Creating a dilemma for public organizations The effect of NPM reforms on the media coverage of PSOs
Demi Herma Hoeve 11798173
Master’s Thesis 26-06-2019
Graduate School of Communication
Master’s programme Communication Science dhr. dr. Piet Verhoeven
2 Abstract
This research conducts a content analysis to compare the use of legitimacy and reputation frames in articles about hospitals in four Dutch newspapers (N = 335) before and after the privatization of hospitals in 2006. The results show that although the total amount of legitimacy and reputation frames did not increase after the privatization of hospitals, the focus on reputation did increase. The cognitive legitimacy of hospitals was largely ignored, while reputation frames aimed at comparing hospitals, stating differentiation and judging reputation were most prominent. The findings suggest that while governmental reforms do impact how the media cover public organizations, the privatization of Public Sector Organizations (PSOs) does not lead to a similar treatments of PSOs and private organizations. This research thereby sheds light on the effects of New Public Management (NPM) reforms on the media coverage of PSOs and on the legitimacy-reputation dilemma.
Introduction
In the western world, managing legitimacy and reputation has become an important issue for Public Sector Organizations (PSOs). In the past, PSOs were only slightly concerned with maintaining their legitimacy and reputation. Rather than deriving it from the media, PSOs were granted satisfactory levels of legitimacy through their connection with the government (Blomgren, Hedmo, & Waks, 2016). Reputation was not one of the main worries of these organizations, as they did not engage in an economic competition for customers. PSOs were financed by the state and thus did not need to immerse in races for being the best or most popular organization (van Essen, 2009). This changed when New Public Management (NPM) gained foothold. NPM is a management theory aimed at reforming governmental organizations in order to make them more efficient and more resemblant of private sector organizations (Groot & Budding, 2008). These reforms, often in the form of privatization or autonomization, made PSOs responsible for their own legitimacy and reputation. No longer
3 could they rely on the government for their legitimacy and reputation. This meant PSOs had to face a new dilemma, formerly unknown to them: the legitimacy-reputation dilemma (Blomgren, et al., 2016).
The legitimacy-reputation dilemma occurs when organizations both need to establish legitimacy, and a good reputation at the same time (Deephouse, 1999). Legitimacy and reputation both are evaluations of the organization’s actions in the eyes of stakeholders and the public (King & Whetten, 2008). Yet, the ways to achieve a good evaluation on both aspects seem contradictory; In order to gain legitimacy, organizations have to focus on conforming to (social) regulations and standards and be on par with similar organizations (Czarniawska & Wolff, 1998; Deephouse, 1999; van Ruler & Verčič, 2009). In order to achieve higher levels of reputation, organizations need to have a unique identity with differentiating attributes (Blomgren, et al., 2016; Deephouse, 1999). This creates a dilemma, whereby strengthening the organization’s legitimacy decreases the organization’s reputation and vice versa (Blomgren, et al., 2016; Deephouse, 1999; Deephouse & Carter, 2005; King & Whetten, 2008; Wæraas & Byrkjeflot, 2012).
One of the main sources from which stakeholders and the public draw in order to evaluate an organization’s actions, is news media. How an organization is covered in the media may thus influence the image of an organization in the minds of its evaluators. One way in which this can be done is by the usage of frames (de Vreese, 2005). By using frames, news media can make certain issues or attributes of an organization more salient than others (Cawley, 2012; de Vreese, 2005). As being legitimate and having a good reputation is crucial for the survival of organizations, the use of legitimacy and reputation frames by the media can have large consequences for organizations (Blomgren, et al., 2016).
4 In the Netherlands, NPM ideals eventually resulted in the privatisation of hospitals in 2006 (van Essen, 2009). Hospitals are exemplary for the NPM reforms around the world, as in most countries, healthcare was one of the focal sectors of NPM (Santiago, Carvalho, & Sousa, 2015; van Essen, 2009). Governments were faced with increasing public costs of healthcare and growing waiting lists. Restructuring the sector according to NPM principles seemed to offer a solution to these issues (van Essen, 2009). Hospitals are important organizations to every citizen and are therefore often scrutinized in the media. If the news media have moved along with the shift from hospitals as public organizations to hospitals as private organizations, this should be visible in the frames they use. Issues and attributes that relate to the reputation of hospitals should have become more pertinent than before the privatisation. This research therefore conducts a content analysis on the usage of framing in print news media, to answer the question: To what extent did the use of reputation and legitimacy frames in
newspaper articles about hospitals change after the privatisation of Dutch hospitals?
To date, there has been research on organizations in the media (Einwiller, Carroll, & Korn, 2010; Wonneberger & Jacobs, 2017) and the effect of NPM reforms on the functioning of the public sector (Groot & Budding, 2008; Klijn, 2012; van Essen, 2009), but these two subjects have not been combined. Analysing the differences in media coverage before and after the privatization of hospitals increases understanding of the effects of governmental reforms on the media coverage of PSOs. Furthermore, the outcomes of this research provide communication practitioners with insights into how their organization is scrutinized and therefore help in setting up communication strategies.
Theoretical framework New Public Management
5 Around the 1980s, new ideas about how PSOs should be managed emerged. Under the name New Public Management (NPM) governments started to reform public administration, state institutions and organizations financed with public money (Groot & Budding, 2008). The main focus of NPM reforms was to abandon classic bureaucratic models and move towards systems that corresponded to the private sector. By adopting commercial practices, governments attempted to increase the effectiveness and efficiency of PSOs (Klijn, 2012; Williams, Rayner, & Allinson, 2011). Depending on the political and cultural context, the way NPM reforms were executed varied greatly between countries (van Essen, 2009). However, in most of the cases three main measures were (partly) used. First of all, the “privatization and contracting out of governmental services, or parts of governmental bodies” (Klijn, 2012, p. 203). Secondly, the use of (semi-)market mechanisms in order to stimulate competition. Thirdly, the use of performance indicators similar to those in the private sector to define and control the goals to be achieved by the PSO (Klijn, 2012).
The NPM reforms are considered to be a radical turning point in public sector management (van Essen, 2009). The break with traditional bureaucratic models had a great impact on PSOs. Privatization, autonomization and the striving for more effectiveness and efficiency forced PSOs to evolve into more commercial organizations (Klijn, 2012). In order to be fully able to follow business-inspired models PSOs needed to become “‘complete organisations’” (Blomgren, et al., 2016, p. 178). This required the establishment of a structured hierarchy, market-driven management, goal-setting and measurement, responsibility for performances and most of all; a clear organizational identity (Blomgren, et al., 2016; Williams, et al., 2011).
PSOs which were autonomized or privatized during NPM reforms no longer derive their identity from the government. Just like private sector organizations, they have become responsible for the creation of their own identity. Consequently, PSOs
6 suddenly have to be concerned about how their actions are being seen by stakeholders, the competition and the government. They need to uphold a good reputation, while staying legitimate (Blomgren, et al., 2016). This burdens PSOs with a problem formerly only known in the private sector: the legitimacy-reputation dilemma.
Legitimacy & Reputation
“A firm should be different and a firm should be the same, respectively”. Deephouse (1999, p. 150), was one of the first to state that organizations are in a quandary when it comes to their legitimacy and reputation. To establish legitimacy, organizations benefit from being similar to the competition. Yet, when it comes to reputation, organizations benefit from positioning themselves far from the competition, through differentiation. In markets where both similarity and uniqueness are required, this leads to an ambivalent situation called the “legitimacy-reputation dilemma” (Blomgren, et al., 2016, p. 177).
Legitimacy.
Organizations need to be legitimate to obtain resources and to increase their performance (Deephouse, 1999). Legitimacy can be defined as a judgment about the actions of the organization and the level to which they are considered desirable, appropriate, sensible and morally justifiable (van Ruler & Verčič, 2009). There are three components on the basis of which the social acceptance of an organization can be evaluated: regulative, normative and cognitive legitimacy (Deephouse & Carter, 2005; Ruef & Scott, 1998). The regulative component entails compliance with official laws and regulations as set by formal institutions, mostly governments. The normative component requires congruence between an organization’s behaviour and the norms and values of the society in which it exists. This also includes professional rules and standards which are not imposed by the government, but to which organizations and their employees subscribe. Cognitive legitimacy is about the raison
7 can function in a morally acceptable way (van Ruler & Verčič, 2009). If an organization has a high level of cognitive legitimacy, it can become taken for granted and unquestioned (Deephouse, 1999; Deephouse & Carter, 2005; Deephouse & Suchman, 2008; Ruef & Scott, 1998).
Legitimacy equals the conformation with “rules, laws, norms, values, and cognitive frameworks in a larger social system” (Deephouse & Suchman, 2008, p. 54). Each category of organizations has its own set of requirements for legitimacy (King & Whetten, 2008). All organizations in the category will have to fulfil these requirements. This results in strategic similarity: competitors within a certain market become similar due to the compliance to the same norms and regulations (Blomgren, et al., 2016; Deephouse, 1999). This process is also called isomorphism (DiMaggio & Powell, 1983).
Reputation.
Where legitimacy is granted by testing an organization’s actions against norms and social expectations, reputation signifies the standing of an organization compared to its peers (King & Whetten, 2008). Stakeholders’ judgement of an organization’s reputation is based on performance indicators (Deephouse & Carter, 2005). Reputation is a zero-sum game, whereby similar organizations get ranked on the basis of what makes them special and unique (Deephouse & Suchman, 2008; King & Whetten, 2008). The most important tool to attain a favourable reputation is differentiation. The rivalry between organizations brings about the need to distinguish oneself from similar organizations, in order to reach a preferential status (Deephouse & Carter, 2005; Deephouse & Suchman, 2008). Reputation, in this view, is an economic instrument: increased reputation indices perceptions of better performance in the eyes of stakeholders. The perception of better performance creates a preference for the organization over other organizations, resulting in increased popularity (Deephouse & Suchman, 2008).
8 Legitimacy and reputation are thus conflicting: To demonstrate their legitimacy, organizations need to conform; to increase reputation and decrease competition, organizations need to deviate (Deephouse & Carter, 2005). Focusing primarily on either one of the two will be considered objectionable by stakeholders and will endanger an organization’s survival. Deephouse (1999) therefore states that organizations should try to maintain a strategic
balance by being “as different as legitimately possible” (p. 148), in order to receive benefits
from being legitimate and having a good reputation.
Traditionally, PSOs had not been concerned with maintaining and improving legitimacy and reputation. Their legitimacy was derived from the fact that they were governmental institutions. Being set up and provided by the government, their right to exist was taken for granted (Blomgren, et al., 2016). As long as PSOs met government regulations, they were legitimate. As PSOs were financed by the state, there was no rivalry between organizations over winning customers. PSOs thus did not try to differentiate themselves with the aim of improving their reputation (van Essen, 2009). When NPM reforms entered into force, this changed. The strategic balance for PSOs got closer to that of commercial for-profit organizations. Organizations no longer obtained their legitimacy solely from the government, but now also needed to be legitimate in the eyes of critical consumers (Klijn, 2012). Furthermore, the importance of having a favourable reputation increased, since organizations were now responsible for their own financing. This resulted in the need to differentiate themselves and outperform the competition (van Essen, 2009).
Legitimacy & Reputation in Media Framing
The media are an important player in conferring legitimacy (Deephouse, 1999; Deephouse & Carter, 2005) and affecting organizational reputation (Wonneberger & Jacobs, 2017). The news media are fundamental in the formation of the public’s perception of organizations
9 (Einwiller, et al., 2010). One way in which the media shape public perceptions is framing, which is defined by Entman (1993) as:
To select some aspects of a perceived reality and make them more salient in a communicating text, in such a way as to promote a particular problem definition, causal interpretation, moral evaluation and/or treatment recommendation. (p. 52) By using frames, the media can thus make certain issues more noticeable, as well as emphasize certain features of the issue.
Due to NPM, the people’s perception of a PSO’s legitimacy and reputation has become increasingly important for survival of the organization (Blomgren, et al., 2016). Media framing can influence this perception as it can put an emphasis on the legitimacy issues (Humphreys & Latour, 2013) and the reputation issues of the organization (Wonneberger & Jacobs, 2017).
Dutch hospitals are an example of PSOs that were reformed according to NPM. NPM managed to gain foothold in the Netherlands in the 1990s, whereupon Dutch healthcare system was revised. After years of planning, a new hospital payment system was implemented in 2006 (van Essen, 2009). Hospitals were privatized and from then on have had to negotiate with health insurance companies on the compensation they would get for performing treatments (van Weel, Schers, & Timmermans, 2012). The main aim of this privatization was to increase competition between hospitals and thereby shorten waiting times, decrease costs, and make hospitals more accountable for their performances (van Essen, 2009). As a consequence of the market-system, hospitals suddenly became “supposed to strategically communicate and market ‘who they are’, what they stand for and their ‘competitive uniqueness’” (Blomgren, et al., 2016, p. 178).
Where the NPM reforms led to a breakpoint in public sector management (van Essen, 2009), it is hypothesized that they also led to a shift in the scrutiny of hospitals in
10 the media. Before the privatization, there was no competition between hospitals, which made reputation insignificant. Legitimacy was granted by the fact that hospitals were governmental institutions, yet hospitals did need to conform to certain requirements in order to stay legitimate. Due to pressures of bureaucratization, hospitals were expected to remain similar to other hospitals (DiMaggio & Powell, 1983). Legitimacy frames were thus more present in the media than reputation frames:
H1: Before the privatization of Dutch hospitals, newspaper articles about hospitals
contained more legitimacy frames than reputation frames.
The implementation of the new hospital payment system in 2006 made hospitals more similar to commercial organizations (van Essen, 2009). As a result, their legitimacy and reputation became more important (Blomgren, et al., 2016). This has led to increased overall scrutiny of hospitals’ legitimacy and reputation:
H2: Newspaper articles about hospitals after the privatization in 2006 contained more
legitimacy frames and more reputation frames than newspaper articles before 2006. The main aim of the NPM reforms was to increase competition (Klijn, 2012). Increased competition induces rivalry and thereby boosts differentiation. Hospital reputation thus became more relevant. After the privatization, hospitals still partly derived their legitimacy from conformation with government standards and from the general belief that hospitals are necessary and therefore have a right to exist (Ruef & Scott, 1998). Consequently, reputation has become the chief topic in the media when it comes to hospitals.
H3: After the privatization of Dutch hospitals in 2006, newspaper articles about
hospitals contained more reputation frames than legitimacy frames.
As hospitals are already taken for granted, discussions about hospital legitimacy will primarily discuss normative and regulative legitimacy, rather than cognitive legitimacy.
11
H4: Frames focusing on cognitive legitimacy are less present than frames focusing on
normative and regulative legitimacy
Apart from the four hypotheses, two explorative research questions will be discussed to get a more general idea of the effect of New Public Management reforms on the media coverage of hospitals. To see whether the privatisation of hospitals has increased or decreased the amount of times certain reputation frames are being used in articles about hospitals, the following question is asked:
ERQ1: Has there been a shift in the presence of certain types of reputation frames after
the privatisation of hospitals?
There has been no previous research on the effect of privatisation on the sentiment of news coverage about an organization. Yet, the valence framing theory states that the valence of messages by the news media play an important role in the forming of citizens’ opinions. How a certain message is interpreted depends on whether the message is presented in a positive or negative way (Hurtíková, 2017). It would therefore be interesting to see if there has been a change in valence framing following the privatisation of Dutch hospitals:
ERQ2: Is there a difference between the valence of newspaper articles about hospitals
after the privatization in 2006 and newspaper articles before 2006? Methodology
A content analysis was conducted on newspaper articles about one or multiple of the five biggest hospitals in the Netherlands, based on the amount of beds in 2006: Erasmus Medisch Centrum, 1221 beds; Universitair Medisch Centrum Groningen, 1339 beds1; Universitair Medisch Centrum Utrecht, 1042 beds; Medisch Spectrum Twente, 1034 beds; Ziekenhuisgroep Twente, 1030 beds (Ministerie van Volksgezondheid, Welzijn en Sport, n.d.). The time frame used is from 2004 until 2008. This time frame was chosen because it
1
For Universitair Medisch Centrum Groningen no data on the amount of beds in 2006 was available, so the amount of beds in 2007 was used as a reference point.
12 includes two years before the introduction of a new healthcare system, as well as two years after the introduction of the new healthcare system, whereby hospitals were privatized.
It was chosen to analyse newspapers because newspapers are, and in the period between 2004 and 2008 were even more, the predominant source of information for the public (Bakker & Scholten, 2014). Furthermore, it has been found that newspapers pay more attention to organizations than television and that online news tends to be very similar to articles in print newspapers (Wonneberger & Jacobs, 2017). The articles were taken from four different newspapers: NRC Handelsblad and De Volkskrant, two quality newspapers and De
Telegraaf and Algemeen Dagblad, two popular newspapers (Bakker & Scholten, 2014). These
are the four daily papers with the highest circulation numbers in the Netherlands and they reach a large part of the Dutch population (Stichting Nationaal Onderzoek Multimedia, 2018). Furthermore, these newspapers have different views with regards to politics, religion and ideology and are therefore representative for Dutch national newspapers in general. Lastly, these newspapers are a mix different kinds of newspapers, enabling to get a more general view of the use of frames in newspapers (Bakker & Scholten, 2014).
Sampling
The articles were retrieved from the online database Lexis-Nexis. The queries included the name of the hospital and possible variations and abbreviations (see Appendix A). For each hospital, a separate search was conducted for the periods 2004-2006 and 2006-2008. Per time period, the sample was grouped into articles by quality newspapers (De Volkskrant & NRC Handelsblad) and popular newspapers (De Telegraaf & Algemeen Dagblad). From each subsample, twenty articles were randomly selected for analysis (see Figure 1). For the combinations of hospital, time period and type of newspaper where there were less than 20 articles available, all articles were selected for analysis.
13
Figure 1: Example of sampling strategy for articles about Erasmus Medisch Centrum
Limitations of this procedure are that search terms are prone to researcher bias and that the database might not be as complete as it claims to be. Since the search terms only include the names of the hospitals and therefore are factual and concrete, the chances of bias are minimized. The completeness of the database can never be guaranteed, so there is a possibility that not all the articles about the hospitals that appeared in the selected newspapers made it into the sample. It is assumed that missing articles in the database are caused by random error rather than systemic error, making that missing articles do not have an influence on the results.
Measures
According to Entman (1993), frames are “the presence or absence of certain keywords, stock phrases, stereotyped images, sources of information and sentences that provide thematically reinforcing clusters of facts or judgments” (p. 52). This research looks at the presence of sentences that indicate the condition of a hospital’s legitimacy and reputational standing. It was chosen to assess articles on sentence-level, because the amount of sentences containing a
Erasmus Medisch Centrum
2004-2006
Quality newspapers (De Volkskrant & NRC
Handelsblad)
20 articles
Popular newspapers (De Telegraaf & Algemeen
Dagblad)
20 articles
2006-2008
Quality newspapers (De Volkskrant & NRC
Handelsblad)
20 articles
Popular newspapers (De Telegraaf & Algemeen
Dagblad)
14 certain frame indicates how salient the frame is in the article. This allows for a more profound analysis of articles that contain both legitimacy and reputation framing.
Independent variable.
Before or after privatisation of Dutch hospitals. The privatisation of hospitals entered
into force on 1 January 2006. Articles published before this date thus fall into the category ‘before privatisation’, where articles published after this date fall into the category ‘after privatisation’.
Dependent variables.
Reputation framing. Firstly it was assessed if the article in one way or another spoke
about the reputation of one of the hospitals. This variable was coded positive if the article spoke about hospitals in a comparative way, literally spoke about reputation, image, status prestige or favourability of a hospital, spoke about how the organization distincts itself from others, indicated preferential status or popularity of a hospital or assessed a hospital on a level that does not fall within the normative, cognitive or regulative dimension (Deephouse & Carter, 2005). If the article contained reputation framing, the amount of sentences speaking about the reputation of a hospital was coded, in order to see how present reputation was in the articles.
Type of reputation frame. If an article contained reputation framing, it was coded
which type(s) of framing were present. The framing could be in the form of a comparison, the literal mentioning of repuation, image or status, discussing differentiation, discussing preference or through a judgement of the hospital not based on grounds of legitimacy (Deephouse & Carter, 2005; Deephouse & Suchman, 2008; Ruef & Scott, 1998).
Legitimacy framing. The article was coded to contain legitimacy framing if the article
spoke about the right to exist of a hospital, contained a judgement about the social or moral acceptability of the acting of the hospital, spoke about the hospitals compatibility with norms
15 and regulations by law, government institutions or professional institutions, stated how hospitals are expected to act in a certain way, spoke about the conformity or uniformity of hospitals (Deephouse & Carter, 2005; Ruef & Scott, 1998). If the article contained legitimacy framing, the amount of sentences speaking about hospital legitimacy was coded, in order to see how present legitimacy was in the articles.
Type of legitimacy frame. If an article contained legitimacy framing, it was coded
which type(s) of framing were present. Articles could be coded as containing cognitive framing, regulative framing, normative framing, and isomorphism frames. Cognitive framing was split up in speaking about a hospital’s right to exist and cognitive judgements of the hospital. Normative framing was split up in sentences about complying with rules and sentences about how hospitals are expected to act in a certain way (Deephouse, 1999; Deephouse & Carter, 2005; Deephouse & Suchman, 2008; DiMaggio & Powell, 1983; Ruef & Scott, 1998).
Valence framing. For coding the valence of the article, the five-point scale by
Wonneberger & Jacobs (2017) was used. This scale ranges from -2 (strongly negative) to +2 (strongly positive), where 0 stands for neutral/mixed (p. 360). To determine the valence, the use of language, frames, context and tone in relation to the hospital was taken into consideration (Hurtíková, 2017; Wonneberger & Jacobs, 2017). When coding the valence, it was assessed if the article contained pieces of text about the hospital or hospitals that were not factual, like judgments or opinions. Following, it was evaluated whether these texts were positive or negative. If the context and tone were positive or negative, without the explicit use of positive or negative words, the piece of text was seen as just positive/negative. If positive/negative words (e.g. slecht, ondermaats, onvoldoende, onbegrijpelijk, incompetent,
zwak / goed, bovenmaats, positief, succesvol, bekwaam) were clearly used to describe the
16 strongly positive/negative. Lastly, the general tone of the article was evaluated; If an article contained solely strongly negative or a mix of negative and strongly negative pieces of text, it was coded to be strongly negative. If an article contained negative pieces of text, it was coded to be negative. The same rule applied for positive pieces of text. If an article contained both (strongly) positive and (strongly) negative pieces of text, it was coded to be positive/negative according to the predominant sentiment, or it was coded to be neutral/mixed if the positive and negative pieces of text balanced each other out.
Administrative variables.
Title of the article. The title of the article with which it appeared in the newspaper.
Name of newspaper. The name of the newspaper in which the article was published.
Publication date. The date of appearance of the article in the newspaper.
Name of the hospital(s). The name(s) of the hospital(s) mentioned in the article.
Coding date. The date of coding the article.
Coder comments. Comments about doubts or problems the coder runs into or
interesting quotes from the article that can be useful for the discussion of the data.
Coding
The articles were coded according to a codebook (see Appendix B). The coding of the articles took place over two weeks, using Qualtrics online survey software. When coding the dependent variables, the presence of reputation or legitimacy frames was evaluated per sentence. If legitimacy and or reputation frames were present in the article, the coder counted and coded the amount of sentences in which this was the case. For the different types of legitimacy and reputation frames, it was only coded if they were or were not present in the article. The amount of sentences in which they were present was not coded.
17 Coder reliability.
Ten percent of the sample (36 articles) was coded twice. Once by the main coder and once by an external coder, who followed the same coder training and codebook. The inter-coder reliability in this case shows in how far the codebook enabled the coders to assess the articles in a scientific way, rather than based on personal interpretation.
Table 1
Inter-coder reliability
Variable Krippendorff’s alpha
Presence of reputation frame(s) in the article .947 Number of sentences about reputation .879 Presence of a reputational comparison .800 The literal mentioning of reputation, image or status .761 Discussion of differentiation Discussion of preference .826 .876 Presence of a reputational judgment 1 Presence of legitimacy frame(s) in the article .917 Number of sentences about legitimacy .884 Discussion of a hospital’s right to exist 1 Presence of a cognitive judgment 1
Regulative framing 1
Discussion of compliance to normative rules 1 Discussion of compliance to normative expectations
18
Discussion of isomorphism .606
For most of the variables the Krippendorff’s alpha value was above .800 and thus acceptable (Krippendorff, 2004). For six variables the inter-coder reliability was perfect, with a Krippendorff’s alpha value of 1. These were all dichotomous questions, which makes them easier to code. Furthermore, some of the legitimacy frames were very straightforward (e.g. a hospital’s right to exist, regulative framing) whereas others were almost never present in the sample (e.g. reputational judgment, cognitive judgment). Two variables scored below the threshold of .800. For the literal mentioning of reputation, image or status this was mainly caused by articles that discussed new, innovative or prestigious research by a hospital. This led to disagreement whether the terms were used to describe the hospital in general, or just the research. For the discussion of isomorphism, the coding turned out to be difficult because articles that discussed differences between hospitals did not always clearly mention the existing expectation of congruence.
Analyses
In order to test the data, statistical analyses were conducted in SPSS. All of the analyses were conducted on the article level. To detect differences in the presence of legitimacy and reputation frames in general and between the two time periods, several t-tests were conducted. Independent samples t-tests were conducted to compare the mean values from two different time periods, 2004-2006 and 2006-2008. Although the group sizes for these two periods were not the same, the Levene’s test showed homogeneity of variance for all tested variables. Paired samples t-tests were conducted to compare the mean values for two different variables in one time period (either 2004-2006, 2006-2008 or 2004-2008).
The presence of certain types of legitimacy or reputation frames was coded as a binary variable (0 = not present in the article, 1 = present in the article). The total score for these
19 variables thus shows in how many articles the specific frame was present. When calculating the mean for these variables (amount of articles in which frame is present divided by total
amount of articles), one thus actually calculates a number between 0 and 1 which shows in
which part of the total the frame occurs. Group means for these variables therefore represent the percentage of articles which contained this type of frame, rather than an average that could be interpreted on article level. These percentages were analysed to discover changes in the occurrence of certain types of legitimacy and reputation frames.
Results
Out of the sample of 353 articles, 18 articles were not coded because they were duplicates or because they did not speak about one of the hospitals. In total, 335 articles were coded. 40.6 percent (N = 136) of the articles dated from before the privatisation of hospitals and 59.4 percent (N = 199) dated from after the privatisation. Of all articles, 36.4 percent (N = 122) contained reputation framing. 22.7 percent (N = 76) articles contained legitimacy framing. 16.1 percent (N = 54) of the articles contained both reputation and legitimacy framing, while 54.1 percent (N = 191) of the articles contained neither legitimacy nor reputation framing. Table 2 shows the descriptive statistics for amount of sentences with reputation framing, the amount of sentences with legitimacy framing and the valence, for each time period.
Table 2
Descriptive statistics of continuous variables
Variable N Total Mean SD
Sentences with reputation framing 335 413 1.23 2.50
2004-2006 136 166 1.22 2.31
20
Sentences with legitimacy framing 335 277 0.83 2.40
2004-2006 136 121 0.89 2.12
2006-2008 199 156 0.78 2.58
Valence 335 -39 -0.12 0.65
2004-2006 136 -10 -0.07 0.63
2006-2008 199 -29 -0.15 0.66
Presence of Legitimacy and Reputation Frames
To see whether before the privatization articles about hospitals contained more sentences with legitimacy frames than sentences with reputation frames, a paired samples t-test was conducted with the articles from 2004-2006 (N = 136). The results indicated no significant difference between the amount of sentences about legitimacy (M = 0.89, SD = 2.12) and reputation (M = 1.22, SD = 2.31), t(135) = 1.42, p = .158, 95% CI [-1.30, 0.79]. In the period 2004 until 2006, newspaper articles about hospitals did thus not contain more legitimacy frames than reputation frames. H1 is thus not confirmed.
Two separate independent samples t-tests were performed to see if the amount of sentences with legitimacy frames and sentences with reputation frames increased after the privatization of hospitals in 2006. For reputation, there was no significant difference between the amount of frames per article in 2004-2006 (M = 1.22, SD = 2.31) and 2006-2008 (M = 1.24, SD = 2.64), t(333) = -0.73, p = .941, 95% CI [-0.57, 0.53], d = -0.01. The same holds for legitimacy, where no difference could be found between 2004-2006 (M = 0.89, SD = 2.12) and 2006-2008 (M = 0.78, SD = 2.58), t(333) = 0.4, p = .693, 95% CI [0.11, 0.27], d = 0.05.
A new variable was created which added up the number of sentences speaking about legitimacy and the number of sentences speaking about reputation. This variable contained the total amount of legitimacy-reputation frames per article. To check if there
21 was a difference in the total amount of legitimacy and reputation frames between 2004-2006 and 2004-2006-2008, an independent samples t-test was conducted. There were no significant differences between the amount of frames per article in 2004-2006 (M = 2.11, SD = 3.50.) and 2006-2008 (M = 2.03, SD = 4.59), t(333) = 0.18, p = .846, 95% CI [-0.83, 1], d = 0.02. H2 is thus not confirmed, as the total amount of legitimacy and reputation frames did not increase after the privatization of hospitals.
To see whether newspaper articles from after the privatization contained more reputation frames than legitimacy frames, a paired samples t-test was conducted with the articles from 2006-2008 (N = 199). The results indicated that in the time period 2006-2008, newspaper articles about hospitals contained significantly more sentences with reputation frames (M = 1.24, SD = 2.64) than sentences with legitimacy frames (M = 0.78, SD = 2.58), t(198) = 2.6, p = .010, 95% CI [0,11, 0,81]. This result supports H3.
Table 3
Presence of different types of legitimacy frames 2004-2008
Type of legitimacy framing Present in number of articles Percentage of total (N=76) χ2(1) value Cognitive legitimacy 15 19.7% 27.84* Right to exist 1 1.3% 72.05* Judgment 14 18.4% 30.32* Normative legitimacy 45 83.4% 2.58 Normative rules 39 51.3% 0.05 Normative expectations 13 17.1% 32.9* Regulative legitimacy 45 59.2% 2.58 Isomorphism 9 11.8% 44.26*
22 * Statistically significant at the p < 0.05 level
Note. Percentages do not add up to 100% as some articles contained multiple types of legitimacy framing. For the 76 articles in the dataset which contained legitimacy framing, it was analysed which type(s) of legitimacy framing were present. A one-sample Chi-square test was conducted to see if all of the types were equally present. The findings are summarized in Table 3. As becomes visible from the Table 3, normative legitimacy and cognitive legitimacy are present in way more articles than cognitive legitimacy. The Chi-square test showed that less articles than expected contained cognitive legitimacy framing, while the numbers of articles that contained normative legitimacy framing and regulative legitimacy framing were very close to the expected count. Cognitive legitimacy frames are thus less present than normative and regulative legitimacy frames. This supports H4. Important to note here is that the groups are not mutually exclusive, as one article could contain multiple types of legitimacy frames. Therefore the results of the Chi-square test might not be completely valid. However, they still give a good indication of how the data is distributed and how this distribution differs from the expected distribution.
Explorative Research Questions Presence of reputation frames.
To check whether there was a difference between the frequencies of certain types of reputation frames used in the periods 2004-2006 and 2006-2008, an independent samples t-test was performed for each type of reputation frame. The presence of a certain reputation frame in an article was coded with either 0 (not present) or 1 (present). The mean of the sample in this case thus represents the percentage of articles in which the reputation frame occurred. Table 4 shows some differences between the time periods, however none of the differences proved to be statistically significant. Therefore,
23 it cannot be said that there has been a shift in the presence of specific types of reputation frames.
Table 4
Presence of different types of reputation frames per time period
2004-2006 (N=136) 2006-2008 (N=199) Comparison 13% 11% Reputation/status/image 7% 5% Differentiation 16% 17% Preference 2% 2% Reputation judgment 14% 18%
Note. None of the differences is statistically significant at the p < 0.05 level.
Valence of the articles.
An independent samples t-test was conducted to see if the valence of the newspaper articles differed for the time periods 2004-2006 and 2006-2008. The results showed no significant difference between the valence of the articles in 2004-2006 (M = -0.07, SD = 0.63) and 2006-2008 (M = -0.15, SD = 0.67), t(333) = 1, p = .318, 95% CI [-0.07, 0.21], d = 0.12. The valence of the articles has thus not significantly changed after the privatization. The negative mean values do show that the general sentiment in articles about hospitals, no matter the time period, tends to be slightly negative.
Conclusions and discussion
The Dutch government introduced New Public Management reforms with the aim of increasing hospital performance. By introducing a new hospital payment system, the
24 differences between the public and private sector were minimized. Consequently, hospitals needed to deal with issues that before only belonged to the private sector. They needed to attract customers, be competitive and be accountable for their performances (van Essen, 2009). In order to do all this, it became necessary for hospitals to create their own organizational identity. As they were suddenly going to be scrutinized by customers and competitors, they needed to ensure their organization had a good reputation. Yet, as they were also still subject to government regulations and controls, they also needed to uphold their legitimacy. This resulted in a legitimacy-reputation dilemma, whereby hospitals needed to be similar to each other, but also needed to be unique (Blomgren, et al., 2016).
This research focused on the representation of this dilemma in the media. The media are crucial in conferring legitimacy and reputation, as the way they frame issues influences public perceptions (Einwiller, et al., 2010). A content analysis of the coverage of hospitals in four Dutch newspapers allowed to compare the frames used with regards to hospitals before and after the introduction of the private hospital payment system.
The results showed that before the privatization of hospitals, legitimacy and reputation frames were equally present in newspaper articles about hospitals. This goes against the expectation that the articles would focus more on legitimacy, because hospitals were already forced to conform to governmental standards and thereby judged on their level of congruency (DiMaggio & Powell, 1983). After the privatization, the total number of frames did not increase. While reputation and legitimacy did become more important for hospitals (Blomgren, et al., 2016), it can thus not be said that the legitimacy-reputation dilemma became more salient in newspapers.
However, the articles from after 2006 did show an increase in the examination of reputation of hospitals. From 2006 on, people have had the opportunity to personally select the hospital they want to be treated at (van Essen, 2009). As a result, it became
25 important to know if a hospital is ‘good’ or ‘bad’. Hospital reputation thus became a more frequently covered topic. This is in line with previous literature that states that NPM reforms lead to increased competition (van Essen, 2009; Klijn, 2012). As increased competition leads to differentiation and thereby a larger focus on reputation (Deephouse & Carter, 2005).
When it comes to the types of legitimacy frames, normative legitimacy and regulative legitimacy are clearly the most discussed topics. Hospitals have a high level of cognitive legitimacy, as their role in society is largely unquestioned and taken for granted (Deephouse, 1999; Deephouse & Carter, 2005; Deephouse & Suchman, 2008; Ruef & Scott, 1998). There is almost no discussion about hospitals’ right to exist and this is reflected in the media by the absence of cognitive legitimacy framing.
With regards to reputation, the most frequently used frames were the same before and after the privatization, namely: comparisons to other hospitals, discussions of how a hospital differentiates itself and reputational judgements. The reputational judgments were mostly focused on the quality and capabilities of employees and directors.
The overall sentiment of articles about hospitals did not change after the privatization. The largest part of the articles was neutral, some were negative and almost none were positive. This resulted in an average sentiment that was slightly negative. Before as well as after the privatization, more attention was thus given to the negative matters than to the positive matters of hospitals. This is in line with the findings of Wonneberger & Jacobs (2017) that although the evaluation of public organizations in the media is close to being neutral, media coverage about public organizations tends to be more negative than positive.
In general, the media coverage of hospitals only showed minor changes after the NPM reforms went into effect. Although the reforms led to serious transformations in
26 the management and financing of hospital (van Essen, 2009; van Weel, et al., 2012), these transformations were not reflected in the media. This suggests that once a PSO is privatized, it will not suddenly be treated as a private organization. An explanation for this could be that the media hold on to the public identity of the organization, even when it becomes a commercial organization. When an organization is long-established and taken for granted as being provided for by the state, it will not lose its public identity. This goes against the theory that NPM reforms put PSOs under a strong external pressure to change their identity (Blomgren, et al., 2016). It would be interesting to see if the retention of PSOs’ former identity also occurs among other stakeholders, such as the public and employees.
While the conclusions of this research are based on a small sample of organizations within a specific sector, the results can be interpreted in a more general way. The reforms that took place are similar to those in other sectors, in the way that they aimed at increasing competition and, through a market-based system, gave more power to customers (van Essen, 2009). These elements contribute to the importance of organizations’ reputations, no matter what specific service the PSOs provide.
Limitations and future research
This research focused on the differences between the time periods 2004-2006 and 2006-2008. It may be, however, that hospitals as well as newspapers already changed their communication strategies before 2006 in anticipation of the planned privatization. Or, on the other hand, that changes in media coverage of hospitals have some incubation time and only become visible years after changing the system. Future research could thus focus on time periods which are further apart from each other, to see if more evident differences can be found.
27 Furthermore, this research did not take into consideration all factors that could contribute to or balance out potential contrasts in media coverage of PSOs. During the coding stage it was observed that when it comes to hospitals, quality newspapers seem to be less factual and neutral than popular newspapers. The presence of (critical) opinion pieces made the valence of quality newspapers more negative than that of popular newspapers. Additionally, there seemed to be a distinction between the coverage of academic hospitals and general hospitals. The sample of this research contained many articles about academic hospitals, which mostly discussed research findings and opinions by principal researchers. Articles about general hospitals, on the other hand, paid more attention to events and incidents happening in the hospital. More research on the influence of these factors could broaden our understanding of the relationship between the media and PSOs.
For communication practitioners in the (semi-)public sector, this research shows that NPM reforms or other measures to decrease the differences between the public and the private sector do to some extent have an influence on the media coverage of PSOs. After such measures, the scrutiny of the organization’s reputation in the media becomes more important. It is thus of major importance to build and maintain a good reputation, even before the measures enter into force. Practitioners should take this into consideration when setting up communication strategies. When doing this, however, they should not try to discard their former public identity, as privatized PSOs will still be treated as public organizations. Practitioners should thus profit from the acceptance and taken-for-grantedness from being a former public organization in order to maintain legitimacy. In the meanwhile, they can use the organization’s new, private identity as an opportunity to increase its reputational standing. Altogether, organizations are still caught in a balancing act, where they need to be “as different as legitimately possible” (Deephouse, 1999, p. 147).
28 Acknowledgments
The author thanks Riordan Verhaar, who assisted as a second coder for the inter-coder reliability test.
References
Bakker, P., & Scholten, O. (2014). Communicatiekaart van Nederland: Overzicht van media
en communicatie. Amsterdam: Wolters Kluwer.
Blomgren, M., Hedmo, T., & Waks, C. (2016). Being Special in an Ordinary Way: Swedish Hospitals’ Strategic Web Communication. International Journal of Strategic
Communication, 10(3), 177-194. doi:10.1080/1553118X.2016.1176569
Cawley, A. (2012). Sharing the paint or shouldering the burden? Journalism Studies, 13(4), 600-615.
de Vreese, C. (2005). News framing: Theory and typology. Information Design + Document
Design, 13(1), 51-62.
Deephouse, D. (1999). To be different, or to be the same? It's a question (and theory) of strategic balance. Strategic Management Journal, 20, 147-166.
Deephouse, D., & Carter, S. (2005). An Examination of Differences Between Organizational Legitimacy and Organizational Reputation. Journal of Management Studies, 42(2), 329-359.
Deephouse, D., & Suchman, M. (2008). Legitimacy in Organizational Institutionalism. In R. Greenwood, C. Oliver, T. Lawrence, & R. Meyer (Eds.), The Sage Handbook of
Organizational Institutionalism (pp. 49-77). Thousand Oaks: Sage Publications.
doi:10.4135/9781849200387.n2
DiMaggio, P., & Powell, W. (1983). The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields. American Sociological Review, 48(6), 147-160. doi:10.2307/2095101
Einwiller, S., Carroll, C., & Korn, K. (2010). Under What Conditions Do the News Media Influence Corporate Reputation? The Roles of Media Dependency and Need for Orientation. Corporate Reputation Review, 12(4), 299-315. doi:10.1057/crr.2009.28 Entman, R. (1993). Framing: Toward clarification of a fractured paradigm, R. Journal of
Communication, 43(4), 51-58.
Groot, T., & Budding, T. (2008). De Ontwikkeling van New Public Management in Nederland. Maandblad voor Accountancy en Bedrijfseconomie, 82(4), 152-160. Humphreys, A., & Latour, K. (2013). Framing the Game: Assessing the Impact of Cultural
Representations on Consumer Perceptions of Legitimacy. Journal of Consumer
29 Hurtíková, H. (2017). The importance of valence-framing in the process of political
communication: Effects on the formation of political attitudes among viewers of television news in the czech republic. Medijske Studije, 8(15), 72-91.
King, B., & Whetten, D. (2008). Rethinking the Relationship Between Reputation and
Legitimacy: A Social Actor Conceptualizations. Corporate Reputation Review, 11(3), 192-207.
Klijn, E. (2012). New Public Management and Governance: A Comparison. In D. Levi-Faur,
The Oxford Handbook of Governance (pp. 201-214). Oxford: Oxford University Press.
Krippendorff, K. (2004). Content analysis: An introduction to its methodology. Thousand Oaks, California: Sage.
Ministerie van Volksgezondheid, Welzijn en Sport. (n.d.). Jaarverantwoording in de zorg. Retrieved from Jaarverantwoording in de zorg:
https://www.desan.nl/net/DoSearch/Search.aspx?action=zoeken
Ruef, M., & Scott, W. (1998). A Multidimensional Model of Organizational Legitimacy: Hospital Survival in Changing Institutional Environments. Administrative Science
Quarterly, 43(4), 877-904.
Santiago, R., Carvalho, T., & Sousa, S. (2015). NPM Reforms and Professionals in Health and Higher Education in Portugal. International Journal of Public Administration,
38(11), 757-768. doi:10.1080/01900692.2014.979199
Stichting Nationaal Onderzoek Multimedia. (2018). NPM 2018-IV. NOM.
van Essen, A. M. (2009). Seeking a Balance?! The emergence of New Public Management in
new hospital payment systems in Germany, the Netherlands and the United Kingdom.
Amsterdam / Enschede: Ipskamp Drukkers.
van Ruler, B., & Verčič, D. (2009). Reflective Communication Management, Future Ways for Public Relations Research. In P. Kalbfleisch, Communication Yearbook 29 (pp. 239-273). New York: Routeledge.
van Weel, C., Schers, H., & Timmermans, A. (2012). Health Care in The Netherlands.
Journal of the American Board of Family Medicine: JABFM, 25 Suppl 1(1), S12-S17.
Williams, H., Rayner, J., & Allinson, C. (2011). New public management and organisational commitment in the public sector: testing a mediation model. The International Journal
of Human Resource Management, 23(12), 2615-2629.
doi:10.1080/09585192.2011.633275
Wonneberger, A., & Jacobs, S. (2017). Media positioning: comparing organizations’ standing in the news. Corporate Communications: An International Journal, 22(3), 354-368.
30 Appendix A
Search queries used in Lexis-Nexis
Erasmus Medisch Centrum:
("erasmus medisch centrum" OR "erasmus mc" OR "erasmus ziekenhuis") (publication(de volkskrant) OR publication(de telegraaf) OR publication(algemeen dagblad) OR publication(NRC handelsblad))
Universitair Medisch Centrum Groningen:
("Universitair medisch centrum groningen" OR "umcg" OR "umc groningen" OR “academisch ziekenhuis Groningen” OR azg) (publication(de volkskrant) OR publication(de telegraaf) OR publication(algemeen dagblad) OR publication(NRC handelsblad)) AND NOT “artsen zonder grenzen”
Universitair Medisch Centrum Utrecht:
("Universitair medisch centrum utrecht" OR "umc utrecht" OR “academisch ziekenhuis Utrecht”) (publication(de volkskrant) OR publication(de telegraaf) OR publication(algemeen dagblad) OR publication(NRC handelsblad))
Medisch Spectrum Twente:
("Medisch Spectrum Twente" OR “ms twente” OR ("mst" AND (twente OR enschede OR ziekenhuis))) (publication(de volkskrant) OR publication(de telegraaf) OR publication(algemeen dagblad) OR publication(NRC handelsblad))
31 ("ziekenhuisgroep twente" OR "ziekenhuis groep twente" OR "zgt almelo" OR “zgt Hengelo” OR “Streekziekenhuis Midden-Twente” OR “Streek ziekenhuis Midden-Twente” OR Twenteborg OR “smt ziekenhuis” OR "ziekenhuis midden twente") (publication(de volkskrant) OR publication(de telegraaf) OR publication(algemeen dagblad) OR publication(NRC handelsblad))
32 Appendix B
Codebook
As the articles are in Dutch and the coder has the Dutch nationality, the codebook is written in Dutch. This enables for examples and descriptions which resemble or can be taken from actual newspaper articles.
Administratieve variabelen
1. Wat is de titel van het artikel? …..
2. Wat is de publicatiedatum van het artikel? (dd/mm/jjjj) …..
3. Wanneer is het artikel gepubliceerd?
- Vóór 01/01/2006 (dus t/m 31/12/2005) - Op of na 01/01/2006
4. In welke krant is het artikel gepubliceerd? - De Telegraaf
- Algemeen Dagblad (AD) - de Volkskrant
- NRC Handelsblad
5. What is de codeerdatum? (de datum van vandaag, dd/mm/jjjj) …..
6. Over welk ziekenhuis of ziekenhuizen wordt in het artikel gesproken? (meerdere antwoorden mogelijk)
- Erasmus Medisch Centrum
- Universitair Medisch Centrum Groningen (UMCG) - Universitair Medisch Centrum Utrecht (UMC Utrecht)
33 - Ziekenhuisgroep Twente (ZGT)
- Medisch Spectrum Twente (MST)
Reputatie
7. Wordt in het artikel gesproken over de reputatie van één of meerdere Nederlandse ziekenhuizen?
Codeer ‘ja’ als:
Er een vergelijking wordt gemaakt tussen ziekenhuizen. Deze vergelijking mag ook zijn tussen een Nederlands en een buitenlands ziekenhuis (Deephouse & Carter, 2005, p. 331).
Er bijvoegelijk naamwoorden of bijwoorden worden gebruikt voor het omschrijven van een ziekenhuis die duiden op een vergelijking. (Bijv. ‘het beste ziekenhuis’ ‘het slechtste ziekenhuis’ ‘het schoonste ziekenhuis’ ‘het ziekenhuis met de kortste wachtlijst’) (Ruef & Scott, 1998).
De volgende woorden (of een vorm daarvan) letterlijk genoemd worden in verband met een ziekenhuis: status, imago, prestige, reputatie, goede naam,
slechte naam, faam, vooraanstaand, toonaangevend, gerespecteerd,
respectabel, vooruitstrevend, voorloper, innovatief, uitzonderlijk, vernieuwend, indrukwekkend, lof, gerenommeerd, berucht, inferieur, minderwaardig, verouderd, achterhaald, onderontwikkeld, achter lopen (Deephouse & Carter,
2005).
Er wordt gesproken over eigenschappen waarmee een bepaald ziekenhuis zich onderscheidt van de rest op positieve of negatieve manier. (Bijv. De toepassing van nieuwe technologiën en behandelingen of het ontbreken hiervan) (Frombrun, 1996 as cited in Deephouse & Carter, 2005).
34
Er wordt gesproken over hoe er een voorkeur of juist afkeur/géén voorkeur bestaat voor een bepaald ziekenhuis. (Bijv. ‘Amsterdammers gaan liever naar het AMC dan naar het OLVG’) (Deephouse & Suchman, 2008).
Er een oordeel over het ziekenhuis wordt gegeven op een gebied dat niet valt onder de normatieve (normen opgelegd door de maatschappij, professionele standaarden) cognitieve (heeft een organisatie bestaansrecht en hoe zou een
organisatie moeten functioneren?) of regulatieve (expliticiete officiële regels en wetten waar een organisatie aan moet voldoen, opgelegd door hogere instanties zoals overheden) dimensie (Ruef & Scott, 1998).
Codeer ‘nee’ als:
Er geen enkele zin in het artikel voldoet aan één of meerdere van de bovenstaande beschrijvingen
- Ja (ga door naar vraag 8) - Nee (ga door naar vraag 10)
8. In hoe veel zinnnen in het artikel wordt er gesproken over de reputatie van één of meerdere Nederlandse ziekenhuizen?
....
9. In welke vorm(en) wordt er over de reputatie van één of meerdere ziekenhuizen gesproken?
Vergelijking. Er wordt een vergelijking gemaakt tussen ziekenhuizen (Deephouse & Carter, 2005, p. 331) / Er worden bijvoegelijk naamwoorden of bijwoorden gebruikt voor het omschrijven van een ziekenhuis die duiden op een vergelijking. (Bijv. ‘het
35 beste ziekenhuis’ ‘het slechtste ziekenhuis’ ‘het schoonste ziekenhuis’ ‘het ziekenhuis met de kortste wachtlijst’) (Ruef & Scott, 1998).
- Ja - Nee
Reputatie/status/imago. Er wordt letterlijk gesproken over de reputatie/status/imago van het het ziekenhuis. Bijvoorbeeld door het noemen van volgende woorden (of een vorm daarvan) in verband met een ziekenhuis: status, imago, prestige, reputatie,
goede naam, slechte naam, faam, vooraanstaand, toonaangevend, gerespecteerd, respectabel, vooruitstrevend, voorloper, innovatief, uitzonderlijk, vernieuwend, indrukwekkend, lof, gerenommeerd, berucht, inferieur, minderwaardig, verouderd, achterhaald, onderontwikkeld, achter lopen (Deephouse & Carter, 2005).
- Ja - Nee
Differentiatie. Er wordt gesproken over eigenschappen waarmee een bepaald ziekenhuis zich onderscheidt van de rest op positieve of negatieve manier. (Bijv. De toepassing van nieuwe technologiën en behandelingen of het ontbreken hiervan) (Frombrun, 1996 as cited in Deephouse & Carter, 2005).
- Ja - Nee
Preferentie. Er wordt gesproken over hoe er een voorkeur of juist afkeur/géén voorkeur bestaat voor een bepaald ziekenhuis. (Bijv. “Amsterdammers gaan liever naar het AMC dan naar het OLVG”) (Deephouse & Suchman, 2008).
36 - Ja
- Nee
Reputatie oordeel. Er wordt een oordeel over het ziekenhuis gegeven op een gebied dat niet valt onder de normatieve (normen opgelegd door de maatschappij, professionele standaarden) cognitieve (heeft een organisatie bestaansrecht en hoe zou
een organisatie moeten functioneren?) of regulatieve (expliticiete officiële regels en wetten waar een organisatie aan moet voldoen, opgelegd door hogere instanties zoals overheden) dimensie (Ruef & Scott, 1998).
- Ja, namelijk ... - Nee
Legitimiteit
10. Wordt in het artikel gesproken over de legitimiteit van één of meerdere Nederlandse ziekenhuizen?
Codeer ‘ja’ als:
Er wordt gesproken over het bestaansrecht van één of meerdere ziekenhuizen (Deephouse & Carter, 2005; Deephouse & Suchman, 2008; Ruef & Scott, 1998).
Er een duidelijk oordeel wordt geveld over het handelen van één of meerdere ziekenhuizen. (Bijv. Door terminologie als: wenselijk, onwenselijk, acceptabel, onacceptabel, etisch, onethisch, sociaal verantwoord) (Ruef & Scott, 1998).
37
Er wordt gesproken over het wel of niet voldoen aan regels zoals wettelijk vastgelegd, vastgesteld door overheidsinstanties of door professionele organisaties. (bijv. Een beoordeling door De Inspectie Gezondheidszorg en Jeugd of het niet voldoen aan de normen van de Nederlandse Vereniging voor Heelkunde) (DiMaggio & Powell, 1983; Ruef & Scott, 1998).
Er wordt gesproken over hoe ziekenhuizen ‘geacht worden’ iets te doen, of hoe er iets van hen ‘verwacht wordt’, waarbij het duidelijk is dat deze verwachting voortkomt uit bestaande regels of vanuit de maatschappij (DiMaggio & Powell, 1983; Ruef & Scott, 1998).
Er wordt gesproken over de gelijkheid, uniformiteit, eenheid, harmonie,
conformiteit, eenstemmigheid of congruentie van ziekenhuizen, of juist het
ontbreken hiervan (Deephouse, 1999; Deephouse & Carter, 2005; Deephouse & Suchman, 2008; DiMaggio & Powell, 1983).
- Ja (ga naar vraag 11) - Nee (ga naar vraag 13)
11. In hoe veel zinnnen in het artikel wordt er gesproken over de legitimiteit van één of meerdere Nederlandse ziekenhuizen?
...
12. In welke vorm(en) wordt er over de legitimiteit van één of meerdere ziekenhuizen gesproken? (meerdere antwoorden mogelijk)
38 Cognitief bestaansrecht/raison d’être. Er wordt gesproken over het bestaansrecht van één of meerdere ziekenhuizen (Ruef & Scott, 1998; Deephouse & Carter, 2005; Deephouse & Suchman, 2008).
- Ja - Nee
Cognitief oordeel. Er wordt een duidelijk oordeel wordt geveld over de wenselijkheid van het handelen van één of meerdere ziekenhuizen. (Bijv. Door terminologie als: wenselijk, onwenselijk, acceptabel, onacceptabel, etisch, onethisch, sociaal verantwoord) (Ruef & Scott, 1998).
- Ja - Nee
Regulatief. Er wordt gesproken over het wel of niet voldoen aan regels zoals wettelijk vastgelegd of vastgesteld door overheidsinstanties. Let op! Alleen regels die zijn vastgesteld door formele (wettelijke) instanties tellen hier mee (DiMaggio & Powell, 1983; Ruef & Scott, 1998).
- Ja - Nee
Normatieve regels. Er wordt gesproken over het wel of niet voldoen aan regelgeving vastgesteld door professionele organisaties (DiMaggio & Powell, 1983; Ruef & Scott, 1998). Let op! Regelgeving door overheidsinstanties of wetten tellen hier niet mee.
- Ja - Nee
39 Normatieve verwachtingen. Er wordt gesproken over hoe ziekenhuizen ‘geacht worden’ iets te doen, of hoe er iets van hen ‘verwacht wordt’, waarbij het duidelijk is dat deze verwachting voortkomt uit bestaande regels of vanuit de maatschappij (DiMaggio & Powell, 1983; Ruef & Scott, 1998).
- Ja - Nee
Isomorfisme. Er wordt gesproken over de gelijkheid, uniformiteit, eenheid, harmonie,
conformiteit, eenstemmigheid of congruentie van ziekenhuizen, of juist het ontbreken
hiervan (Deephouse, 1999; Deephouse & Carter, 2005; Deephouse & Suchman, 2008; DiMaggio & Powell, 1983).
- Ja - Nee
Wat is de algehele toon van het artikel over het betreffende ziekenhuis? Let hierbij op taalgebruik, framing, context en/of toon die worden gebruikt in verband met het ziekenhuis (Hurtíková, 2017; Wonneberger & Jacobs, 2017).
- -2 (sterk negatief) - -1 (negatief)
- 0 (neutraal of gemixt) - 1 (positief)
- 2 (sterk positief)
