No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
AI-replicas as ethical practice: introducing an alternative to traditional anonymisation techniques in image-based research
Abstract
This article introduces the use of AI-replicas as an alternative to traditional anonymisation methods in image-based qualitative research. It emphasises the ethical and practical dilemmas posed by current anon-ymisation methods, such as distortion or loss of emotional and contextual information in images, and proposes the use of AI-replicas to preserve the integrity and authenticity of visual data while ensuring participant anonymity. The article outlines the technological foundations of generative artificial intelligence (AI) and the practical application of Stable Diffusion to generate AI-replicas for anonymisation and fiction-alisation purposes. Furthermore, it discusses the potential biases present in generative AI to suggest ways to mitigate these biases through careful prompt engineering and participatory approaches. The introduced approach aims to enhance ethical practices in visual research by providing a method that ensures participant anonymity without compromising the data's qualitative richness and interpretative validity. Ethical and research-practical challenges in using images for qualitative research Over the last three decades, there has been a growing interest in the visual dimensions of social life, and research methods that utilise visual data of some kind have become legitimate and
ResearchGate has not been able to resolve any citations for this publication.
Diffusion models have emerged as a powerful new family of deep generative models with record-breaking performance in many applications, including image synthesis, video generation, and molecule design. In this survey, we provide an overview of the rapidly expanding body of work on diffusion models, categorizing the research into three key areas: efficient sampling, improved likelihood estimation, and handling data with special structures. We also discuss the potential for combining diffusion models with other generative models for enhanced results. We further review the wide-ranging applications of diffusion models in fields spanning from computer vision, natural language processing, temporal data modeling, to interdisciplinary applications in other scientific disciplines. This survey aims to provide a contextualized, in-depth look at the state of diffusion models, identifying the key areas of focus and pointing to potential areas for further exploration. Github: https://github.com/YangLing0818/Diffusion-Models-Papers-Survey-Taxonomy.
The storage of facial images in medical records poses privacy risks due to the sensitive nature of the personal biometric information that can be extracted from such images. To minimize these risks, we developed a new technology, called the digital mask (DM), which is based on three-dimensional reconstruction and deep-learning algorithms to irreversibly erase identifiable features, while retaining disease-relevant features needed for diagnosis. In a prospective clinical study to evaluate the technology for diagnosis of ocular conditions, we found very high diagnostic consistency between the use of original and reconstructed facial videos (κ ≥ 0.845 for strabismus, ptosis and nystagmus, and κ = 0.801 for thyroid-associated orbitopathy) and comparable diagnostic accuracy (P ≥ 0.131 for all ocular conditions tested) was observed. Identity removal validation using multiple-choice questions showed that compared to image cropping, the DM could much more effectively remove identity attributes from facial images. We further confirmed the ability of the DM to evade recognition systems using artificial intelligence-powered re-identification algorithms. Moreover, use of the DM increased the willingness of patients with ocular conditions to provide their facial images as health information during medical treatment. These results indicate the potential of the DM algorithm to protect the privacy of patients’ facial images in an era of rapid adoption of digital health technologies.
Biometric facial recognition is an artificial intelligence technology involving the automated comparison of facial features, used by law enforcement to identify unknown suspects from photographs and closed circuit television. Its capability is expanding rapidly in association with artificial intelligence and has great potential to solve crime. However, it also carries significant privacy and other ethical implications that require law and regulation. This article examines the rise of biometric facial recognition, current applications and legal developments, and conducts an ethical analysis of the issues that arise. Ethical principles are applied to mediate the potential conflicts in relation to this information technology that arise between security, on the one hand, and individual privacy and autonomy, and democratic accountability, on the other. These can be used to support appropriate law and regulation for the technology as it continues to develop.
Learning how to conduct qualitative research may seem daunting for those new to the task, especially given the paradigm ’s emphasis on complexity and emergent design. Although there are guidelines in the literature, each project is unique and ultimately the individual researcher must determine how best to proceed . Reflexivity is thus considered essential, potentially facilitating understanding of both the phenomenon under study and the research process itself . Drawing upon the contents of a reflective journal, the author provides an inside view of a first project, making connections between theory and practice. This personal narrative highlights the value of reflexivity both during and after a study, and may help to demystify the research process for those new to the field.
This chapter draws from a commercial netnography we conducted to explore Chinese cosmetics consumers' changing notions of female beauty. Specifically, we were tasked with informing the new product development efforts of ArtCosmetics, an Italian B2B contract manufacturer operating in the global cosmetics market. We use the project to discuss how we used a team of researchers to collect, interpret, translate, and understand data about the central role of cultural codes of beauty in cosmetics tastes and routines. In response to the complexity involved in the project, we improvised a netnographic research design. As a result, our netnography with ArtCosmetics was a methodological and intellectual journey that challenged us in many ways: transnational and effective sampling, appropriately bricolaged research design, collecting and comprehending the nuance of foreign language and foreign culture data from new platforms, resolving heterogenous data, managing vast cultural complexity, and translating sophisticated ethnographic finding into pragmatic consumer, brand, and new product development insights. Today, Chinese notions of beauty and identity are firmly rooted in historical ethnic and national identities but are also fluidly global. Adapting netnography to this fluid transnational context allowed us to grasp a flow of beautyscapes, infoscapes, brandscapes, selfscapes, and usagescapes as shifting elements in a primordial process that oscillates between East and West, traditional and futuristic, and symbolic and functional.
Field notes are widely recommended in qualitative research as a means of documenting needed contextual information. With growing use of data sharing, secondary analysis, and metasynthesis, field notes ensure rich context persists beyond the original research team. However, while widely regarded as essential, there is not a guide to field note collection within the literature to guide researchers. Using the qualitative literature and previous research experience, we provide a concise guide to collection, incorporation, and dissemination of field notes. We provide a description of field note content for contextualization of an entire study as well as individual interviews and focus groups. In addition, we provide two “sketch note” guides, one for study context and one for individual interviews or focus groups for use in the field. Our guides are congruent with many qualitative and mixed methodologies and ensure contextual information is collected, stored, and disseminated as an essential component of ethical, rigorous qualitative research.
This article analyzes how pregnant women perform their pregnancies on Instagram. We ask whether they rely on and reproduce pre-existing discourses aimed at morally regulating pregnancy, or reject them and construct their own alternatives. Pregnancy today is highly visible, intensely surveilled, marketed as a consumer identity, and feverishly stalked in its celebrity manifestations. This propagates narrow visions of what a “normal” pregnancy or “normal” pregnant woman should be like. We argue that pregnant women on Instagram do pregnancy via three overlapping and complimentary discourses of “learn it,” “buy it,” and “work it.” Together these form the current authoritative knowledge of pregnancy we call “intensive pregnancy” as performed on Instagram. Concurrently, this article highlights how the combined discursive power of hashtags, images, and captions may influence and enforce discursive hegemonies.
Mobile dating applications such as Tinder have exploded in popularity in recent years. On Tinder, impression management begins with a motivation to download the app, the choice of one’s profile photos and an assessment of the expectations of potential Tinder matches. These processes occur in a technologically mediated environment of reduced cues and increased control, local proximity and a reduced filtering process. My focus in this paper is this first stage of impression management, which consists of both impression motivation and impression construction. Specifically, what are the pre-match impression management practices of Tinder users? I present the results of interviews with Tinder users in the Netherlands. Participants were recruited via a Tinder profile that advertised the study using the University emblem and a brief description. Interview questions focused on user understandings of self-presentation practices and profile construction. The interviews also examined how users evaluated their potential matches. Results show users’ motivations for using Tinder range from entertainment to ego-boost to relationship seeking, and these motivations sometimes change over time. Profile photos are selected in an attempt to present an ideal yet authentic self, and chosen as an illustration of not only one’s desirability but also of other indicators such as education level. Tinder users ‘swipe’ not only in search of people they like, but also for clues as to how to present themselves in order to attract others like them. This research offers insight into user experiences and perceptions within the still under-researched area of inquiry.
The use of images and other visual data in qualitative research projects poses new ethical challenges, particularly in the context of participatory research projects that engage research participants in conducting fieldwork. Little is known about how research participants deal with the ethical challenges involved in conducting fieldwork, or whether they succeed in making balanced ethical judgments in collecting images of identifiable people and places. This study aims to increase our understanding of these ethical challenges. From an inductive analysis of interview data generated from nine participants recently involved in a photovoice research project we conclude that raising awareness about ethical aspects of conducting visual research increases research participants’ sensitivity toward ethical issues related to privacy, anonymity, and confidentiality of research subjects. However, personal reasons (e.g., cultural, emotional) and cautions about potential ethical dilemmas also prompt avoidance behavior. While ethics sessions may empower participants by equipping them with the knowledge of research ethics, ethics sessions may also have an unintentional impact on research.
Photography serves as an important tool for researchers to learn about the contextualized lives of individuals. This article explores the process of integrating photo elicitation interviews (PEI) into research involving children and families. Much literature is dedicated to the general debate surrounding the ethics of visual methods in research, with little attention directed at the actual process of gaining study approval and publishing one's findings. There are two main critiques that researchers must face in order to conduct and disseminate studies involving visual images—ethics committees and peer reviewers. In this article, I identify and discuss some of the challenges that emerged across gaining protocol approval from an ethics committee in the United States. Ethical concerns and restrictions related to the use of photography can delay data collection and create barriers to research designs. Similarly, I describe the process of responding to reviewers' concerns as part of the publication process. Peer reviewers' lack of familiarity with the use of photography as a research tool may lead to misunderstandings and inappropriate requests for manuscript changes. While many concerns are sound, the range of benefits stemming from the use of visual data help to justify the time and energy required to defend this type of research. Implications are discussed for researchers using visual methods in their work.
URN: http://nbn-resolving.de/urn:nbn:de:0114-fqs1503274
This article addresses potential effects on reflexivity of researcher’s social position (e.g. gender, age, race, immigration status, sexual orientation), personal experiences, and political and professional beliefs. Because reflexivity is a major strategy for quality control in qualitative research, understanding how it may be impacted by the characteristics and experiences of the researcher is of paramount importance. Benefits and challenges to reflexivity under three types of researcher’s position are discussed and illustrated by means of case examples: (1) reflexivity when researcher shares the experience of study participants, (2) reflexivity when researcher moves from the position of an outsider to the position of an insider in the course of the study, and (3) reflexivity when researcher has no personal familiarity or experience with what is being studied. Strategies are offered for harvesting the benefits of researcher’s familiarity with the subject and for curbing its potentially negative effects. Directions for future research are suggested.
Anonymity is often taken-for-granted as an ethical necessity. Discussions around reuse of qualitative data have added further weight to its importance for protecting research participants from unknown future use of data. Yet, anonymisation also compromises the possibilities of future reuse. This paper argues that default anonymisation forecloses discussion of the ethics of naming; it calls for a reorientation of debate, away from an assumption of the universal/ist ethical good of anonymity, towards a politics and ethics of the question of naming. Rather than a slide from an ethic of avoiding harm to a paternalistic notion of protection, I suggest that questions of naming in research, and avoidance of harm, could productively be approached through a feminist ethics of care.
The considerable progress in qualitative methods is directly connected with developments in the field of text-interpretation. On the basis of a thorough reconstruction of their formal structures texts are treated as autonomous domains of self-referential systems. Such a methodological status has been denied to pictures in empirical research in the field of social sciences up until now. The documentary method, based on Karl MANNHEIM's Sociology of Knowledge, opens up methodical access to pictures. Methodologies from art history (PANOFSKY, IMDAHL) can thus become relevant for empirical research in social sciences. Connections to semiotics (BARTHES, ECO) and philosophy (FOUCAULT) are worked out in their consequences for qualitative methods. Thus verbal contextual and pre-knowledge can be controlled methodically in the documentary interpretation of pictures. The reconstruction of formal structure of pictures becomes of central importance in analysis. All of this will be demonstrated by examples from research practice.
URN: urn:nbn:de:0114-fqs0803267
Text-to-image generation (TTI) refers to the usage of models that could process text input and generate high fidelity images based on text descriptions. Text-to-image generation using neural networks could be traced back to the emergence of Generative Adversial Network (GAN), followed by the autoregressive Transformer. Diffusion models are one prominent type of generative model used for the generation of images through the systematic introduction of noises with repeating steps. As an effect of the impressive results of diffusion models on image synthesis, it has been cemented as the major image decoder used by text-to-image models and brought text-to-image generation to the forefront of machine-learning (ML) research. In the era of large models, scaling up model size and the integration with large language models have further improved the performance of TTI models, resulting the generation result nearly indistinguishable from real-world images, revolutionizing the way we retrieval images. Our explorative study has incentivised us to think that there are further ways of scaling text-to-image models with the combination of innovative model architectures and prediction enhancement techniques. We have divided the work of this survey into five main sections wherein we detail the frameworks of major literature in order to delve into the different types of text-to-image generation methods. Following this we provide a detailed comparison and critique of these methods and offer possible pathways of improvement for future work. In the future work, we argue that TTI development could yield impressive productivity improvements for creation, particularly in the context of the AIGC era, and could be extended to more complex tasks such as video generation and 3D generation.
We present ControlNet, a neural network architecture to add spatial conditioning controls to large, pretrained textto-image diffusion models. ControlNet locks the productionready large diffusion models, and reuses their deep and robust encoding layers pretrained with billions of images as a strong backbone to learn a diverse set of conditional controls. The neural architecture is connected with “zero convolutions” (zero-initialized convolution layers) that progressively grow the parameters from zero and ensure that no harmful noise could affect the finetuning. We test various conditioning controls, e.g., edges, depth, segmentation, human pose, etc., with Stable Diffusion, using single or multiple conditions, with or without prompts. We show that the training of ControlNets is robust with small (<50k) and large (>1m) datasets. Extensive results show that ControlNet may facilitate wider applications to control image diffusion models.
As participatory methods place an emphasis on ‘collaborative research practices … to produce knowledge in collaboration,’ a transition to virtual platforms created challenges in effectively facilitating these collaborations. Because participatory approaches to digital methods are still growing, there is limited information on best practices for working remotely with ‘vulnerable’ individuals from refugee backgrounds using participatory methods. Given the unexpected requirement to transition to remote research practices as a result of the pandemic, in this article I reflect on my experiences navigating my university’s institutional ethics process, transitioning my participatory visual research project with Turkish and Kurdish asylum seekers to a virtual platform, and the challenges that I have encountered in facilitating digital focus groups.
The use of participatory visual methods and integration of cellphone technology is expanding in global public health research. Cellphilm method capitalises on these trends by inviting participants to use mobile devices to create short videos about health topics. This paper presents the quilted cellphilm method, which supports the participation of stigmatised populations to engage in research. We present the method with reference to the Celling Sex project, which worked with young women who have transactional sex experience. Four key steps in our unique model are discussed: (a) individual cellphilm-making; (b) participatory analysis; (c) creating a composite video; (d) publicly screening the work. We consider how working individually with participants in the cellphilm-making process built trust. We unpack how offering participants opportunities to engage in either group or one-on-one activities promoted participation in collaborative analysis. We outline how creating a composite video of the cellphilms and organising screenings facilitated knowledge translation and exchange. Overall, the quilted cellphilm method created a supportive community for vulnerable participants to generate products that challenged social stigma. Increased reliance on mobile media, especially during the COVID-19 pandemic, makes the quilted cellphilm method an opportune, exciting and accessible approach for participatory public health research.
Described by the philosopher A.J. Ayer as a work of ‘great originality and power’, this book revolutionized contemporary thinking on science and knowledge. Ideas such as the now legendary doctrine of ‘falsificationism’ electrified the scientific community, influencing even working scientists, as well as post-war philosophy. This astonishing work ranks alongside The Open Society and Its Enemies as one of Popper’s most enduring books and contains insights and arguments that demand to be read to this day. © 1959, 1968, 1972, 1980 Karl Popper and 1999, 2002 The Estate of Karl Popper. All rights reserved.
In this article, I take on the problem of the face in images and visual research on children. This is a problem that is engendered through the visual representations of children and the act of deploying the visualizing techniques associated with visual methods (pictures, video, etc.). It nevertheless is a problem, I argue, that has been couched singularly within a question of ethics in child studies, criminology, and sociology, among other disciplines. Here, I utilize the work of Gilles Deleuze and Felix Guattari to challenge the unquestioned ethical commitment to the pixilation of children’s faces in publications. To trouble and reconceptualize the problem of visual representations of children, I assert that this problem is intimately connected to the cultural politics of childhood. For illustrative purposes, I analyze how children are represented in Today’s Child advertisements and Roman Vishniac’s Children of a Vanished World. This article concludes with a broader discussion of the (child’s) face, digital images, and (micro)politics.
Openings of interaction are critical for maintaining social relationships. We explore the practices of managing openings in a situation where parents are separated from their children and they only encounter each other in a video-mediated environment. As a result of this, these openings have a particular emotional significance. Based on detailed analysis of 56 video calls openings between Chinese migrant parents and their left-behind children, we show that grandparents, who are the caregivers of these children, “orchestrate” these openings. We demonstrate, firstly, that grandparents help to establish a parent-child “talking head” configuration upon connection. We show, secondly, that after the establishment of mutual talking head, both grandparents and parents expect the child to treat the production of this VMC-mediated configuration as sequentially implicative, that is to speak and greet first, and to greet largely. Our findings suggest that such orchestrated openings are an important component in the enactment of family bonding and intimacy across distances.
This article seeks to trouble the conflation of photo-anonymisation with ethical research practice, as is often promoted by institutional ethics review. Contributing to discussions of ‘visual ethics’ and the application of photographic methods in visual research, it poses the question: what is the effect of the anonymisation of photographs on the integrity of visual research and generation of new knowledge? Discussion unfolds via the exploration of school-based research around young people and sexualities. Examples of photo-anonymisation are examined from a project concerning the sexual cultures of schooling, employing a combination of photo-diary and photo-elicitation methods. The way in which anonymisation techniques meant to protect participants from harm, such as photo-cropping, facial blurring and pixelation, may be counterproductive to this aim, is revealed. Applying such techniques can undermine the agency of participants to convey their ideas and experiences as they intended. It is also argued that anonymisation processes can make a mockery of the integrity of visual methods by casting their meaning and content into obscurity. Such practices can consequently impoverish the generation of new meanings and the advancement of a disciplinary area – which, in the current study, is research on sexualities and schooling. This discussion seeks to encourage visual researchers to reflect upon anonymisation, inciting them to consider the implications of this practice for the ethical treatment of participants and the generation of new meanings in their chosen area.
A brief review is presented of existing forms of transcription of talk that incorporate visual, spatial and temporal elements. The most common forms use text and a line-by-line-based system, and conversation analytic transcripts have been successful in making a number of other features of talk visible. The desire of geographers to draw upon video recordings and time-lapse photography has lead to time-series images being used to bring those visual materials into documents. The graphic transcript is proposed as an alternative form of transcription that hybridises the qualities and evidentiary criteria of the transcript with the representational conventions of the comic strip. The comic strip itself has recently been undergoing a period of experimentation and hybridisation with other forms. The graphic transcript brings together familiar comic strip features such as panels, guttering, speech bubbles and captions with the transcript's criteria of providing an evidentiary record of earlier events that is available for re-inspection and re-interpretation by other analysts.
In this article I discuss my growing worry over the use (and nonuse) of photographs of young children in social research. Taking a view of child protection as paramount, I situate the discussion in the context of ethical governance of research, young children as research participants, changing views of childhoods, and the juxtaposition of the Arts with/in Arts-Based Educational Research. The article sets out a worry over the increasing tendency to pixilate—and hence distort—images of young children for inclusion in research reports or presentations. This (largely unchallenged) practice is slipping into research conventions whilst at the same time research governance and research involving children profilerate. The article argues that the pixilation of photographs of children, may represent a further “crisis of representation” and is an example of the “Othering” of young children in research.
This article focuses on innovative methods for protecting privacy in research of Internet-mediated social contexts. Traditional methods for protecting privacy by hiding or anonymizing data no longer suffice in situations where social researchers need to design studies, manage data, and build research reports in increasingly public, archivable, searchable, and traceable spaces. In such research environments, there are few means of adequately disguising details about the venue and the persons being studied. One practical method of data representation in contexts in which privacy protection is unstable is fabrication, involving creative, bricolage-style transfiguration of original data into composite accounts or representational interactions. This article traces some of the historical trends that have restricted such creative ethical solutions; emphasizes a researcher's obligation to protect research participants' privacy in mediated research contexts; and offers an introductory framework for reconsidering how to make case-based decisions to better protect the interests of participants in situations where vulnerability or potential harm is not easily determined.
The International Visual Sociology Association (IVSA) Code of Research Ethics and Guidelines introduces the principles and ethical standards that underlie professional responsibilities involved in doing visual research. It consists of an introduction, a preamble and a statement of general principles, followed by ten statements of ethical standards. The Code is prefaced by an introduction by Diana Papademas, who led the development and wrote the Code on behalf of the IVSA.
The ethical regulation of social research in the UK has been steadily increasing over the last decade or so and comprises a form of audit to which all researchers in Higher Education are subject. Concerns have been raised by social researchers using visual methods that such ethical scrutiny and regulation will place severe limitations on visual research developments and practice. This paper draws on a qualitative study of social researchers using visual methods in the UK. The study explored their views, the challenges they face and the practices they adopt in relation to processes of ethical review. Researchers reflected on the variety of strategies they adopted for managing the ethical approval process in relation to visual research. For some this meant explicitly \'making the case\' for undertaking visual research, notwithstanding the ethical challenges, while for others it involved \'normalising\' visual methods in ways which delimited the possible ethical dilemmas of visual approaches. Researchers only rarely identified significant barriers to conducting visual research from ethical approval processes, though skilful negotiation and actively managing the system was often required. Nevertheless, the climate of increasing ethical regulation is identified as having a potential detrimental effect on visual research practice and development, in some instances leading to subtle but significant self-censorship in the dissemination of findings.
In this companion volume John van Maanen's Tales of the Field, three scholars reveal how the ethnographer turns direct experience and observation into written fieldnotes upon which an ethnography is based. Drawing on years of teaching and field research experience, the authors develop a series of guidelines, suggestions, and practical advice about how to write useful fieldnotes in a variety of settings, both cultural and institutional. Using actual unfinished, "working" notes as examples, they illustrate options for composing, reviewing, and working fieldnotes into finished texts. They discuss different organizational and descriptive strategies, including evocation of sensory detail, synthesis of complete scenes, the value of partial versus omniscient perspectives, and of first person versus third person accounts. Of particular interest is the author's discussion of notetaking as a mindset. They show how transforming direct observations into vivid descriptions results not simply from good memory but more crucially from learning to envision scenes as written. A good ethnographer, they demonstrate, must learn to remember dialogue and movement like an actor, to see colors and shapes like a painter, and to sense moods and rhythms like a poet. The authors also emphasize the ethnographer's core interest in presenting the perceptions and meanings which the people studied attach to their own actions. They demonstrate the subtle ways that writers can make the voices of people heard in the texts they produce. Finally, they analyze the "processing" of fieldnotesâthe practice of coding notes to identify themes and methods for selecting and weaving together fieldnote excerpts to write a polished ethnography. This book, however, is more than a "how-to" manual. The authors examine writing fieldnotes as an interactive and interpretive process in which the researcher's own commitments and relationships with those in the field inevitably shape the character and content of those fieldnotes. They explore the conscious and unconscious writing choices that produce fieldnote accounts. And they show how the character and content of these fieldnotes inevitably influence the arguments and analyses the ethnographer can make in the final ethnographic tale. This book shows that note-taking is a craft that can be taught. Along with Tales of the Field and George Marcus and Michael Fisher's Anthropology as Cultural Criticism, Writing Ethnographic Fieldnotes is an essential tool for students and social scientists alike.
Photovoice is a process by which people can identify, represent, and enhance their community through a specific photographic technique. As a practice based in the production of knowledge, photovoice has three main goals: (1) to enable people to record and reflect their community's strengths and concerns, (2) to promote critical dialogue and knowledge about important issues through large and small group discussion of photographs, and (3) to reach policymakers. Applying photovoice to public health promotion, the authors describe the methodology and analyze its value for participatory needs assessment. They discuss the development of the photovoice concept, advantages and disadvantages, key elements, participatory analysis, materials and resources, and implications for practice.
Background. Observation is used in research in two ways – structured and unstructured. Which of these methods to choose depends on the research question but will be defined predominantly by the paradigm underlying each study. Positivistic research generally uses structured observation and interpretist/naturalistic paradigms use unstructured observation. This paper is concerned with unstructured observation.
Aim. To discuss the importance of unstructured observation as a research method and to examine critically the problems associated with both access and field notes.
Application to nursing. Although unstructured observation has been used in nursing research, it is less favoured than the interview. Furthermore, although some of the dilemmas concerned with observation such as the role of the researcher, covert research and informed consent have received attention in the nursing literature, other aspects remain relatively unexplored. In particular little has been written concerning access to the field and field notes. Both of these aspects have practical and theoretical considerations that require the attention of the researcher. This omission may partly explain nurses' apparent reluctance to use unstructured observation in their studies. This paper seeks to redress this situation.
Qualitative content analysis as described in published literature shows conflicting opinions and unsolved issues regarding meaning and use of concepts, procedures and interpretation. This paper provides an overview of important concepts (manifest and latent content, unit of analysis, meaning unit, condensation, abstraction, content area, code, category and theme) related to qualitative content analysis; illustrates the use of concepts related to the research procedure; and proposes measures to achieve trustworthiness (credibility, dependability and transferability) throughout the steps of the research procedure. Interpretation in qualitative content analysis is discussed in light of Watzlawick et al.'s [Pragmatics of Human Communication. A Study of Interactional Patterns, Pathologies and Paradoxes. W.W. Norton & Company, New York, London] theory of communication.
This paper is a description of inductive and deductive content analysis.
Content analysis is a method that may be used with either qualitative or quantitative data and in an inductive or deductive way. Qualitative content analysis is commonly used in nursing studies but little has been published on the analysis process and many research books generally only provide a short description of this method.
When using content analysis, the aim was to build a model to describe the phenomenon in a conceptual form. Both inductive and deductive analysis processes are represented as three main phases: preparation, organizing and reporting. The preparation phase is similar in both approaches. The concepts are derived from the data in inductive content analysis. Deductive content analysis is used when the structure of analysis is operationalized on the basis of previous knowledge.
Inductive content analysis is used in cases where there are no previous studies dealing with the phenomenon or when it is fragmented. A deductive approach is useful if the general aim was to test a previous theory in a different situation or to compare categories at different time periods.
- M Banks
Banks M (2001) Visual Methods in Social Research. London: Sage Publications.
Laion-5B: A new era of open large-scale multi-modal datasets
- R Beaumont
Beaumont R (2022) Laion-5B: A new era of open large-scale multi-modal datasets. Available at:
https://laion.ai/blog/laion-5b/ (accessed 14 January 2024).
DALLE3: Improving image generation with better captions
- J Betker
- T Brooks
- J Wang
Betker J, Brooks T, Wang J, et al. (2023) DALLE3: Improving image generation with better captions. arXiv. https://cdn.openai.com/papers/dall-e-3.pdf
Stable diffusion v1 model card
- Compvis
CompVis (2022) Stable diffusion v1 model card. Available at: https://github.com/CompVis/stablediffusion/blob/main/Stable_Diffusion_v1_Model_Card.md#training (accessed 14 January 2024).
Anonymizing people in images using generative adversarial networks
- K Douglas
Douglas K (2022) Anonymizing people in images using generative adversarial networks. MSc
Thesis, University of Amsterdam.