitly rejects the idea of associating vulnerability with specific groups of people. Instead, he describes the following six characteristics that he believes constitute vulnerability: cognitive, juridic, deferential, medical, allocational and infrastructural (Kipnis 2001, p. G6). By identifying the factors that underlie different forms of vulnerability in this way, research projects can go deeper and account for socially and culturally determined problems of justice. To address the complexity of the phenomenon of vulnerability, Luna (2019, p. 88) also makes explicit reference to the particular context in which the people concerned find themselves. If this is not included, there is a risk of stereotyping by assigning research participants to a particular vulnerable group and not taking into account the individual situation of the person. The observation that vulnerability is always linked to specific situations is another challenge for concrete research practice. What does it mean for the research process if the vulnerability of research participants is only identified during the actual research activity? How should this be handled? At this point, the relevance of an ‘Ethics of Practice’ and the ethical competence to respond flexibly to the ethical challenges of everyday research becomes clear once again (see Chap. 2.1). We acknowledge that vulnerability is a highly fuzzy concept that is, in some perspective, individualised and therefore starts at personal capabilities and circumstances. However, it is also a collective characteristic that pertains to social groups. This multifaceted phenomenon requires multi-dimensional responses, and this article seeks to help sort and overlook the palette of options and ensuing situations, focusing on communication research.

How can the concept of vulnerability be used in media and communication studies from a research ethics perspective? It seems useful to take the tension between broad and narrow approaches to the concept seriously. It is important to clarify which fundamental factors have led to the target group being described as vulnerable and to focus on the respective social, cultural and societal conditions. It is also important to be sensitive to the danger that the mere labelling of people as vulnerable can lead to a perpetuation of stereotypes and clichés, which are then transferred to the people concerned. At the same time, attempts to narrow the concept should not completely lose sight of the opportunities offered by a broad understanding of vulnerability. For example, the recognition of human vulnerability per se can also help justify the general human duty of care and responsibility to derive from it “a special duty to protect or else the duty to change social circumstances” (Coors 2022, p. 5)—not only of study participants but also, for example of student coders or researchers themselves. In this sense, the narrow and broad understandings of the term would be complementary—rather than opposite—approaches to an important field, which, in the first sense, provides perspectives for describing and investigating and, in the second sense, points to ethical-normative possibilities for action. To take into account the complexity of the concept of vulnerability, this article will therefore refrain from categorising it on the basis of selected groups. Instead, to discuss the various ethical challenges that may arise in the context of research with vulnerable groups in a practical way, the following explanations follow the classic sequence of research processes—from planning a study to the recruitment of participants and publication of sensitive personal data—and concern both quantitative and qualitative research (Scherer 2013).

To develop actionable recommendations, we first discuss potential principles and models that can address the relevant challenges in research practice. This discussion is guided by approaches of ethical judgment.

2.3 Ethical decision-making throughout the research process

In traditional research processes within media and communication studies, various responsibility groups and regulatory bodies become evident. Moreover, different dimensions of research ethics emerge. While central ethical challenges in upcoming research projects can be planned and reviewed in advance, such as through ethics committees, researchers often face additional ethically relevant situations during actual research practice. Guillemin and Gillam (2004, p. 263) categorise these dimensions into ‘procedural ethics’ and ‘ethics in practice’ or situational ethics. Both dimensions are essential to an ethically reflective research process.

‘Procedural ethics’ highlight that many ethical challenges in the research process can be anticipated. By acting with foresight, researchers can identify situation-specific challenges. Detailed research plans, the involvement of ethics committees and adherence to formal ethical standards help ensure that fundamental ethical principles, such as avoiding harm and obtaining informed consent, are maintained (Guillemin and Gillam 2004, p. 272). However, even with meticulous planning, unforeseen problems and ethical issues that are not addressed by formal guidelines can arise. In such cases, researchers must be able to respond appropriately and make ethically justifiable decisions. This is where personal ethical competence and judgment become crucial (Prinzing et al. 2020, p. 352). Ethical competence encompasses three dimensions: ‘seeing,’ ‘judging,’ and ‘acting’. Together, these constitute the decision-making process (Feeser-Lichterfeld and Heyer 2010, p. 156):

  1. 1.

    Seeing: This involves perceiving ethical challenges, which requires sensitivity to the situation and knowledge of relevant values and norms. Recognising and accepting one’s own responsibility is essential (Beck 2017, p. 82).

  2. 2.

    Judging: This dimension refers to the ability to weigh relevant values and justify why one value may take precedence over another and may guide a decision.

  3. 3.

    Acting: The decision made must be implementable in practice.

In applied ethics, various models support individual decision-making situations. A well-known set of ethical principles, originally proposed by Beauchamp and Childress (2019) for biomedical ethics, provides valuable guidance for research with vulnerable groups in media and communication studies. Beauchamp and Childress identify four central moral principles: respect for autonomy, nonmaleficence, beneficence and justice.

  1. 1.

    Respect for autonomy: This principle emphasises individuals’ self-determination and freedom of choice. In research involving vulnerable groups, this means ensuring that participants are fully informed and voluntarily participate without pressure or manipulation.

  2. 2.

    Nonmaleficence (non-harm): This principle calls for avoiding harm. Researchers must ensure that their methods and the publication of their results do not harm participants. This is particularly crucial for vulnerable groups who may be more susceptible to negative consequences, such as the risk of re-traumatisation during interviews.

  3. 3.

    Beneficence: In research, this principle means that studies should contribute to the well-being of participants and have a positive impact. Researchers should aim to produce findings that benefit the groups involved and design their studies to maximise benefits for participants and society.

  4. 4.

    Justice: This principle highlights the importance of fairness and the equitable distribution of research risks and benefits. When working with vulnerable groups, it is essential to ensure that these groups are not disproportionately burdened and that the benefits of research are fairly distributed.

Following Beauchamp and Childress’ principles of ethics in biomedicine, Paganini (2018) developed a set of principles for media ethics using the reconstructive method. These principles are directly relevant to the media sector and extend to the field of media and communication studies, as well as research ethics in general. Parallels between media creation and research practices are evident—from conducting research to engaging with protagonists, informants and interviewees and the publication process. Paganini identifies five key principles in media ethics: transparency, fairness, respect, responsibility and competence. The relevance of these principles to the research process is briefly explained here:

  1. 1.

    Transparency: Transparency is a central principle of research ethics and emphasises the need to make research processes comprehensible and to justify methodological decisions. It is crucial for ensuring research quality. In the context of vulnerability, transparency involves explaining one’s perspective on vulnerability, detailing how vulnerability was addressed during the research and how it influenced the chosen methods. Transparency, such as through clearly stating the study’s objectives, is also important in communication with vulnerable groups.

  2. 2.

    Fairness: This principle closely aligns with Beauchamp and Childress (2019) principle of justice, although Paganini prefers fairness for its less judgmental and complexity-reducing nature. In research, fairness involves treating all participants equitably and upholding the integrity and scientific ethos of researchers.

  3. 3.

    Respect: Respect is especially vital when researching vulnerable groups. It involves scrutinising the relationship and personal attitudes researchers hold towards their subjects, ensuring recognition and respect, and trying to reduce power asymmetries. Challenges may arise in approaching vulnerable individuals, where ‘othering’ might occur in sampling. The research process itself creates power imbalances, which should be acknowledged and managed without undermining respectful interaction.

  4. 4.

    Responsibility: Responsibility, while not prescribing specific behaviours, emphasises the importance of caring for others beyond the fulfilment of duty. In research with vulnerable groups, it highlights the ongoing responsibility researchers have towards their participants, potentially extending beyond the research context. This stands in relation with nonmaleficence and beneficence.

  5. 5.

    Competence: Paganini underscores the need for the ability to handle complex and rapidly changing situations. This is essential when dealing with vulnerable groups and varied research scenarios. Although principles can guide research, they cannot cover every situation. Hence, individual decision-making skills, training, supervision in anti-discriminatory settings and reflection on one’s normative positionality as a researcher are crucial for effective research practice.

In the previous sections, we examined selected theories and approaches to demonstrate how ethical considerations in the research process can aid decision-making. In the following sections, these theoretical frameworks will be applied to practical experiences. The goal is to develop recommendations for conducting research with vulnerable groups by integrating both theory and practice within the context of holistic research ethics. To this end, possible challenges and recommendations will be outlined along the typical course of research processes. To streamline the research process, we divided it into three phases: the ‘pre-study phase’ (research interest, examination, preparation), the ‘study phase’ (data collection) and the ‘post-study phase’ (data evaluation, publication). Each phase presents distinct situations in which ethical decisions must be made and focus on either 1) vulnerable groups as research participants or 2) vulnerable groups as involved assistants/researchers. Recognising these situations is the first step in exercising fundamental ethical skills and taking responsibility.

Responsibility in the research process is multifaceted, encompassing both individual actions and the broader role of researchers. Researchers must be accountable for their actions (or inactions) concerning all participants and all stages in the research process (Beck 2017, p. 82; Edwards et al. 2021). Sensitivity to vulnerability is crucial and requires researchers to critically evaluate their own perspectives on participants’ vulnerabilities. This critical evaluation is essential for making informed judgements and taking appropriate actions. Therefore, it is vital to identify who may become the object of responsibility during the research process. This includes not only the researchers but also the study participants and, in the context of content analyses, the content creators whose media are evaluated. During data evaluation, responsibility extends to coders or student assistants. In the publication phase, responsibility again shifts towards the recipients of the research findings and the study participants (Fig. 1).

Fig. 1
figure 1

Constellations of responsibility in the ethical reflection on vulnerability

Full size image

3 Planning research (pre-study phase)

In the phase of planning research, various crucial points have to be considered. This section reflects on questions of accessibility, principles of participatory research, preparation of documents that have to pass the data manager and ethics commission (IRB) of the researcher’s institution, field access, training on intercultural competence and incentives. As a disclaimer, we have to state that in all areas of ethical considerations, self-reflection is key. The following explanations are therefore always bound to a process of constantly calling procedures and decisions into question in the light of ethical reasoning.

Planning research with vulnerable groups means to think about who might be excluded (justice/fairness) already when formulating the research interest and examining a topic’s conditions. Reasons for exclusion can include poor eyesight, hearing impairment, poor language skills or a lack of writing skills. For example, when preparing an online survey, it must be designed according to the European Accessibility Act. In the case of a survey being distributed in an elementary school class, some pupils might not be able to participate as they cannot read the questions. This can lead to a high level of frustration in these children (nonmaleficence/non-harm). Further, researchers have to think of the appropriate level of participatory research (Wansing et al. 2022; Wilkinson and Wilkinson 2018) or even inclusive research (Buchner et al. 2016) that they might want to reduce power asymmetries and ‘learn’ from the ‘research objects’ that will then be regarded as partners. The question behind this approach is: who should benefit from academic knowledge (beneficence) and which modes of knowledge production should be used in science (transparency) (Schuppener et al. 2016). The aim is to emphasise mutual learning so that participants do not feel like passive subjects and instead feel like active participants (fairness).

The timely and target group-specific planning of research proves to be central as vulnerable groups are often difficult to access, or participants might mistrust the research process. This might be because they had negative experiences in the past that led them to become disenfranchised or because they were subject to harm, stigma or discrimination, and some might even be hiding in remote areas because they fear persecution (Ellard-Gray et al. 2015). If confronted with intersectional identities (e.g. disabled migrants or gay prison inmates), this might become even more difficult, as possible participants might see greater risks in sharing sensitive information. In any case, “researchers need to ensure that recruitment is inclusive of all who fit within sampling parameters” (Ellard-Gray et al. 2015, p. 4) (justice/fairness). Ellard-Gray et al. (2015) suggest working with broad labels in recruiting materials (e.g. different identity-open labels for ‘gay’ or ‘migrant’) and avoiding stigmatisation (nonmaleficence/non-harm/responsibility). Further, using easy to understand/plain language in as early as the recruiting phase can diminish participation barriers. Sometimes, it is even mandatory when distributing data privacy consent forms (transparency).

Choosing the right location, e.g. for an interview, is another point to be considered especially in research with vulnerable groups. For instance, one’s own home might be considered a place of safety and trust and might be the right place to talk about gaming addiction; “such an in situ approach to interviewing is safer and more appropriate for the refugee context than structured meetings where they are removed from the only community they have in the midst of displacement.” (Akesson et al. 2018) Here, it is important to perceive and recognise the needs of the person (respect/autonomy). Prior to conducting research with a vulnerable group, it is essential to get acquainted with the specific needs and requirements of the respective group, e.g. take part in intercultural training before talking to people of different cultural backgrounds, keeping in mind that cultural homogeneity within the researcher’s team might increase power imbalances (Akesson et al. 2018) (responsibility). This includes competence in addressing vulnerabilities without at the same time re-emphasising prejudices. To see the individual is a sensitive precaution to avoid over-generalisation. Thus, blunders such as not taking off shoes when entering Arabic families’ homes can be avoided. In addition, one can learn about, for example, how to encourage a Muslim wife to more actively take part in an interview if the husband happens to dominate the interview situation. In one of our projects (Rothenberger et al. 2019), we were advised by a native worker to go into Arabic families with a male and female researcher to keep a gender balance and ensure that neither the Arabic wife nor husband feel offended (respect). In addition, when considering working with disabled people, pre-training and exchange with experts are mandatory. Thus, researchers feel better prepared when presumably ‘awkward’ situations challenge their competency to conduct an observation (e.g. someone permanently standing at one’s side and talking during an observation in a sheltered workshop or trying to touch you) (competence). If possible, it always is advisable to “engage with community organizations and leaders in order to effectively enter the community and build trust between the research team and the research participants” (Akesson et al. 2018) (transparency).

Several studies have dealt with the question of incentives in survey research (guided interviews or online surveys) as this aspect and its pros and cons are not often debated in ethical guidelines: “there is a lack of guidance on payments in the ethical guidelines provided by various research associations” (Ellard-Gray et al. 2015, p. 335; Head 2009). Incentives are most often used to increase response rates, not so much out of courtesy and ethical consideration (the interviewee gives something to me, the researcher, so I should give something back); they could also be seen as a restriction of autonomy. If monetary incentives or gift vouchers are the only motivation for people to take part in research, this might have an effect on recruiting strategies and willingness/enthusiasm of answers. “Some ethical committees regard the payment of money as an apparent inducement to participate in a study and as being unacceptable, because the money may influence the participant’s autonomy and ability to consent freely to participate; in other words, it could be viewed as a form of coercion to participate” (van Wijk and Harrison 2013, p. 575). The researcher should also consider providing transportation and childcare reimbursements (justice/fairness by enabling participation). Whatever incentive paid, it should be mentioned in the research report, and the advantages and disadvantages (e.g. distorting the principle of free consent to participate/feeling of coercion to participate, especially if dealing with vulnerable groups subject to poverty) should be discussed critically. In all these considerations, the researchers should always reflect on their role, and the wellbeing of the future participants should always be an irrevocable premise of research (beneficence).

In addition to the personal ethical perspective, one has to consider that there might be legal and formal requirements set by the university or institution funding the researcher (responsibility/competence), which are addressed by awareness of procedural ethics. Many universities nowadays have ethic committees/institutional review boards (IRBs), and some journals even require IRB statements with the submission of manuscripts of research involving human beings. Moreover, funding institutions often require a data management plan that has to be coordinated with the data security officer of the researcher’s institution. Here, the researcher has to show in advance how one will deal with data issues and data privacy in all stages of the research, from recruiting to interview recording, data processing and storage (transparency). Going through these administrative processes might be time-consuming and has to be considered at the very beginning of the planning phase of the study.

4 Sampling and data collection (study phase)

As clarified in the section on designing a study, research ethics considerations are essentially designed to reflect on the procedures used by researchers to ensure the integrity of all participants in the research process.

In the phase of preparing and conducting data collection, there are two aspects that require particularly critical self-reflection on one’s own role in the context of research with vulnerable groups (competence): (1) the question of a possible power asymmetry, which concerns the principles of respect, fairness and justice (see Chap. 2.3), and (2) the need to ensure voluntary participation to guarantee the autonomy of the involved parties. First and foremost, this requires researchers to take responsibility for the transparency of the research process. It is important to recognise power asymmetries between all kind of research participants. In this context, one must not only question their own embeddedness in (hierarchical power) structures but also ensure that study participants are not perceived as mere research subjects and objects of investigation but are treated as partners with their own agency. To respect the autonomy of participants, it is also important for all parties involved to ensure that participation in empirical social research is always voluntary, intrinsically motivated and non-violent (do no harm/nonmaleficence). The central instrument here is the declaration of consent in the form of “informed consent” to secure transparency, which is usually obtained in writing prior to data collection. Compliance with the legal minimum standards (for Germany: DSGVO 2021, Kap. 2, Art. 7) is less problematic in most cases due to standardised documents and procedures. It becomes more challenging when it comes to creating real transparency about the objectives and content of the research project, as well as about the rights and use of the data (transparency). Thus, informed consent goes beyond the clarification of legal issues. It includes the provision of information to participants about the purpose of the research and its procedures, potential risks, benefits and alternatives to ensure that they are able to make a voluntary and uncoerced decision. It must be ensured that “individuals control whether or not they enrol in clinical research and participate only when the research is consistent with their values, interests and preferences” (Emanuel et al. 2000, p. 2706). In research designs involving vulnerable groups, it is often particularly challenging from a research ethics perspective to create a starting point that not only guarantees comprehensive information but also takes responsibility for ensuring that participants really gave consent voluntarily (competence/autonomy/respect).

We will explore these aspects in more detail below, with a focus on the two methods chosen for this article: survey and content analysis. Although methods from the qualitative paradigm are often used in connection with vulnerable groups, quantitative research must also be considered in the context of ethical challenges, especially if, as in this article, a broad, flexible and, above all, situational understanding of vulnerability is advocated. We will do this by taking examples of communication research concerning different vulnerable groups.

4.1 Surveys and guided interviews

There is a relatively large body of research on the ethical challenges of collecting data for a survey (e.g. Friedrichs 2019; Liamputtong 2007), in part due to the interdisciplinary nature of the method. If it can be determined before or during data collection that vulnerable people are involved in the research process, ethical considerations need to be taken into account with particular care. The ethical challenges that arise in each case may vary depending on the reason for the vulnerability.

The design of the research situation itself is crucial to scientifically meaningful and ethically appropriate data collection. For example, a study involving school children (Riesmeyer et al. 2023) may require ethical considerations concerning the setting in which the survey can take place during the school day to accommodate the needs of the pupils. Is the classroom a good setting because it is familiar? Or is it perhaps too closely associated with classroom rules that trigger certain role behaviours in the pupils? For the same reason, whether the presence of the teacher might trigger reassurance or bias in the interviewed children must be considered (competence, responsibility). Adolescents are also a potentially particularly vulnerable group. This is true not only because of their age but also because of particularly sensitive research interests in communication studies, such as the self-presentation of young people in social networks (Zillich and Riesmeyer 2021), the use of right-wing extremist online media (Pohl and Riesmeyer 2023; Reinemann et al. 2019) or recent experiences of violence or forced migration (Rothenberger et al. 2019). Questions of anonymisation arise not only in advance of conducting the study but also during data collection, especially in cases where problematic, self-injurious or even potentially criminally relevant behaviour can be identified (responsibility).

Ethical challenges are exacerbated in many ways when collecting data from migrant populations, especially if they have a history of forced migration (von Unger 2018). Proponents of critical qualitative research suggest that attention be paid in the data collection phase to the various issues of power and unequal power relations between researchers and research subjects that reflect existing political, social and possibly even global differences (Lammers 2007; Marshall and Batten 2004). Migration studies are therefore an area where the principles of justice, fairness and respect are both an issue and a challenge.

The spectrum of consequences of these power asymmetries in research with vulnerable groups is broad, affecting both qualitative and quantitative studies, and ranges from an intercultural variant of social desirability in standardised surveys to re-traumatisation in interview situations, especially when studying war-affected populations or people who have suffered physical violence (nonmaleficence). In an online evaluation of an educational program for refugees, the almost unqualified praise suggests that participation in independent empirical social research is either not a known social practice in their countries of origin or that asylum seekers are afraid that critical statements might jeopardise their procedure because they cannot assess the political consequences (Dechau et al. 2017). The legal limbo in which refugees find themselves after their arrival in their new host country creates uncertainty with regard to contact with official bodies and social elites, including academics—especially if they appear to belong to the western white majority society.

This does not necessarily have to be the case, but it can lead to overly conformist behaviour in certain settings during the survey. This is also true for migrant respondents, who may adjust their responses to conform to assumed conventions based on perceived cultural differences. A study on media consumption in Arabic-speaking families in Bavaria showed that this was the case on several levels (Rothenberger et al. 2020). In this example (and we acknowledge that it might be different in other Arabic families), the willingness of female family members to speak seemed to be inhibited not only by the role of the husband or father, who often dominated the conversation on the spot (see Chap. 3), but also by the visit of the research team, which usually includes at least one interpreter in these foreign-language constellations and thus brings several strangers into the home, a protected private space. In such cases, it may make sense in terms of the autonomy of the participants to examine the advantages and disadvantages of different interview situations and, if necessary, switch from a face-to-face interview to a video or telephone interview during the survey (Götzenbrucker et al. 2022) (competence, respect).

Ethical research therefore requires special competences not only to constantly reflect on one’s own actions (Berger 2015) but also to respond to and prepare flexibly for acute challenges that arise in different situations (situational ethics). These barriers can also be linguistic in nature and have different causes. Participants who speak a different native language, as well as very young people or those with lower educational levels or cognitive impairments, benefit from the use of simple or plain language (Mälzer 2016). Since 2022, a project at the University of Münster has been investigating how publicly funded empirical social research can be made more accessible and thus more inclusive (justice/fairness). For example, online surveys often use scales that are difficult for people with reading disabilities to understand or that people with visual impairments sometimes cannot complete (Gehrau et al. 2022).

All these examples and efforts are aimed at ensuring that research participants, against the background of their autonomy, are linguistically and socially capable of deciding whether to participate in a study without being forced to do so. Informed, voluntary participation is a central prerequisite in the context of vulnerability. As a dimension of power, informed consent generally means obtaining the formal agreement of potential participants (Tuckman 1999). In order to help vulnerable groups accurately assess their consent, documents must first meet three linguistic criteria: understandable, comprehensive and concise. Finding the right tone is usually a balancing act, because “just as overzealous protection is paternalistic, overly simplistic language is disrespectful” (Szala-Meneok 2009, p. 512) (respect). For this reason, alternative forms of consent, such as verbal, must always be discussed, and consent should also be renewed at various points in the research process through either re-consent (Szala-Meneok 2009) or process consent (Emanuel et al. 2000; Mackenzie et al. 2007). Vulnerable people must therefore always have the option during data collection to either skip some questions they feel uncomfortable answering, take a break or even end the survey situation altogether should the need arise (Akesson et al. 2018). This needs to be done on a case-by-case basis (situational ethics). Meeting with a counsellor or including a debriefing session may help prevent the survey from being terminated. This option to prevent or at least reduce potential harm (nonmaleficence) was also discussed with refugee women in a research project investigating refugees’ representation in the coverage of the five most popular international news channels in Europe. The group discussion included stimulus material from the content analysis, whose content could have had a potentially discriminatory or, depending on the individual history of the participants, a traumatising effect (Rothenberger and Schmitt 2023).

Ethically responsible data collection in the context of vulnerability must therefore be understood first and foremost as a fluid concept “that requires constant re-examination and redefinition within informed consent, which is viewed and implemented as an ongoing process” (Marshall and Batten 2004).

4.2 Content analysis

Looking at content analysis as a method, it is true that it is very closely related to research areas in communication studies, as the analysis of journalistic and media messages makes up a large part of the empirical work in the discipline (Altmeppen et al. 2011). However, it is less obvious in which constellations content analysis, which is generally considered a non-reactive method (Rössler 2017), poses ethical challenges in research in terms of vulnerability. Nevertheless, sensitive situations may arise for different groups involved, especially during the collection of content analysis data. These include, first, the researchers themselves and, to an even greater extent, students and student assistants involved in data collection, for example, as part of classroom research projects. Second, under the conditions of collecting digital traces of all kinds, the authors of media content are increasingly under scrutiny. Nowadays, social media users become data providers without being asked. Their informed consent to publishing the content in the fragile online public sphere where the content is permanently available for research can at least be questioned (fairness).

Against this background, the concept of vulnerability takes on a new connotation and a different extension because it is often not so easy to predict in advance who will become vulnerable in the course of a content analysis, or when and for what reason they become vulnerable. This can happen because, for example, the text and/or image material to be analysed is stressful to the individual. When using content analysis in research designs, there is a special and novel need for the ability to recognise vulnerability (competence) and to ensure the ethical standards of non-harm and transparency. Researchers are not just responsible for paying attention to self-care; as study directors/principal investigators, they also have a special duty of care towards their coders, as there is a hierarchical relationship of dependency, especially in constellations where students are assessed based on their participation in the research, and this performance may even be relevant for passing the course. In such contexts, it seems extremely important from an ethical perspective to establish an open and trusting culture of respectful communication from the outset, raise awareness of psychological stress, offer supervision during or after the coding work and establish clear rules of the game. For example, in a research project on the media representation of refugee women (Rothenberger and Schmitt 2023), during the coding process clear stop signals were agreed upon. If a YouTube video had a trigger warning at the beginning, for example because the content showed violence against refugee women, the coders were to skip the analysis unit and mark it accordingly so that the research supervisor could decide at a later stage whether to expose herself to the material (similar, e.g., Meltzer 2023).

For a long time, the authors of media messages of all kinds were less in the focus of ethical considerations. This is partly due to the fact that, in the pre-digital era, ethical analysis tended to focus on journalistic reporting or on the texts of professional communicators, such as politicians. It could be assumed that these messages were produced in the knowledge that they would circulate in the public sphere and that there was therefore conscious consent to their publication. However, new ethical questions of responsibility arise when a research design includes the analysis of semi-public spheres, such as internet forums or social media postings, instead of mass media content. The ethics of internet research must address issues, such as the nature of public spaces (McKee and Porter 2009; Schlütz and Möhring 2018). For example, even if content is shared on social networks by users whose privacy settings are formally stored as public, the ethical question arises as to whether users really wanted to give their informed consent for their Facebook statement to be made available for research reuse for an indefinite period of time. The discussion on data donation (users let apps store their usage data on their mobile devices and transfer it to the researchers’ databases) takes the autonomy of social media users into account and seeks to obtain the consent of the originator to make sure that the use of the data is transparent.

The permanent availability and automatic discoverability of one’s own media messages, which are often written spontaneously, situationally and emotionally, ensures that private individuals as authors of media content online can, under certain circumstances, be classified as vulnerable. A case-related ethical review would also be necessary when analysing digital trace data, but this is hardly or not at all possible given the collection of huge datasets in automated processes (e.g. scraping). Ethical considerations regarding content analysis in the digital space lead to the following questions. Can informed consent be obtained retroactively? What harm can the use of online statements have for their originators? How can users withdraw their consent to data use in the sense of re-consent? Finally, there should at least be an open discussion about whether there needs to be an expiration date for publicity online. Do users perhaps even have the right to be forgotten? The questions of how researchers in digital media environments can fulfil their responsibilities to social media users and how to ensure the principles of nonmaleficence and respect for autonomy need to be an integral part of big data research.

5 Data analysis, data processing and publication (post-study phase)

After data acquisition is terminated, researchers enter the process of data wrangling, data storage and data analysis. They translate, transcribe and re-hear audio-recorded interviews or video material prior to data analysis. Here, various challenges occur, especially with regard to treatment of data from vulnerable groups. One scenario might be that a person withdraws their consent, which leads to the deletion of all their data (respect for autonomy). Another scenario might contain bad articulation/pronunciation and thus poor intelligibility of interview data (e.g. due to language barriers or physical disability). Thus, the question of whom to consult while not breaking the rules of confidentiality arises (responsibility). In general, it is essential to maintain confidentiality (Surmiak 2018), e.g. when vulnerable people hide their identity because they are member of a stigmatised group. Confidentiality can thus be considered crucial at all stages of the research process, including data archiving and re-usage. Sometimes, the researcher involuntarily gives meaning to sentences that was not intended by the participant. Here, reflecting one’s own role and always questioning whether instrumentalisation might occur are essential (competence).

This is also accomplished by anonymising or pseudonymising research material such as names, places, and occupations as well as by password-protection of documents to avoid unauthorized access. Special care and caution are required when working with spatial/geodata (GIS data/GPS tracking), e.g. with migrants that are not allowed to leave the district or county they are bound to. However, data protection has to be implemented in such a way that traceability and transparency of rigorous scientific research are ensured (having ‘proof’ that the research project was conducted). Sometimes, the participants and student-researchers need a debriefing. The researcher should offer to be a contact person beyond the time of the project (aftercare/follow-up support/responsibility).

Especially in cases of emancipatory or engaged research, the following question arises: Shall the publication lead to improvements for the vulnerable population (beneficence)? For example, in the case of giving voice to imprisoned terrorists, reflexivity proves to be a helpful tool in understanding research ethics and becoming aware of one’s own responsibility and limited perceptions as a researcher. Researchers should be aware that with their publication, they make certain populations more visible, which might have positive as well as negative implications. Some researchers are convinced of initiating change with their results: “we have made the dissemination of our findings a priority. After completing data analysis, we have begun to engage in multiple methods of knowledge mobilization to share these experiences with a broad audience and potentially affect change.” (Akesson et al. 2018) Openly accessible data can also benefit vulnerable groups as the data can be used for future studies by other researchers. However, informed consent has to be given by all participants. In any case, to communicate the results to the participants or the affected community should be a priority (respect).

When writing journal articles, researchers are confronted with journal policies that require a paragraph about ethical aspects and challenges (which is also a point in reviews) or a statement that inclusive language guidelines were followed (e.g., Intellect Style Guide, pp. 55–70Footnote 3). Some journals require IRBs’ consent statements. It is important not only to follow ethical guidelines’ rules (procedural ethics) but also to share experiences with ethical challenges that could not have been anticipated and needed ethical judgements in a certain situation (situational ethics).

6 Recommendations and implications for future research

In the last three chapters, the ethical principles systematised in the theory section were linked to examples from research practice. We discussed various ethically relevant situations encountered in research involving particularly vulnerable groups (‘seeing’; Feeser-Lichterfeld and Heyer 2010). By developing a fundamental sensitivity to these situations and critically reflecting on individual perspectives on vulnerability, researchers can apply principles from Beauchamp and Childress (2019) and Paganini (2018) to make responsible decisions during the judgement phase (‘judging’) that might lead to implementation in practice (‘acting’). Practical application reveals that these principles can dominate different scenarios. We suggest dividing the list of principles into four clusters (Fig. 2). First, in direct interactions with research participants, the principles of ‘respect’, ‘autonomy’ and ‘justice’ or ‘fairness’ are crucial. Second, in a broader context, the principles of ‘nonmaleficence’ and ‘beneficence’ gain importance. These principles underscore the need to consider and address the fundamental objectives and consequences of the research for those affected throughout the implementation process.

Fig. 2
figure 2

Cluster of principles of ethical reflection on vulnerability

Full size image

Third, Paganini’s principle of ‘transparency’, which is closely related to values such as ‘truth’, ‘truthfulness’ and ‘accuracy’, is significant for both internal and external communication in the research process. Communication with all study participants should be guided by a commitment to truth. This principle also applies to the publication of the final results, where a balance between truth and non-harm must be maintained, particularly concerning sensitive participant data.

Fourth, the principles of ‘responsibility’ and ‘competence’ are specifically relevant to researchers and broader research organisations. Researchers must possess basic ethical competence, including knowledge of central ethical principles and methods of ethical judgement and a critical understanding of the concept of vulnerability and its challenges, i.e. applying procedural and situational ethics. Researchers are ultimately responsible for recognising ethically challenging situations, taking them seriously, making informed decisions and acting accordingly, with the obligation to justify their actions (Fig. 2).

Research ethics is always a balancing act between the interests of free science in the pursuit of knowledge and the protection of human subjects. Because each situation is unique, there are no prescriptive guidelines for how a research project can be conducted in an ethically correct manner in all cases. Rather, we need a heuristic that can serve both as a “code of conduct for researchers” and as a “norm-guided review of research processes” (Krause 2023, p. 202).

Ethical challenges in communication research arise at all stages of the research process, for both qualitative (von Unger et al. 2014) and quantitative (Maxwell and Kelley 2011; Scherer 2013) study designs and for all methods used in our discipline, such as content analysis, survey, guided interview, focus group or observation (Friedrichs 2019). As demonstrated, there is a manageable but certainly audible critical discussion of research ethics issues in the field (e.g. Heise 2017; Prinzing et al. 2020; Schlütz and Möhring 2016). However, proposals for dealing with vulnerable groups in research have so far come from other disciplines (Biddle et al. 2013; Bluvstein et al. 2021; Clark 2017; Swartz 2011; von Köppen et al. 2020). This may be due to the fact that communication studies questions outside of experimental settings, e.g. on media effects, do not per se seem to be suspicious of addressing sensitive areas of life, as is more widely accepted in psychology, sociology or social work. Two developments mark this assumption as fallacy. First, a society understood as plural and post-migrant (Foroutan 2016) increasingly brings marginalised groups into the public eye and thus ushers their media activities into the focus of communication research. Second, the rapidly changing object of study under the conditions of digitalisation and its research with digital methods constantly raises new ethical questions, such as about the self-determination of internet users and new vulnerabilities that arise from the permanent storage and replicability of digital data (Schlütz and Zillich 2023).

Against this background, and with the intention of supporting researchers in their ethical reflection, there is a large number of guidelines and codes, promoted by German-speaking professional societies (e.g. DGPuK 2017; DGS 2017; DFG 2019) and internationally, with a particular focus on vulnerability (for an overview, see Inter-Agency Standing Committee [IASC] 2007; van Wijk and Harrison 2013). It is precisely this link between theoretical and practical knowledge that is essential to grasp the concept of vulnerability, whereby despite the academic debates “the descriptive and normative meanings ascribed to the concept have remained disconnected from the perspective of the users of the concept and those affected by its use” (Lajoie et al. 2020, p. 128). Our article should therefore be understood as an attempt to contribute to making vulnerability fruitful both as a theoretical concept and as an ethical premise for action in communication research (see Figs. 1 and 2).

For this reason, we would like to add some recommendations for action to the existing canon and reemphasise selected ones. The list is based on the sequence of a prototypical research process and is inspired by the discussions during the workshop at the ‘Methods’ section conference:

Comprehensive research process

  • Vulnerability is not dependent on paradigms and methods but affects the entire research process, including the choice of theory, the construction of the instrument, the access to the field and the data collection phase and the ethical reflection of the results with regard to the political and social consequences that the research findings could have for the vulnerable people involved.

Pre-study phase

  • Careful advance planning (procedural ethics): It is important to consider the resources required for ethical support.

  • Ethics is not only a question of desire but also a question of time, personnel and, not least, financial resources. These aspects have to be taken into account when applying for third-party funding and have to be shown as cost items.

  • Research with vulnerable populations is time-consuming. The time factor should be taken into account, especially when preparing for the research, e.g. time for trust-building with potential participants.

  • Personal approaches and skills acquisition (responsibility and competence).

    • Ethical reflection always involves reflection on one’s own positionality as a researcher. If one’s own value structure, religious and political beliefs, prejudices and privileges are not critically questioned (Loue and Loff 2013), ethical research often remains an empty phrase rather than a real conviction. This process is only successful if it is uncomfortable.

    • No guidelines and no ethics committee accompany the researcher into the field. There is a difference between procedural ethics and ethics in practice (Guillemin and Gillam 2004, p. 263–265). This means that ethical decisions always have to be made in the specific situation itself, because ethics committees cannot give advice for unforeseeable situations. This is particularly true when working with marginalised groups, given the facets of their vulnerability and their intersectional connections. However, this also occurs in research projects that do not initially appear to be ethically challenging, when initially unknown stressors or traumas are expressed during the process.

    • Integrating ethical reflection into methods training and other university courses is an investment in the future (see the following for suggestions for German universities: https://www.forschungsethik-kmw.de/handreichungen).

Study phase

  • Direct interaction with stakeholders (respect of autonomy, justice and fairness).

  • Research with vulnerable populations requires a sensitive approach to language. How much technical language can I assume? Where can I get help translating my consent form into simple language? Is it necessary to use an interpreter? Researchers can also be or become vulnerable and may need protection, institutional support and, increasingly, self-care skills (Bluvstein et al. 2021). This is true in the context of confrontation with stressful media content and in the balancing act of maintaining scientific distance while still being emotionally involved in the context of engaged research, for example when a person in the sample is deported.

  • Engaging in research with particularly vulnerable target groups might involve promoting greater social participation and strengthening their visibility. This approach does not conflict with the researchers’ neutrality, which is intrinsic to the research mandate. However, the research methods may need to be reevaluated and adapted to address and prevent power imbalances that can emerge during the research process. The distinct roles of the researchers should be carefully considered.

  • Consider the consequences of the research process for the participants (nonmaleficence/non-harm/beneficence).

    • Visibility can serve as a tool for participatory justice and for highlighting the importance of vulnerability in life. In our view, this approach differs fundamentally from merely thrusting individuals into the spotlight for display purposes. A critical aspect to consider is the continuous need to evaluate the potential consequences of such visibility. Additionally, it is crucial to be mindful of the risk of re-traumatisation in interview settings.

  • Designing communication processes (transparency and truthfulness).

    • To address the existing power imbalance in the research process, ensuring transparent communication between participants is crucial. The objectives and procedures of the research must be openly conveyed, promoting clarity and mutual understanding.

    • The research procedure, including preliminary assumptions, personal approaches and ethical standards, should be meticulously documented and shared both internally among the research team and externally with the wider community. This comprehensive communication fosters accountability and trust.

    • Additionally, challenges encountered during the research should be openly discussed. By sharing these difficulties, researchers engaging in similar inquiries or utilising comparable methods can benefit from these insights, ultimately enhancing the overall quality and robustness of future research endeavours.

Post-study phase

  • Committed research also takes responsibility after the field phase and publication (nonmaleficence).

  • What benefit does my research have not only for the scientific interest of knowledge and my personal career but also in terms of positive and negative impacts on the vulnerable group being studied?

  • Many recommendations for action lack precision and concrete references (transparency).

    • It is therefore important to create transparency about the ethical challenges in the various projects, record them in detail and make them a relevant part of the presentation of results. Only in this way can they help future research and raise awareness of ethical issues in the field.

In times of polarised discourse in a changing public sphere, communication research oriented towards a pluralistic society requires not only new theories and methods but also, above all, a high degree of critical self-reflection (Guillemin and Gillam 2004) and sensitive antennae for old and new vulnerabilities in this mixed situation. In addition to the ethical challenges that arise in research contexts with vulnerable people, which have been discussed extensively in this article, there are also considerable opportunities for the vulnerable group itself, since one of the goals of engaged scholarship is “to give vulnerable people a voice” (Liamputtong 2007, p. 21). Research designs with marginalised groups offer the opportunity to gain insights into life-worlds that otherwise remain closed to us as predominantly white and privileged members of so-called majority society. In this case, however, it is always necessary to weigh the need for protection of the individuals involved against the potential benefits of the research for the vulnerable groups and for science. At the same time, care must always be taken to ensure that protection does not become overprotection or even paternalism, since a classification of ‘vulnerable’ may protect individuals while also somehow limiting their agency (Loue and Loff 2019, p. 7).

The following diagram (Fig. 3) offers a comprehensive overview of the central practice-oriented guidelines developed in this article. These guidelines are organised according to the different phases of the research process. They are designed to assist in decision-making during ethically relevant situations in research practice, in alignment with the principles presented earlier (Fig. 2).

Fig. 3
figure 3

Framework and steps of ethical reflection on vulnerability in the research process

Full size image

7 Conclusion and outlook

We have written this text in the knowledge that we represent a privileged perspective. This self-reflection, as we have noted in various places, is not only a basic prerequisite for planning and conducting empirical studies; it also applies to the writing of review articles or theoretical essays. We have emphasised that there are guidelines according to procedural ethics, such as the International Communication Association (ICA) Code of Ethics (2019); however, this code refers to vulnerable groups only once: “ICA expects researchers to take care to address the specific needs of vulnerable populations in their research, such as children, refugees, people who are mentally unwell, etc.” (International Communication Association 2019) Furthermore, it does not refer to specific challenges of methods, such as surveys or content analyses.

Building on this, we find that a common ethical guideline, or at least key questions for research with vulnerable groups, across the communication research sub-disciplines would be helpful. With our article, we tried to map the field of possible interventions and points of self-reflection. In this conclusion, we want to reference the following (slightly adapted) questions of Liamputtong (2007, p. 7) as a starting point for further developing an integrative approachFootnote 4:

  • What makes your participant group different to others?

  • How might their vulnerability hinder you in accessing the field? Will safety issues play a role?

  • How can you ensure that your research will not lead to further marginalisation of this group?

  • What research method would be the most appropriate to research your envisaged vulnerable group?

With regard to the last question, Liamputtong (2007) makes innovative suggestions that go beyond the traditional (single, focus group or conjoint) in-depth interviewing methods, such as arts-based methods, video diaries or reflexive photography (Liamputtong 2007, p. 140–163). Additionally, instead of conservatively written research findings, she suggests new forms, such as staged plays or poetic representations (Liamputtong 2007, pp. 164–188). Researchers also have to be aware of new technological possibilities, such as geo tracking or facial action recognition, that might be especially dangerous for vulnerable people (e.g. refugees). Efforts in getting to more equal terms become visible in the notable shift towards greater collaboration in research, specifically in participatory research with participants who may be considered vulnerable, marginalised or socially excluded. This has provided “new opportunities for equalising, and even transposing, researcher-participant relationships and facilitating participant ‘voice’” (Aldrige 2014, p. 1).

Researching vulnerable individuals needs especially responsible and sensitive researchers. Moreover, formal support guidelines to assist the researcher and student-researchers should be developed, such as forming discursive teams of mutual exchange and support. In general, the ethics of research with vulnerable groups should be a topic not only at conferences but also in communication studies education.