Introduction. This article focuses on the lack of a minimum overview of the way in which social problems have been analyzed in the Romanian media during the last years as reflected by domestic literature. More specifically, at first sight, one can notice that scientific studies focused on the way in which Romanian media exposed the developments recorded in the last decades, being rather concerned with issues related to the political life than with purely social elements.

On a general, theoretical level, the need for a systematic approach of the specialized literature in various fields was explained by L. Uman (2011), who noticed the existence of an increasingly high number of articles or volumes that treat subjects from various fields from a scientific perspective. She outlined that, based on this reality, at present, it is harder for a researcher specialized in human and social sciences to develop an overview of what is still to be explored, more specifically, what are the points and themes not covered by the specialized literature (Uman, 2011). In essence, according to Uman (2011), the narrative analysis is no longer enough to address the scientific literature, as it is focused exclusively on the descriptive aspects, with an emphasis on the systematic analysis or/and meta-analysis, which implies the existence of a data collection plan and clear criteria for their prioritization. Most frequently, the researcher says (Uman, 2011), the systematic analysis also includes the meta-analysis, which addresses from a statistical point of view the synthesis of data collected from numerous studies, in order to obtain a quantitative result that could describe the effect of the studies included in the analysis regarding the effect on the targeted population.

In view of these extremely general considerations, I was interested in identifying the main trends recorded in recent years in the scientific analysis of the way in which the Romanian media presented the internal social aspects. This is why I have carried out a systematic analysis of the method used by the researchers in order to study the media representation of disadvantaged social groups.

The article is structured in three main parts. The first section consists of a review of the international literature regarding what is commonly known as Framing Theory, which is the main conceptual framework used in our approach to media representations. In the second part, I included the research project carried out on the above-mentioned subject, while the last section was focused on the conclusions derived from the juxtaposition of the specialized literature with the results of the systematic analysis.

Methodology and methods. General theoretical framework. The Framing Theory of a media subject or theme is still a debated topic in academic analyses focusing on the relationship between media and the society in the Eastern Europe countries, including Romania. In recent years, a growing number of studies using this theoretical framework have been recorded in a number of interlinked disciplines and academic fields, such as: sociology, communication sciences, media studies, cultural anthropology, psychology of communication (D’Angelo, 2002; Shah, Domke & Wackman, 1996). Researches using the framework offered by the Framing theory can be found in cognitive, constructive and critical studies (D’Angelo, 2002), in sociology, economics, psychology, cognitive linguistics and communication studies (Scheufele & Tewksbury, 2006), or in political sciences and media studies (Hertog & McLeod, 2001). As Hertog and McLeod (2001) show, this diversity is “both a blessing and a curse”.

One of the most important general-theoretical distinctions is the one between frames in thought and frames in communication (Chong & Druckman, 2007a; Lim & Jones, 2010; Scheufele & Scheufele, 2009). The first type of frames refers to “a person's cognitive understanding of a given situation” (Chong & Druckman, 2007a). However, frames in communication could be described as “an idea of a central organization of a text or a narrative direction which gives meaning to a string of events” (Gamson & Modigliani, 1989). A main assumption of the framing theory is that there is a causal relationship between the two types of frames mentioned above, i.e. the frames in communication can affect the frames in thought (Scheufele, 1999). An example of this is the way in which certain frames in communication influence the particular way in which news are read (De Vreese & Claes, 2005).

Such varied perspectives have not only encouraged the manifestation of creativity (Hertog & McLeod, 2001), but also the existence of a paradigmatic diversity that has led to the spread of prospects for media “framing” (D’Angelo, 2002). However, the lack of a clear conceptualization and operationalization (Scheufele, 1999) has led to the use of researches derived from this theory as well as research approaches that are structurally different (Scheufele, 2000; Scheufele & Tewksbury, 2006). Furthermore, there are important differences regarding the essential points of this theory (Hertog & McLeod, 2001). For example, in contradiction to Entman (1991), D’Angelo (2002) argues that the call for a single framing paradigm expressed by Entman would be neither possible, nor desirable (D’Angelo, 2002). Furthermore, D’Angelo (2002) states that it was precisely the diversification of theoretical and methodological approaches that led to a more comprehensive understanding of the “framing” (Potter & Riddle, 2007).

From a methodological point of view, this paper used the systematic analysis of a sample of scientific articles focused on the analysis of the way in which disadvantaged Romanian social groups have been presented by the media. The literature distinguishes between two methods used in the analysis of literature relating to a problem: Meta-analysis and systematic analysis (Hunter, Schmidt & Jackson, 1982; Emmers-Sommer & Allen, 1999).

Meta-analysis is a form of a reviewing the existing scientific literature on a given subject. As a quantitative method, the meta-analysis allows, by definition, the statistical testing and the generalization of the results (Hunter, Schmidt & Jackson, 1982). By definition, systematic analysis is a “systematic quantitative technique used to establish relationships between variables” (Emmers-Sommer & Allen, 1999). Generally, it is considered that the systematic analysis helps to elucidate misconceptions in specialized literature related to a research topic, examine methodological arguments and provide a full assessment of theoretical views. However, as Li and Tang suggest, meta-analyses and systematic analysis can sometimes limit the examination of a subject that contains a variety of underlying themes (Li & Tang, 2012). However, as Cook and Leviton show, both meta-analysis and systematic analysis are superior to other methods of research used in the literature, including the narrative approach (Cook & Leviton, 1980).

In order to select a sample for the study, I started by using “Google Scholar” which was considered the main database for scientific articles. The timeframe used for the study was the period between the years 2000 and 2019. In order to identify the articles, I used several keywords paired with the term “media”, such as: “old people”; “vulnerable groups”; “poor people”; “disabilities”; “violence”; “social problems”; “unemployment”; “social movements”; “social crisis”; “children”. The research was carried out in both Romanian and English language. For each combination of terms, I took into consideration the articles displayed on the first fifteen pages of “Google Scholar”. The summaries of the articles were subsequently read, and, as a result, a new selection was made. In the third phase, all articles were read and those that were not focused on the analysis of the media presentation of at least one social group in Romania were removed. It resulted in a final sample of 70 articles published in the mentioned period.

The iterative way of selecting the studies is shown in the figure below.

The analysis grid for the articles included in the final sample consisted of twenty-nine items, out of which, ten were closed (pre-coded) and nineteen, open (See Table 1 below).

Data processing was performed using the SPSS program (version 11.5) and for the present analysis we used exclusively descriptive statistics (frequencies and contingency tables).

ResearchResultsandDiscussion. According to the data resulting from our analysis, the largest number of scientific articles on how the media covered disadvantaged groups in Romania was published in 2015 (20.41%), followed by 2018 (12.24%) and 2013 (10.20%). At the same time, except for 2019, the lowest number of articles on the topic of interest was published in 2003, 2008, 2009, 2010 and 2012 (2.04% of the total sample for each year).

Most studies were published in journals in the field of communication sciences (30.64%), followed by articles published in journals in the field of media studies (24.49%). In third place, but at a considerable distance, there were studies published in journals in the field of social sciences in general (8.16%) and ethnic studies (6.12%). Articles published in journals in the field of security studies, human geography, gender studies and social work were the least represented in the total sample included in the analysis: 2.04% each category.

Regarding the language in which the articles were written, our data indicate that an overwhelming percentage (95.92%) were studies published in English (both in foreign journals and those published in Romania), only 4, 08% being written and published in Romanian.

Only an extremely small percentage of the articles (7.5% of the total sample) did not clearly specify at least one research objective, the vast majority of studies (92.5%) having included it in both the abstract and the body text.

Regarding the explicit inclusion of research objectives, our results indicate an extremely high percentage (55% of the total sample) of studies that did not include them in their text. Only 45% of the articles focused on the analysis of the way in which the media presented the disadvantaged groups in Romania also included clearly formulated research objectives.

Most studies (73.47%) only analyzed the way in which the media presented the disadvantaged groups in Romania, while 20.41% presented these groups comparing Romania and other countries, and 6.12% of the articles analyzed the situation of these groups with direct reference only to the European Union (without making any reference to Romania).

Almost half of the total sample included in our analysis (43.59%) were studies that used the qualitative methodology exclusively, while only 20.51% of the articles used an altered form of qualitative methodology. What is surprising is the high percentage, almost a quarter of the total sample (25.64%), of articles that used a mixed methodology, combining qualitative and quantitative methods.

Regarding the type of sample used, our results indicate that there was a balance between the quantitative samples (which were present in 38.77% of the items included in the analysis) and the qualitative samples (present in 36.73% among the studies analyzed). An extremely small number of articles (2.04%) presented both quantitative samples and qualitative samples.

A second unexpected result was the non-inclusion in the body text of clearly formulated research hypotheses or presuppositions: a percentage of 92.50% of the analyzed articles did not have at least a single and clearly identifiable research hypothesis or presupposition. Only for an extremely small percentage (7.5%) of them we identified the existence of such elements.

However, a significant percentage (90% of the studies analyzed) included a presentation of the research methods used, which was missing only in the case of a significantly low percentage of sampled items (10%).

The most frequently used research methods were quantitative content analysis (24.49%), qualitative content analysis (18.37%), and speech analysis (12.24%). Methods such as pragmatics, visual analysis, secondary analysis of social documents or visualization of semantic networks were extremely scarce in the articles included in the analysis (the percentages for each of these methods were 2.04% for each).

Regarding the explicit presentation of the data collection tools, the results of our analysis indicate that more than half of the studies (65%) had no mention of these elements intrinsically related to the research process. Only 35% of the items included in the sample studied contained either a brief presentation of the data collection tools or indications related to their structure.

From the perspective of the presentation of the research results, our data indicated that the majority (60%) were presented as a general text and only 20% of the total sample dealt with studies that clearly and meticulously presented the final results. However, a relatively high percentage (17.5%) of the total sample of analyzed articles presented the results in a vague, unclear way and, in addition, for 2.5% of the total sample we found that we were dealing with articles in which the section dedicated to the presentation of results was completely missing.

Problems regarding the ethics of scientific research were mentioned only in the case of 10% of the total studies analyzed, as the vast majority (90%) of them did not make the slightest reference to such issues.

Most of the studies included in the analysis (89.74% of the total sample) also included a section dedicated to clearly formulated comments and final conclusions. It is, again, extremely unusual that 10.26% of the total studies analyzed did not include such a section.

From the perspective of the implications (theoretical and/or practical) that the studies included in our analysis may have, the results indicated that over half of them (62.5%) contained either a separate section dedicated to these aspects or included references to them in the body text. However, more than a third of the articles analyzed did not even include the implications that the research could have theoretically and/or practically.

Conclusions. According to the literature, the framework has been analyzed from the perspective of various scientific disciplines (D’Angelo, 2002; Reese, 2007; Van Gorp, 2006). Sociologists have used framing theory to examine, for example, how social movements frame a problem through the media, in order to get help from the public (Gerhards, 1995; Pan &. Kosicki, 2001; Snow & Benford, 1992; Snow, Rochford, Worden & Benford, 1986). They found that successful “frame sponsors” are based on the cultural codes that exist in a society at one time or another (Benford & Snow, 2000; Gamson & Modigliani, 1989; Pan & Kosicki, 2001; Reese, Gandy & Grant, 2001), linking the messages they convey to the values and beliefs of their target groups (Snow, Rochford, Worden & Benford, 1986). Others examined media as a forum for public deliberation, concluding that these frameworks were either “specific to each issue” (De Vreese, 2005; Neuman, Neuman, Just & Crigler, 1992; Price & Tewksbury, 1997; Semetko & Valkenburg, 2000), so they are used exclusively to define a single theme or problem, or they are “generic”, in which case they are used repeatedly to define the meaning of certain topics, themes or problems (Ihlen & Nitz, 2008).

Studies dedicated to the framing process have often been connected with those dedicated to the “agenda setting” analysis and the priming effect, all three being subsumed to the broader category of cognitive effects of the media (Scheufele, 2000; Scheufele & Tewksbury, 2006). Challenging the model of limited media effects, McCombs’ proposed theory argues, in essence, that through the way it makes daily selection of news, media influences the public agenda (McCombs & Shaw, 1972). McCombs later argued that media coverage can even be seen as the second dimension of agenda setting research (Cappella & Hall Jamieson, 1997; Maher, 2001).

All these theoretical aspects were mentioned (totally or partially) in the articles included in our systematic analysis. Unfortunately, as our results have indicated, in recent years (2000-2019), academic articles on this topic (scientific analysis of how the Romanian media presented disadvantaged groups) have recorded significant shortcomings in terms of content. The aspects that were underrepresented in these studies were the explicit presentation of the research objectives and some clearly formulated research hypotheses or presuppositions, as well as the problems related to the ethics of scientific research. At the same time, the loose presentation method (in more than half of the sample this was extremely general), the omission of the implications in theoretical and/or practical terms and even the 10.26% of the studies that did not contain a section dedicated to the conclusions are a second subset of deficient elements highlighted in our analysis. It is obvious that in the absence of these elements, the analyzed studies have become, in fact, simple validations in Romanian context of established theories and have not presented elements of novelty, further development, both conceptually and methodologically.

The existence of this status quo raises serious questions about the validity of the academic approach within the last almost two decades. At the same time, however, these results indicate the main directions that can be corrected in the future, in order for the approaches in our country to contribute to the progress of knowledge, in various segments of the procedures of socio-human disciplines.