Framing Studies Evolution in the Social Media Era. Digital Advancement and Reorientation of the Research Agenda

: Framing studies remain a powerful line of research in political communication. However, in recent years, coinciding with the emergence of social media, theoretical and operational advances have been detected, as well as a signiﬁcant reorientation of its research agenda. The interaction between media and platforms such as Twitter or Facebook has built a clearly hybrid communicative environment and profoundly transformed the organization of public debate. This is the case, especially, with processes such as the setting of the public agenda or the construction of interpretive frames. Based on a systematic review of the international reference literature (2011–2021), this article analyses the inﬂuence of social media on the evolution of framing studies. Moreover, speciﬁcally, the beginning of a new stage of digital development is contextualized, and a triple research impact is explored. The main contributions of the text are that it (1) identiﬁes advances in the theoretical and empirical organization of these studies; (2) explores its reorientation of content towards a greater balance between the analysis of media and political frames; and (3) reviews the recent experimental development of effects studies. Finally, the main challenges for future research in this ﬁeld are detailed.


Introduction
Setting the agenda and building interpretive frames to guide public debate continue to be two key journalistic functions in democratic societies.To this day, the media maintain their leading role when it comes to identifying the main social problems and offering simple interpretive schemes to citizens (Casero-Ripollés 2018).Consequently, they maintain a notable influence in shaping public attitudes about current affairs.
Despite its prior application to other disciplines, framing studies have had a regular presence in the communication field since the mid-1990s.Currently, they constitute a powerful line of research in both journalism and political communication (Entman and Usher 2018).The concept of "framing" has proven very useful to connect and explain globally the processes of production, distribution, and consumption of political information.Moreover, it has achieved relevant results, above all, in the study of two processes: the strategic orientation of the journalistic message and the social construction of public debate (D 'Angelo 2012).However, it also carries a general dispersion problem, some theoretical ambiguities, and methodological limitations.Furthermore, it has been characterized by a certain imbalance in its research agenda.While the study of the message has played a major role when dealing with political issues in the reference press, other issues have been left in the background, especially the sociological dimension of the concept, the analysis of the audience frame, and the experimental study of its effects (Muñiz 2020).
At present, the situation of framing as a research program is complex.On the one hand, intense research activity is maintained which, together with its classic conceptual and operational "fragmentation", has led to a certain saturation in the field (D' Angelo et al. 2019).On the other hand, the digital context has brought obvious advances.The conceptual debate offers more integrative definitions and a better operationalization of the frames (Brugman and Burgers 2018).Furthermore, an expansion of the research focuses on the study of the framing of parties and social movements in their strategic use of networks.Above all, a more comprehensive approach is consolidated when reviewing the frame building process (Scheufele and Iyengar 2017).
Undoubtedly, the arrival of social media and its consolidation as a political tool, about a decade ago, has had important consequences (Chadwick 2017).Specifically, platforms such as Twitter or Facebook have profoundly transformed the professional dynamics of political communication.To a large degree, the media have lost their monopoly on the management of current affairs and their ability to influence has been limited: political actors and citizens distribute their messages directly to mass audiences and achieve greater prominence in public debate.
These new roles have generated intense interactions between new and old media and in-depth changes in the rules of the game between journalism and politics (Tewksbury and Riles 2018).Among them, communicative hybridization has energized key aspects of the framing process, generating new professional relationships and influences, new types of content and communication styles, and opening a specific line of work that connects the emergence of fake news and misinformation strategies that alter the functioning of the communication system.In this context, the study of social media has become a dynamic factor in framing studies.This leading role, together with theoretical and operational advances, anticipates the opening of a new stage in its research development (Saperas and Carrasco-Campos 2015).
Different studies have reviewed the evolution of framing in political communication in recent years (among others, Borah 2011; López-Rabadán and Vicente-Mariño 2013; Ardévol-Abreu 2015; Knüpfer and Entman 2018;Tewksbury and Riles 2018;Muñiz 2020).However, the impact of social media on its research program has not been specifically addressed.This review article aims to offer an updated vision of the transformation of framing studies in the digital context from a double approach: on the one hand, to critically contextualize its main theoretical and operational advances and its pending challenges; and on the other, to explore the renewal of its research agenda, which includes new content, dynamics, and actors in the process of building political frameworks.
Based on a strategic review of the available literature, our study is organized around the following objectives regarding the evolution of framing studies:

•
O1. Identify its current strengths and limitations when explaining the process of the construction of the informative message and organization of the public debate; From these objectives, the article is organized into five sections.After the explanation of the methodology, the reasons for the research success of the framing theory are detailed and its main limits and contradictions are also updated (Section 3).Next, the current moment is contextualized within the general evolution of these studies (Section 4).Subsequently, several theoretical and operational consensuses under construction are identified, key in a more effective reorganization of the specialty (Section 5).Regarding the diversification of its research agenda, new lines of work and methodological approaches proposed in the last decade are explained (Section 6).Innovation in the study of effects and the increase in experimental activity are addressed in depth (Section 7).Finally, a set of conclusive ideas about the present and future challenges of this line of research is offered (Section 8).

Method
This work is based on a systematic review of the literature organized in three phases.In the first place, more than 180 scientific articles on framing and political communication published in the last decade (2011)(2012)(2013)(2014)(2015)(2016)(2017)(2018)(2019)(2020)(2021) were located in Google Scholar, while understanding this period as a new stage of development of the specialty in the context of social media.In a second phase, a final sample of 78 articles was selected applying criteria of quality and scientific relevance: only works published in journals indexed in the two academic reference databases, Web of Science (WoS) and Scopus (preferably in quartiles Q1 and Q2) were finally included; and, likewise, its impact on the specialty measured in the number of appointments was taken into account.All bibliographic consultations have been carried out individually between June and September 2021.
In a third phase, this strategic sample of articles was classified into two time periods: start of digital political communication studies, especially Twitter (2011Twitter ( -2015)); and research on new networks (2015)(2016)(2017)(2018)(2019)(2020)(2021).Finally, the texts have been analysed in depth, taking into account two thematic variables: first, distinguishing between theoretical work, state of the art, and applied studies; and later, according to its object of study, media framing, specific works on social media, comparative studies on media hybridization, and experimental studies on framing effects.

Success and Limitations of Framing Studies in Political Communication
Framing theory has experienced intense application in recent decades to become one of the most active and relevant research areas in communication (Matthes 2009;Borah 2011;Cacciatore et al. 2016).Since its origin in the field of sociology, and with a significant connection with disciplines such as psychology, behavioural economics, or political science (D 'Angelo 2002;Brugman and Burgers 2018), the study on the concept of "frame" has found the most favourable field for its development in the 21st century in political communication (Brugman et al. 2017).In particular, this approach can analyse the connections amongst the media process, the political debate, and the social definition of meanings (Matthes 2012;López-Rabadán and Vicente-Mariño 2013) while explaining the public opinion formation process.All these factors have positioned it as one of the predominant perspectives in the field (Muñiz 2020).
On the other hand, there is significant data on its level of activity and impact in the field (Weaver 2007).In 2004, several authors placed Framing Theory as a significant theory of the second half of the 20th century, occupying the first place in references from the period 2001-2004, well ahead of perspectives such as the agenda-setting, the theory of uses and rewards, or the theory of selective exposure (Bryant and Miron 2004).In the following decade, this high level of activity intensified (Borah 2011), reaching theoretical and operational improvements, high empirical applicability, and a constant presence in the journals with the greatest impact (Saperas and Carrasco-Campos 2015).
However, interest and research success of framing studies in recent decades has emerged with frequent criticisms of its conceptual and operational inconsistency (De Vreese 2005; Ballesteros-Herencia and Gómez-García 2020).In consequence, its intense and imprecise application has ended up generating saturation problems in the field (D 'Angelo et al. 2019).Next, the main arguments of interest and the limitations of this theory are specified, which, paradoxically, are closely related.

Reasons for a Research Success
There are at least five arguments to explain the success of framing studies in political communication.First, it deals in-depth with a significant communication process in democratic societies.This theory offers a very effective metaphor to explain how political information is produced and transmitted to citizens (Muñiz 2020).While the idea of agenda synthesizes thematic priorities ("the what"), the concept of "frame" has been defined, to a large extent, as the form of strategic presentation of events through informative or political messages ("the how").It does so through two mechanisms: emphasizing or excluding possible aspects of reality (De Vreese 2005) and developing certain discussion functions (Entman 1993).
On the other hand, its success is also explained by its global but well-segmented view on the communication process (Entman et al. 2009).The Framing Theory is defined by its ability to explain the mediation process of political reality, both in a panoramic way and in-depth in the parts that make it up (Matthes 2012;Hänggli and Kriesi 2012).In this sense, it connects three relevant fields: the strategic production of content, the characteristics of the message and, finally, its individual, social, or cultural impact (De Vreese and Lecheler 2012).
This theoretical breadth comes from its multidisciplinary origin.The original concept of "frame", understood as meaningful schemes that give meaning to the social world is located between the sociological and the psychological, landing in very different ways in communication studies (D 'Angelo 2018).The classic communicative definitions (Entman 1993;Scheufele 1999) locate the frame in four points of the communicative process: sender (mind of strategists), message (texts), receiver (mind of citizens), and finally, culture (shared mindsets).This way, the message construction process goes beyond the journalistic work itself while the frames can be four things at the same time: a psychological process, a professional organization process, a final product, and a political strategy tool (Entman et al. 2009).
Third, it stands out for its realism in its investigative application.Although defined as a multidisciplinary paradigm while potentially allowing a comprehensive analysis of the entire communication process (D'Angelo 2012; Matthes 2012), its actual development has focused on studying its most significant parts.Specifically, it is a certified key tool for the strategic analysis of the journalistic message (Borah 2011;López-Rabadán and Casero-Ripolles 2014).However, the experimental study of the media effects on individuals has been approached in a more specific way (Cacciatore et al. 2016;Brugman and Burgers 2018).
The methodological flexibility shown by framing studies is also significant when analysing all kinds of issues: from specific subjects such as electoral campaigns or the image of a leader (Sahly et al. 2019;Louie and Viladrich 2021) to long-term problems such as climate change, social protest, the refugee crisis, or cases of sexual harassment (Lück et al. 2018;Ahmed et al. 2019;Starkey et al. 2019), from very different techniques and designs.In general, framing studies are characterised by proposing a great diversity of models of content analysis of the informative or political message (Matthes and Kohring 2008), while surveys and experimental designs have had less application up to a few years (Brugman and Burgers 2018;Banks et al. 2021).
Finally, it is relevant to highlight its ability to adapt to the digital context.As a consequence of its conceptual extent and its operational flexibility, framing studies have almost directly incorporated the political use of social networks into their research project (Chadwick et al. 2018;D'Angelo et al. 2019).The strategic management of Twitter, the interaction between the media and the public (Manor and Crilley 2018;Hopke and Hestres 2018), and finally, the attitudinal effects it generates (Wicke and Bolognesi 2020), have become relevant objects of study and a dynamic factor in the field.

Limitations for Its Scientific Study
However, the popularity of the concept of framing has landed with a notable inconsistency in its application (De Vreese 2005).Its multidisciplinary origin, though useful in development, has generated notable problems of theoretical imprecision (D 'Angelo 2002;Miceviciute 2013).In perspective, this theory has been a blessing and a curse for communicative research (Borah 2011).To a large extent, it has become a powerful explanatory metaphor that hides a complicated conceptual and operational definition.
Paradoxically, despite its investigative use for decades, uncertainty remains about what precisely constitutes a frame and how the framing process is articulated (Lecheler and De Vreese 2016).This lack of theoretical consensus has been denounced as a problematic issue (Matthes 2009;De Vreese and Lecheler 2012).Definitions of frames are usually scarce and very general in explaining "what they are".For example, some speak of "principles of selection, emphasis and presentation" (Gitlin 1980) without actually defining them.Moreover, the more specific definitions focus on "what they do".This is the case of the four framing functions proposed by Entman (1993).Consequently, it has been impossible for a long time to reach definitions and a conceptual consensus shared by the entire research community (Carragee and Roefs 2004;Vliegenthart and Van Zoonen 2011).
This situation is explained by the persistence of two contradictory trends (Saperas and Carrasco-Campos 2015).First, this lax conceptual framework favours high applicability to all kinds of approaches, topics, and research fields: from the quantitative and qualitative, the empirical and interpretive, the psychological and sociological, and, finally, the academic and professional.Likewise, this ambiguity has created a very dispersed line of research, lacking consensus on its central concepts and operational mechanisms.After almost three decades, the "fractured paradigm" situation denounced by Robert Entman (1993) remains, since framing studies continue to lack a unified theory regarding how frames are constructed, represented, and how they are influential (Entman et al. 2009).To a large extent, framing studies have been victims of their applicability and research success (D 'Angelo and Kuypers 2010).
This has resulted in a lack of theoretical consensus generating an enormous methodological dispersion when identifying and measuring media frames, parties, and among citizens (Matthes and Kohring 2008;Ardévol-Abreu 2015).Based on imprecise methodological criteria, the operational diversity has been enormous.Almost every author defines and operationalizes frames in general, different, or even divergent ways (Matthes 2009).This situation has generated significant reliability and validity problems and has prevented the scientific standardization of the field (Van-Gorp 2010; López-Rabadán and Vicente-Mariño 2013).
Another derived methodological problem is the commitment to research based on typologies of specific frameworks (Matthes 2009).This type of work ends up being very descriptive and prevents the formulation of hypotheses and comparative studies that advance the theoretical basis of the field.The predominance of research based on specific frames has been "grouping" unconnected empirical data, instead of "fitting" the results within a shared theoretical base (D 'Angelo 2012).
Finally, it is relevant to indicate the problems generated by the excessive segmentation of framing studies (Miceviciute 2013;Brugman and Burgers 2018).On the one hand, research on framing has not considered the sociological basis on news production raised by Tuchman (1978) or Gitlin (1980).This disconnection between the journalistic and social frame has reduced the depth and critical look of the field.Thus, the idea of power is absent in most current approaches, and it is not proposed to investigate the responsibility of certain settings or their social or political consequences (Vliegenthart and Van Zoonen 2011;Ardévol-Abreu 2015).On the other hand, the separation between the journalistic frame and the psychological scheme has also reduced profundity in defining its effects at different levels.
Concerning the field fragmentation, different authors have criticized the excessive prominence of the analysis of the journalistic message (Hänggli and Kriesi 2012;Valera-Ordaz 2016;Muñiz 2020).Beyond its results, this "mediacentric" approach to framing studies has consolidated a reductionist and partially problematic pattern.On the one hand, it obviates the sociological dimension of the process and limits its critical capacity.On the other hand, it has slowed down further development of the analysis of the frame of other types of actors (parties, social movements, or audiences) and the experimental study of its effects (Borah 2011;Powell et al. 2019).

Evolution in Three Stages
Framing studies have their origin in two concepts from other disciplines: Bateson's psychological frames (Bateson 1972) and the definition of the frame as a basic sociocognitive mechanism established by Goffman (1974) from micro-sociology.
This interdisciplinary starting point has allowed for flexible application in different fields (López-Rabadán and Vicente-Mariño 2013).Among others, its pioneering use in studies of journalistic production as an organizational and interpretive framework of news stands out (Tuchman 1978;Gans 2004;Gitlin 1980), but also in psychology (Kahneman and Tversky 1984), or the study of social movements (Snow and Benford 1988).After this landing phase, the following decade is characterized by the appearance of specific works that combine theoretical reflections on media framing with the empirical analysis of news discourse (Gamson and Modigliani 1989;Iyengar 1991;Gamson 1992).
From the definition of Robert M. Entman, a second stage opens in the evolution of media framing studies.This work is key to defining two questions.On the one hand, he defines it as a "fractured paradigm" and places it together with other theories, such as priming, which review and reinterpret the effects of agenda-setting.On the other, he establishes the key points of the field (Entman 1993, p. 52):

•
He defines the framing process as a strategic action that develops four discursive functions (definition of the problem, attribution of causality, moral judgment and treatment recommendation);

•
And he places it at four points throughout the communication process (emitter, text, receiver, and culture).
Despite this attempt at coordination, this second stage is characterized by a growing empirical application, ambiguous in many cases, and an intense theoretical and operational debate against other perspectives.On the one hand, they link their development to the analysis of journalistic treatment ("informative how") as a step beyond the journalistic selection ("informative what") that the agenda-setting theory represents (Semetko and Valkenburg 2000).On the other hand, the lack of a solid theoretical base generates a notable imprecision and the fragmentation of the field in different domains.Only after the appearance of reference works (Scheufele 1999(Scheufele , 2000;;D'Angelo 2002) does the field begin to clarify its conceptual and operational basis.Above all, this underlines its independence and complementarity with the tradition of the agenda-setting (Reese et al. 2001;Carragee and Roefs 2004;De Vreese 2005).
From that moment onwards, the third stage of consolidation began (2007-), characterized by the recognition of its scientific autonomy, the enormous commented empirical applicability, and the opening of a methodological debate to increase research precision and quality (Matthes and Kohring 2008;Saperas and Carrasco-Campos 2015).The key moment in this stage was the publication of a monographic volume of the Journal of Communication that represents the definitive consolidation of the field (Scheufele and Tewksbury 2007).

A Fourth Stage of (Re)organization and Digital Advancement
In the middle of the last decade (2015-), a new stage emerged.It is distinguished by a broader theoretical and methodological coordination and, above all, the reorientation of the research agenda (Scheufele and Iyengar 2017).Among the new priorities, the following stand out: (1) the integration of social networks as an object of study; (2) the commitment to the study of effects; and (3) the adoption of a more global perspective when exploring digital dynamics, both in the selection of topics and the adoption of a more comparative and international perspective (Knüpfer and Entman 2018).
Coinciding with the generalization of the political and journalistic use of social media, a series of operational consensuses begin to consolidate key in the organization of the field (Matthes 2012;Entman and Usher 2018).In this new digital context, framing studies have been presented as an increasingly precise tool to analyse phenomena such as political (self) communication, digital activism, and the dynamics of hybridization of new and old media (Hamdy and Gomaa 2012; Qin 2015; Nee and Maio 2019), or the advance of the spectacularization and populist style in the political message (Araújo and Prior 2021).

Consensus under Construction: Current Keys in Framing Studies
The transversality of framing studies in parallel to the development of research on social media and journalism (Segado-Boj 2020) has generated an intense debate to try to organize its development in the digital context (Borah 2011; López-Rabadán and Vicente-Mariño 2013; Saperas and Carrasco-Campos 2015; Muñiz 2020).As a result, it is possible to identify a growing consensus on several key aspects that mark a certain research pattern in the specialty for the future: general definition, concept of frame, organization of studies, and analytical model.

Plural Definition as a Research Program
From the experience of the last decades, and assuming the lack of a single concept that provides global coherence (Entman 1993), framing studies are currently considered a plural research project assuming its limitations and attempt to take advantage of its theoretical and researcher diversity (Muñiz 2020).Given its multidisciplinary nature and its objective of global analysis, many experts consider that it is neither possible nor desirable that these studies constitute a single paradigm (D 'Angelo 2002'Angelo , 2012;;Matthes 2009Matthes , 2012) ) while defining framing theory as a research program that serves as a meeting point for:

•
Paradigms such as the cognitive, the constructionist, and the critical;

•
To quantitative and qualitative methods;

•
To diverse epistemological approaches, from the sociological to the psychological;

•
To empirical and interpretive approaches; • Innovation in the academic and professional field.

Application of a Dual Concept of Frame
From this integrative general vision, the frame concept is defined by the connection and complementarity of its two dimensions: the sociological and the psychological (Muñiz 2020).After several phases of a certain reductionism, current studies find it essential to delve into the dual condition of the concept and integrate it into a shared theoretical base (Vliegenthart and Van Zoonen 2011).Each frame is defined as a shared social convention, which at a professional level is outlined in rules (journalistic practices, for example), as well as at a discourse level, in the use of symbols, images, or keywords (strategic orientation of the journalistic message) (Tankard 2001).However, at the same time, the frame concept presents a powerful psychological dimension understood as a cognitive structure that determines the individual interpretation of the message and its possible attitudinal effects (Scheufele and Iyengar 2017).A realistic and current definition of the concept must always include both dimensions and combine them in the analysis of political and mediatic phenomena.

Organization on the Study of Three Processes
Despite the possibility of conducting global studies, those studies centred on analysing the parts in the context of framing research predominate (De Vreese 2012).Among the few cases of comprehensive research, several studies that measure the media treatment of European elections and connect it with effects on audience attitudes through surveys stand out (De Vreese 2003;Matthes 2012).
Within the enormous diversity of these studies, it is possible to identify three large specific fields of study (D 'Angelo 2002;Muñiz 2020): construction of frames, which reviews the strategic organization of the issuer (frame building); definition of frames, which studies their presence in messages (news or political framing); and effects of frames on the public and culture (framing effects).Next, several shared keys about their uneven research development are specified.
Undoubtedly, frame building is the field with the least empirical development.Likewise, it is key to delving into a more sociological and professional vision of the framing process.This field offers the possibility of going beyond mere discourse, delving into determining factors while analysing the power relations between media and political actors from a critical point of view.From this perspective, the media is a central actor in the framing process, but not the only one.Furthermore, the dynamics of framing are defined as a struggle for the meaning of political events between different actors who have unequal material and symbolic resources (Vliegenthart and Van Zoonen 2011;Hänggli 2012).
The main object of this field is the professional norms and dynamics that determine the proposal, promotion, and consumption of frames at the social level.Identifying two types of factors that influence the work of the media are amongst its advances (Hänggli 2012;Hänggli and Kriesi 2010).On the one hand, the factors internal to the organization are those norms that simplify routines and help standardize the development of the journalistic product.This type of professional organization frame can occur at an individual or collective level, even arriving to develop stable strategic approaches (López-Rabadán 2010).Its most appropriate technique is conducting in-depth interviews.On the other, certain factors external to the environment are decisive in the creation of media frames; mainly, the influence capacity of political actors, businessmen, and pressure groups.Direct identification is complicated, and its indirect study is based on a content analysis that allows measuring correlations between the framing of press releases and the final content in the media.
News framing addresses the journalistic message as the central concept of the theory.Furthermore, its intense research activity has determined the evolution of the field since the 1990s (Matthes 2009;Saperas and Carrasco-Campos 2015).However, the arrival of social networks has broadened its focus of analysis to other fields such as the political, activist, and citizen message (Snow et al. 2018).At an operational level, content analysis stands out as the main technique (Matthes 2009).Moreover, it is mainly applied as an inductive approach trying to systematize the identification of news frames from very different typologies (De Vreese 2005; Brugman and Burgers 2018).Among them, the use of more specific "thematic frames" predominates, compared to the "generic" ones such as conflict, attribution of responsibility, episodic, or human interest (Lecheler and De Vreese 2016;Brugman et al. 2017).Within the diversity of existing textual analysis models (Matthes and Kohring 2008), the most frequent is the "holistic manual".Mixed in nature, it is based on the qualitative detection of frames in key content and its subsequent quantitative measurement in larger samples.The "computer-aided" or "deductive" models have been presented as interesting add-ons, still under development.While other qualitatively based models such as the "hermeneutic" or the "linguistic" have had less application.
Framing effects addresses the final result of the theory and is specified in an object of enormous social relevance: the influence of the informative and political message on citizens (Scheufele 1999;D'Angelo 2018).The study of framing effects implies the empirical connection of two phenomena: identifying frames in the messages from the news (independent variable), and the subsequent location of some type of sociocognitive or attitudinal effect on the audience (dependent variable).The study of effects has been present since the sociological start of the field (Goffman 1974) and its development in communication (Scheufele 2000).In political communication, they have become a frequent approach (De Vreese and Lecheler 2012) based on the premise that subtle changes in the way of presenting the message (the informative how) can generate relevant changes in the attitudes of the citizens (Lakoff 2004;Weaver 2007;Vicente-Mariño and López-Rabadán 2009).
Regarding its development, the framing effect is understood as a process of cognitive channelling in three steps (Scheufele 2004;Matthes 2012):

•
First, it goes from the cognitive impact of the media and/or political message to the audience's thinking; • Second, it is established as an individual mental scheme that will be used for several purposes: to think about a topic while building opinions about it; • Ultimately, to trigger behavioural changes.
Therefore, frames become mechanisms that generate effects at the attitudinal and conative level, determining what people think about key aspects of the political agenda (De Vreese 2005; D'Angelo 2018).In the last decade, theoretical developments and the improvement of experimental designs as the main technique have resulted in a significant rise in the empirical study of framing effects (Muñiz 2020).

Approach to a Comprehensive Analytical Model
Both the complexity of the framing process and the interpretative nature of the frames in the context of the informative or political message have complicated the scientific study of frames.Frames are latent structures, and their identification is obtained from indirect indications in the message (arguments or keywords) or expert testimonies that express their perceptions in this regard.Moreover, they are not exclusive.In the same content, several frames of different intensities can coexist.These factors have complicated the application of standardized and rigorous techniques (Van-Gorp 2010).However, the evolution of the field has also generated a series of conventions that guide, to a large extent, the analysis of frames in political communication (Matthes and Kohring 2008;López-Rabadán and Vicente-Mariño 2013;Tewksbury and Riles 2018).
Regarding the analysis of the message, five main conventions are pointed out:

•
To place the research focus on two processes: thematic selection and discursive organization, as key professional actions in the informative setting; • Seek the methodological balance, betting on designs that combine quantitative content analysis of the message, together with the complementary use of qualitative techniques, such as observation or interviews, that confirm or qualify the detected patterns;

•
Bet on longitudinal analysis samples that allow identifying stable discursive patterns in the medium term; • Whenever possible, opt for an inductive approach that builds analysis categories, validate them from previous farms and, finally, measure them in the final sample.It is a more complex approach but offers more interesting results than the deductive approach.Frame mapping techniques that automate the study of the presence or absence of keywords are relevant inductive examples (Matthes and Kohring 2008);

•
Depending on the objective, to take advantage of the entire tradition of existing generic and specific typologies (Semetko and Valkenburg 2000;De Vreese 2005) to improve the proposal of their analysis models and try to be integrated into a debate of broader results.
Regarding the analysis of effects, two main conventions are indicated (Muñiz 2020):

•
To reorganize the experimental studies based on the use of more realistic informative materials.Instead of using artificial samples as a stimulus, it is recommended to obtain the samples beforehand by identifying the news frames through an analysis of real content (Scheufele 2004);

•
To increase the external validity of the experiments, a more realistic exposure to the materials is proposed (Lecheler and De Vreese 2016).In particular, to repeat the exposure of the same frames several times or to carry out the simultaneous exposure of several contradictory frames is recommended.

Triple Impact of Social Media on Framing Research
The advent of social media has transformed the research map of framing studies in two directions.On the one hand, it has propelled the renew of its objectives, improving its investigative techniques (D 'Angelo 2018;Entman and Usher 2018).On the other hand, it has reoriented and digitized its research agenda towards a more balanced distribution of content between the study of media, political, and citizen settings (Scheufele and Iyengar 2017).A third relevant consequence involves the improvement and the intense empirical development of the studies of the effects addressed in the final section of the text.

Methodological Advances
In the first place, there are improvements in the operationalization of the frames.More precise keys are gradually offered to identify frames in the texts.Furthermore, organization and argumentation mechanisms are deepened, such as, for example, the metaphor, the hyperbole, and the irony (Burgers et al. 2016).
On the other hand, innovative designs for the study of social media are identified; particularly, comparative models that connect the massive analysis of digital content (for example, publication of news on Twitter) with the conduct of surveys on contemporary phenomena, such as dynamics of increasing selective exposure (Aruguete and Calvo 2018), or the predisposition towards misinformation (Chadwick et al. 2018).
Improvements are also detected when applying the frame mapping technique (Matthes and Kohring 2008) to the study of digital objects.This is the case of the detection of political frames on Facebook (Ballesteros-Herencia and Gómez-García 2020), or the review of mixed content and visual-text about the refugee crisis on Instagram (Radojevic et al. 2020).
Finally, it is relevant to point out the use of big data techniques in analysing massive samples of political content on networks.Two examples of this are, on the one hand, the identification of frames on the refugee crisis of 2015 from a massive analysis of 7.5 million tweets collected through hashtags (Siapera et al. 2018) or from the interactions in networks between media and citizens (Pöyhtäri et al. 2021), and, on the other hand, the media frame of demonstrations and social protests during the presidency of Trump (Ophir et al. 2021).

Reorientation of the Research Agenda
In this fourth stage of evolution, framing studies maintain intense activity while presenting a more complete and balanced research agenda.Hence, this activity has been classified into three categories according to their main study object: analysis of the media, the political frame, and the hybridization between new and traditional media.

Innovative Trends on Media Frames
To a large extent, the traditional analysis model (coverage of a current issue in the reference press) maintains its predominance, though innovative trends in this field of research are also detected.
In the first place, key concepts and phenomena in current political communication are introduced as an object of study.Three relevant examples would be the presence of populism in the media during Bolsonaro's ascent to power in Brazil (Araújo and Prior 2021), or the personalization of politics during Trump's term (Ophir et al. 2021), or the xenophobic attack on the gypsy community by Minister Matteo Salvini in Italy (Cervi and Tejedor 2020), and, from another perspective, new dynamics such as the irruption of comedians to the TV prime time (Ödmark 2021).
A second trend is the analysis of increasingly global issues.These are medium-term social problems and not current affairs.Among these issues is climate change, which is allowing the development of comparative studies between countries (Lück et al. 2018) or long-term ones over time (Stecula and Merkley 2019).Other equally relevant issues are, first, the construction of the Euro crisis in five EU countries (Joris et al. 2018); second, the definition of sexual harassment uncovered in the #MeToo campaign in four different national contexts (Starkey et al. 2019); third, the degree of informative independence in the coverage of the Covid-19 pandemic (Milutinović 2021).
In this new stage, there is a growing interest in studying the visual frame.Given the generalization of digital journalism and networks, the study of the journalistic text is insufficient since the roles of photo and video gain weight when reviewing media frames.In this sense, those works that address the definition of negative visual frames about refugees or migrants (Amores et al. 2020) and the framing of war conflicts such as Ukraine on television in Russia and Germany stand out (Lichtenstein et al. 2019).

Development of Social Media as Object of Study
In this new stage, the analysis of the political management of networks becomes an autonomous and relevant object of study in framing studies.After Twitter, social media have expanded their presence in the research agenda, incorporating Facebook and Instagram while organising the study of these platforms on a series of significant trends.
The first is that the political framing analysis in networks reproduces a very similar approach to that of the media, namely, the selection of relevant cases and the search for consolidated strategies in the accounts of a leader or party.A very recurring example is the case of President Trump's social networks, highlighting here an innovative work that connects him with the construction of a frame of white supremacy (Louie and Viladrich 2021), organized around a triad of functions ("divide, divert and conquer").
Likewise, there are new works that analyse the Twitter strategy of certain international institutions.One example is the Israeli Ministry of Foreign Affairs during the 2014 Gaza War (Manor and Crilley 2018) in an extension of the theory of digital diplomacy framing during war conflicts.Another example is the competitive use of networks between parties, from differentiating themselves on Facebook in the 2019 Spanish electoral campaign (Ballesteros-Herencia and Gómez-García 2020) to the attacks on Twitter against the candidate Hillary Clinton through negative memes (Nee and Maio 2019).
The second trend is newer and is related to expanding the studies towards social movement strategies or citizens' debates in social media (Snow et al. 2018).There are significant examples of this.First, the work on the framing of climate change by the environmental movement, in particular the comparison of content on Facebook produced by 289 NGOs in 18 countries (Vu et al. 2021); second, the digital debate on highly relevant issues, such as the appearance of anti-immigration political rhetoric on Facebook (Nortio et al. 2021); and third, the conceptualization of the Covid pandemic as a war conflict on Twitter (Wicke and Bolognesi 2020).However, unquestionably, the issue that has attracted the most attention in recent years has been the construction of citizen frameworks around the Syrian refugee crisis of 2015, both the emergence of negative frameworks with a populist base on Twitter (Siapera et al. 2018) and the dissemination of humanitarian-based visual frames on Instagram (Radojevic et al. 2020).

Framing Analysis on Media Hybridization
The interaction between social media and traditional media has become, perhaps, the most relevant research area of this new stage of framing studies.The construction of frames within the hybrid media system, and the study of its consequences, has generated intense research activity in recent years and offers innovative aspects in its techniques and results (Tewksbury and Riles 2018).Regarding content, it is possible to identify two main lines of work: on the one hand, comparative studies of coverage; on the other, the analysis of the consequences of sharing content.
The most active and relevant comparison has been between media coverage and its replication on social media.Since the "Egyptian spring" of 2011, we have found pioneering works that explore the differences between traditional press framing and Twitter.The networks are a place of greater freedom and interaction in times of political crisis (Hamdy and Gomaa 2012).The different characterization of controversial characters such as Edward Snowden is also relevant since he was portrayed as a hero in the networks and a traitor in the North American media (Qin 2015).
Today, this first line continues offering interesting results on the digital treatment of global issues.Sometimes, these results are very different from those of the reference media.Climate change (Hopke and Hestres 2018) or the "refugee crisis" of 2015 (Pöyhtäri et al. 2021) are clear examples of this.In both cases, there are powerful visual frames and similar polarization strategies.Another global issue addressed in recent years is the differences between the new media and the press in treating political protests.Recent research shows that factors such as the type of protest (peaceful or violent), the place (own country or abroad), and the type of media are relevant factors that connect or not with the "protest paradigm" that demonizes protesters and marginalizes their causes (Harlow et al. 2020).In any case, citizen debates on Twitter are always perceived as a space of greater freedom and internationalization than the traditional media (Ahmed et al. 2019).
A secondary line of work is the one that explores the consequences of sharing news on networks, understood as an individual framing mechanism of the news reality (Aruguete and Calvo 2018).As network users select or discard content, they highlight facets of events or issues to promote a particular interpretation, especially on key issues such as social protests and their different treatment at a domestic or international level (Kilgo et al. 2018).Within the hybrid media context, a step further is to try to connect, through surveys, the common practice of sharing sensational news on networks, with a greater predisposition to disinformation and democratically dysfunctional behaviours (Chadwick et al. 2018).

Development of Experimentality in Framing Effects Studies
In this fourth stage, there are advances and increasing activity in the study of effects and the application of experimental models (Muñiz 2020).Parallel to the appearance of several review-works in this field (Cacciatore et al. 2016;Lecheler and De Vreese 2019;D'Angelo 2018), the publication of experimental studies has increased, reaching 300 (Brugman and Burgers 2018).Next, the main advances in the theoretical debate on effects are synthesized and a review of the most relevant research areas is offered.

Conceptual and Operational Classification
At the conceptual level, the proposal by Cacciatore et al. (2016) stands out to reorganize the theoretical basis of effects studies based on a series of principles that are somewhat ground-breaking with the previous tradition.

•
On the one hand, they link the effect of framing to its sociological definition while selecting a set of facts or arguments about a fact ("the what") and differentiate it from their psychological notion.Finally, there is a presentation of two equivalent ideas about a theme ("the how"); • Therefore, they define the framing effect as an applicability process.There is a connection with previous mental schemes invoked by the media message.This helps to process the information in a certain way (Scheufele 2004).
This definition takes away the idea of "accessibility" as exposure to certain content, which characterizes the agenda-setting theory and the definition of framing by Entman (1993).
At the operational level, several trends are detected.On the one hand, analysis protocol is established with practical instructions on the design and application of experimental studies (Lecheler and De Vreese 2019).On the other hand, there are objects of study typical of the current hybrid context, such as the effects generated in the interaction between new media and journalism, the growing influence of media frames that combine visual and verbal content, or the comparative and international-scale design approach (D 'Angelo 2018).Given the current challenge of increasing the realism and external validity of the experiments, a current of "experimental realism" is gaining ground.This line proposes moving beyond the laboratory to work with informative materials.Furthermore, it involves more realistic content consumption measurements, thus allowing a better understanding of the framing effect (Muñiz 2020).
Finally, time becomes crucial to explain in detail the framing effect while deepening its study.Thus, recent works identify two speeds of information processing: fast, automatic, and emotional on the one hand; slow, controlled, and rational, on the other.Furthermore, these works consider time a primary variable in the establishment of effects (Powell et al. 2019).Moreover, a growing number of experimental studies are investigating the duration of the effects of news framing.Most experts suggest that the effects persist beyond the initial exposure while possibly influencing subsequent decision-making.However, this debate is still open as other factors intervene, such as exposure to competitive or emotional frameworks (Lecheler and De Vreese 2016).

Increased Experimental Activity
In this phase, there is regular and innovative empirical activity in the context of research on framing effects (Lecheler and De Vreese 2019).Although not very standardized, a large number of experiments are carried out that connect the selective exposure to frames with the appearance of politically-based effects (Wu 2018).
Among the effects, one of the most outstanding is the increase in political engagement of citizens, especially in the context of the electoral campaign as a decisive moment in current political communication.Based on "experimental realism" designs, a recent work links thematic or strategic settings of the media with opinion measurements and changes in the level of citizen engagement in networks during the 2018 Mexican campaign (Muñiz and Echeverría 2020).On electoral engagement, there is a relevant study that connects the frameworks proposed by two candidates (Trump and Clinton) on two different platforms (Twitter and Facebook) during the 2016 presidential elections, with different levels of audience participation on each platform.The content analysed goes from "likes" or "retweets" to more elaborate comments (Sahly et al. 2019).
Another analysed phenomenon from this view is the rise of political polarization.Different studies have addressed it using different approaches, from experiments that connect exposure to polarizing tweets about candidates and parties to surveys that evaluate significant changes.In this sense, two events stand out: first, the case of a triple experiment on generic frames and negative messages in the North American presidential elections stands out (Banks et al. 2021).Second, an experiment based on surveys on the consumption of social networks and the increase in partisan radicalization in Brazil during the first months of the COVID-19 pandemic (Calvo and Ventura 2021).From a similar perspective, recent studies detect effects in the debate on gun control in the US.While emotional content exacerbates prejudices, news from a neutral frame generates reflection and a possible change of opinion among different audiences (Wu 2018).
The third group of relevant experiments connects the consumption of certain digital frames with the increase in social rejection of minorities, especially immigrants and refugees.In this sense, the work of Lecheler et al. (2015) stands out.In this study, the social influence of certain positive (emancipation and multiculturalism) or negative (assimilation or victimization) frameworks on immigration is measured.After the exposure to framing and conducting surveys, the conclusion is that positive emotions operate as mediators of more powerful framing effects.
Regarding the influence of the visual frame, there are exploratory results of two studies that stand out.While an experiment shows that informational videos on the European refugee crisis do not produce more powerful opinion shifts than news texts (Powell et al. 2018), another study detects interesting trends in the effect of photographs on immigrants in the US media.Through different experiments, Parrott et al. (2019) conclude that human interest frames increase positive emotions, enhancing positive attitudes about the group.Furthermore, political frameworks increase negative emotions and ultimately lead to negative attitudes.
Other relevant areas of experimentation, but less constant so far, have been those that affect changes in the evaluation of candidates and the intention to vote (Von Sikorski and Knoll 2019) or those that measure the impact on citizens' attitudes of disinformation strategies.This is the case of conspiracy theories about the origins of COVID-19 (Bolsen et al. 2020).Finally, another 100% hybrid line of research is the one that reviews the effects derived from sharing news on social media.In this sense, important experiments indicate a rise of racial identity issues from a selective exchange of content on Twitter (Bigman et al. 2019).Another experiment shows how morality frameworks increase the exchange of news on Twitter and Facebook while those that involve conflict reduce the exchange (Valenzuela et al. 2017).

Conclusions
According to the research objectives, it is possible to synthesize some conclusive ideas about the current moment of framing studies and pose various challenges regarding their development in the medium term.On the one hand, its intense research activity in the last decade returns a reasonably positive diagnosis regarding its evolution.In this sense, significant progress has been observed in some of its "classic" challenges, such as the need for greater theoretical clarity, operational precision, and a global study of the process (Borah 2011;López-Rabadán and Vicente-Mariño 2013;Saperas and Carrasco-Campos 2015).In recent years, the academy has chosen not to exaggerate its theoretical and operational weaknesses and to try to improve specific aspects of the study of frames in the hybrid context (D' Angelo et al. 2019).
Throughout the text, advances such as the construction of certain theoretical consensuses and a better definition of key concepts have been highlighted.For example, the link between the idea of framing and the deep meaning of the message, and for its analysis to propose a complementary vision that takes into account both the frames based on selection and "emphasis" (common in journalism) and those of interpretive "equivalence" (less frequent).Likewise, it is committed to staying the course as a research program that integrates and takes advantage of different approaches instead of seeking uniformity.After several decades of evolution, there is consensus in accepting the complexity of the concept of framing and understanding operational diversity as a value that allows for the deepening of its study.
On the other hand, the diversification of its research agenda is clear.Although political journalism remains the main area, social media have generated a significant reorientation of its objectives.First, a global vision of the framing process has been imposed that better contextualizes it within the journalistic and political dynamics.Second, although the media and parties are the main sources of the process, the research focus already includes the strategies of social movements and the interpretation of the audience.Third, the hybridization between media and political settings is posed as a growing area of research.Likewise, a certain methodological standardization is detected, both in the operational identification of frames in the message and in the study of effects.Greater analytical precision is observed on the processes of selection and discursive organization and a more realistic approach to effects analysis from experimental designs.

Challenges and Future Approaches
Despite the advances indicated, the framing studies still present an important margin of theoretical and methodological development (Muñiz 2020) and, above all, a better adaptation to the dominant hybrid context in current political communication (Qin 2015;Entman and Usher 2018).In this sense, there are several future challenges that serve as continuity to the objectives of this work.

•
Comprehensive approach to the process.Although the concept of frame allows it, there are still few works that study the framing process as a whole, from the production of the message to the effects.An in-depth study of media and political frames needs to take into account both the professional context where they are built and the society where they are interpreted and disseminated; • Balanced methodological designs.As it occurs latently, the study of the frame requires sophisticated techniques for its detection and measurement.Given the limitations offered by content analysis, it is recommended to advance in inductive approaches that combine a quantitative base with a qualitative refinement.Although more complex, this approach is more rigorous and realistic.Beyond the use of big data techniques, it is recommended to always provide examples that clarify and contextualize the categories and their interpretation (D' Angelo et al. 2019).On the other hand, in the current context, the study of the frame demands increasingly complete comparative and longitudinal perspectives.Regarding the study of effects, "experimental realism" opens an interesting way for methodological improvement; • Expansion of the research agenda towards the study of the visual frame.Photographs and videos are key content of digital journalism and social media.The very idea of framing contains a powerful explanatory metaphor and in the current context, its investigative interest has increased.Although relevant work has already been observed, it is a field of enormous potential.However, its study requires specific methodological improvements that take into account the framing power and uniqueness of the image and its combination with the text; • Analysis of new factors in the hybrid context.Among the consequences of the hybrid media system is the appearance of new factors that influence the framing process (Entman and Usher 2018;Knüpfer and Entman 2018).Standing out among them is the role of platforms that are almost monopolistic, such as Google or Facebook, which can determine access to political information and its interpretation.Furthermore, the power of algorithms and strategic technologies (such as digital analytics) that can be used to monitor online debates, refine communications, and quantify opinion and political engagement.For its part, it also highlights the influence exerted by new digital actors that subvert the traditional dynamics of political communication.Among them stand out "ideological media" that are committed to orienting their message towards the polarization of the public sphere.Furthermore, "rogue actors", such as hackers and disinformation platforms, who are disrupting the classic news ecosystem.Integrating all these factors is key to understanding the competition between frames and its political consequences in the future; • Go beyond the message, delve into the consequences.Starting from a more determined sociological approach, the critical study of the political effects of framing can become an interesting field of research in current political communication and, in this way, review in detail how networks and media hybridization are altering the process of the construction of public debate (Entman and Usher 2018).In this way, it can delve into relevant phenomena such as the fragmentation of the media system, the increase in transnational information flows, or the growing control of information by economic and political elites and analyze others in expansion such as the ability of new actors to manipulate media messages and distort the functioning of the democratic system without forgetting the role of traditional journalism in full transformation, with its limitations and its new roles.
Funding: This research is part of the research project UJI-B2020-14, funded by Universitat Jaume I within Research Promotion Plan 2020.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.