diagnosis
How to improve the perceived
level of service quality of a
radiology department in an
intra-hospital setting
Thesis W.P. Joosten (s1334999) September 2009
Dependent on diagnosis
How to improve the perceived level of service quality of a
radiology department in an intra-hospital setting
Master thesis
Msc. Business Administration
Specialization Business Development Faculty of Economics and Business University of Groningen
Supervisor University: prof. dr. R.T.A.J. Leenders Co-assessor University: dr. M.A.G. van Offenbeek
Supervisors Radiology: ir. T. Hoogstins and ir. J.W. Hoorn.
Abstract
This thesis regards the concept of perceived service quality in an intra-hospital setting. The organization involved in this case-study is an institutional radiology department in a Dutch academic hospital. The customer is defined as the referring physician. The radiology department performs a service for these internal customers by providing them with knowledge about the nature and location of a disease in a patient’s body. In preliminary interviews complaints were heard about the service delivery to these physicians and diverse scholars recognized these problems in other institutional radiology departments (Hoe 2007).
The concept quality can mean different things to different people. Therefore the concept perceived service quality was used, defined as a form of attitude resulting from a comparison of expectations with perceptions of service performance (Kang 2006). The research’s question was: How can the radiology department improve the perceived level of service quality to their intra-hospital customers?
An adequate model designed to measure the perceived level of service quality and to find the roots of possible quality problems is the service quality model or gap model by Parasuraman et al. (1985). One gap, the customer gap, is the difference between expectations and perceptions of the service performance, which is influenced by four so-called provider gaps. The perceived level of service quality can be assessed by measuring the customer gap. The magnitude of the provider gaps can be assessed by researching their antecedents, elaborated in the extended gap model (Parasuraman et al. 1988). Diverse scholars have already used these models in comparable internal and/or healthcare settings (Wisner and Stanley 1999; Reynoso and Moores 1995).
Four hypotheses were made. The first deals with the consequences of an improvement of the perceived level of service quality, assuming that it will benefit the patient. The second hypothesis assumes that referring physicians differ in their perceptions on the level of service quality based on their departmental origin. The third hypothesis involves the question on which aspects the referring physicians weigh the most importance and the fourth hypothesis gave the management team of the radiology department more insight into the causes of the perceived level of service quality.
The research was carried out in two studies. First, a survey with a small sample size was administrated by 33 referring physicians in five selected departments, measuring the perceived level of service quality on a technical dimension (output) and a functional dimension (process). As an instrument for measuring the functional dimension, SERVPERF was chosen in favour of SERVQUAL. Second, more in-depth knowledge and detail was obtained in interviews (n=32) with radiology staff members, managers of the radiology department, referring physicians, and managers of the five selected customer departments.
Among the results is that referring physicians need a high perceived level of service quality provided by the radiology department to improve their care for patients and that they weigh the most importance to the assurance, responsiveness, and reliability of the service delivery. Moreover, their perceptions on the level of service quality do not differ based on departmental origin. Finally, the perceived level of service quality is low, especially the perceived excellence of radiology reports. Related problems were based on inadequate internal and external communication. All these problems are probable caused by ‘silo thinking’ of the departments.
Table of content
1. Introduction... 1
1.1 The radiology department... 1
1.2 Service provision of the radiology department... 2
2. Problem ... 4 2.1 Problem area... 4 2.2 Quality ... 5 2.3 Problem statement ... 6 2.4 Organizer... 6 3. Theoretical framework ... 7 3.1 Gap model... 7
3.2 Extended gap model... 10
3.3 Proposition to link provider gaps with dimensions ... 12
4. Hypotheses... 13
4.1 Hypothesis one... 13
4.2 Hypothesis two ... 14
4.3 Hypothesis three... 15
4.4 Hypothesis four ... 17
4.5 The relationships between the four hypotheses ... 20
5. Methodology ... 21 5.1 Hypothesis one... 23 5.2 Hypotheses two ... 23 5.3 Hypothesis three... 25 5.5 Hypothesis four ... 26 6. Results... 47 6.1 Hypothesis one... 49 6.2 Hypothesis two ... 51 6.3 Hypothesis three... 53 6.4 Hypothesis four ... 55 7. Conclusion... 64
7.1 Conclusion based on the four hypotheses... 48
7.2 Recommendations... 49
8. Discussion ... 53
Appendix 1 SERVPERF vs. SERVQUAL ... 60
Appendix 2 Data analysis plan ... 61
Appendix 3 Survey (in Dutch) ... 63
1. Introduction
1.1 The radiology department
Radiology is a medical specialism that uses radiation, sound, and magnetic waves to diagnose the location and/or the nature of a disease. In addition to this diagnostic activity, radiologists can treat various blood vessel disorders. This is called intervention radiology. This field of medicine started with W.G. Roentgen’s discovery in 1895, and it opened the living body noninvasively making it accessible for the study of anatomy and function (Oudkerk et al. 1994: 1). Nowadays, radiology is an important speciality in clinical medicine and Chan (2002: 639) argues it will continue to grow in the future. Computed tomography (CT) and magnetic resonance (MR) imaging offer exponential benefits to the field of medicine (Duford 2009: 16) and physicians view these techniques as the two most important medical technologies to have been developed in the past thirty years (Fuchs and Sox 2001).
Besides the medical application, the radiology department in the University Medical Centre Groningen (UMCG) wants to make a contribution to the training of radiology specialists, the medical science, and the education of general medicine students. The department has a staff of hundred seventy employees; consisting of researchers, laboratory technicians, medical specialists, managers, administrative staff, and supporting staff. The department is subdivided in three teams, where each team is assigned to a set of customer departments. Still, there is a lot of overlap between the teams.
Conceptually, the referring physicians in other departments are internal customers, defined by Gremler et al. (1994: 34) as “anyone in an organization who is supplied with products or services by others in the organization.” For an institutional radiology department, Marasco and Linton (1990: 192), Alderson (2000: 320), Johnson et al. (2006: 248), Lau (2007:4), and Hoe (2007: 644) also identified referring physicians as the primary customers.
Because referring physicians in other departments are labelled internal customers, the services provided by the radiology department to these physicians can be classified as internal services. Stauss (1995: 65) defined internal services as “services provided by distinctive organizational units or the people working in these departments to other units or employees within the organization”.
1.2 Service provision of the radiology department
On the day of the appointment, a laboratory technician accompanies the patient to the diagnostic room. This member of staff is also responsible for taking an image. Dependent on the potential disease, condition of the patient, and available time, a radiologist will decide when to view the picture. Some standard images do not come with a report, unless requested by the referring physician. With the help of voice recognition software the radiologist makes a report about the picture, in which the nature and/or the location of the disease is diagnosed. Like mentioned before, most often this knowledge is available to the requesting doctor one day later. The images of a patient are directly available for the referring physician, at the same time as a radiologist can see the images. This process is shown in figure one.
Figure 1: Process of service delivery to referring physician.
Legend: boxes are persons who perform or receive some form of activity, the black arrow line is the sequential flow in which the activities are performed, and the dotted arrow line is where information is transferred back to the referring physician. The figure shows that the referring physician receives the image at the same moment as the radiologist receives it. The patient is excluded from the figure.
Referring physician: request Technician or radiologist: application judgment Administration: planning Technician: makes image Radiologist: makes report Referring physician:
receives image and report
Request not accepted
Date and time communicated to referring physician
2. Problem
2.1 Problem area
The aim of the radiology department is to provide excellent service for their customers. But the problem the management team identified is that some departments in the hospital are not that satisfied with the service the radiology department provides. Where the focus of the radiology department lays on perfecting the pictures and corresponding reports on their own quality indicators, some customers have a different demand. The management team told that physicians in the neurosurgery department are not primarily interested in the picture and a report, but would like a radiologist to be present at their patient discussions. Some doctors at other departments have a different opinion on the quality of the reports the radiology department delivers, want other aspects to be discussed in such a report, or claim that they do not even need a corresponding report.
In the introduction, the radiology department was labelled a service organization. Services can be defined as deeds, processes, and performances (Zeithaml and Bitner 1996:5). Four characteristics distinguish services from goods (Fitzsimmons and Fitzsimmons, 1994). First, most services are intangible; precise manufacturing specifications concerning uniform quality can rarely be set. Second, services are most often heterogeneous; their performance varies from producer to producer, from customer to customer, and from day to day. Third, services are perishable and can not be stored. Fourth, the production and consumption of many services are inseparable.
The problematic service provision by the radiology department is supported by a statement made by Hoe (2007: 643) as he notes that: “Most departments based at academic hospitals need to focus more on their referring physicians than most are currently doing”. Other scholars recognize the same problems for internal services in general. Farner et al. (2001: 352) suggest that internal operations do not face market forces and therefore may not provide quality service to other departments they serve. In addition, Harari (1991:42) claims that “so-called customers feel often more controlled and confused than served by their internal supplier”. And although Erturk et al. (2005: 985) claim that the concept of quality has a long history in the field of radiology, this would not apply to the service provision to referring physicians. Hoe (2007: 646) namely plainly argues: “If we were to be honest, most of us would probably agree that most institutional radiology departments are barely able to provide good customer service”. Hence, the problem area is recognized in literature.
2.2 Quality
To the start of the analysis, a clear understanding of the concept of quality is needed. Deming was one of the first scholars in this field (Dale 2004: 53). He defined quality as “quality of design, quality of conformance, and quality of the sales and service function” (ibid.). In 1979, Crosby gave another definition of quality: “Quality is conformance to agreed and fully understood requirements” (Dale 2004:8). This definition inhabits that quality is an attribute; a characteristic which, by comparison to a standard or reference point, is judged to be correct or incorrect (ibid.).
Nevertheless, this definition is based on research in the goods sector and knowledge about quality of goods is insufficient to understand quality in services (Parasuraman et al. 1985: 41). In services, an objective assessment of quality is nearly impossible because of the earlier mentioned characteristics of a service. A more usable concept for services is
perceived quality, defined as a consumer’s judgment about an entity’s overall excellence
2.3 Problem statement
The radiology department provides a service to other departments in the hospital. The management team of the radiology department wants to know what their perceived level of service quality is and how they can improve this perceived level of service quality. This leads to the main question, which needs to be answered:
“How can the radiology department improve the perceived level of service quality to their intra-hospital customers?”
The purpose of this thesis is to “write a report in which the current state of the radiology department’s perceived level of service quality to their intra-hospital customers is described, and how this perceived level of service quality can be improved.”
2.4 Organizer
Hereafter, an overview of the thesis is given. The theoretical framework (chapter 3) handles the theoretical models and a proposition. In the next chapter (chapter 4), four hypotheses were based on literature and pilot-interviews. In the last paragraph of this chapter the relation between the four hypotheses and the research question is clarified.
In chapter 5 the method of data collection is explained to answer the hypotheses. In the appendix a theoretical discussion about the SERVQUAL instrument, a data analysis plan, and the survey (in Dutch) can be found.
3. Theoretical framework
3.1 Gap model
Since the start of the 1990s the concept of perceived service quality got a lot of (research) attention with specialized professional journals, academic journals, conferences, institutes, and books (Ballantyne et al. 1995: 7). These initiatives were necessary, as scholars claim that perceived service quality is an elusive and abstract construct that is difficult to define and measure (Parasuraman et al. 1985: 42; Brown and Swartz 1989: 92; Carman 1990: 34; Cronin and Taylor 1992: 55; Abdullah 2006: 33). Kasper et al. (1999: 184) described five approaches to quality: the transcendent approach based on psychology, the product-based approach based on economics, the user-based approach originating from marketing and operational management, the manufacturing-based approach with roots in operational management, and the value-based approach.
One user-based model became widely cited in literature and widespread in the industry, namely the service quality—or gap-model (Brown et al. 1993: 127; Asubonteng et al. 1996: 62; Carrillat et al. 2007: 473). This model for the measurement of perceived service quality was developed in 1985 and is refined in the years thereafter. The gap model is illustrated in figure 2. The model shows how perceived service quality emerges. The upper part of the model includes phenomena related to the customer; the lower part demonstrates phenomena related to the service provider (Blois and Grönroos 2000). This model therefore perfectly matches a quotation by Mindak and Folger (1990:1): “In striving for a competitive edge, service producers learn early on that marketing’s basic role is to make promises and to build expectations for a service’s performance. The task then remains for the organization’s operations support system to deliver on these promises the most efficient way possible”.
The model is designed to measure service quality as perceived by the customer. This is the most important gap; in their original research the authors argued that the quality a customer perceives in a service is a function of the magnitude and direction of the gap between
expected service and perceived service (Parasuraman et al. 1985:46). The expectations of
Figure 2: GAP model (Zeithaml et al. 1988: 36)
Legend: The dotted black line distinguishes the service provider from the customer. The black arrows represent lines of influences. For example, the external communication of the service provider influences the perceived and expected service. The red arrows are the gaps, which represent a potential misfit between two items in the model. Like provider gap two, this is a misfit between what management thinks customers expect and the translation into service quality specifications.
Word of mouth communications
Personal needs Past experience
Expected service
Perceived service
The process of closing this ‘customer gap’ (gap 5 in figure 2) can be subdivided in closing four ‘provider gaps’ (Zeithaml and Bitner 1996: 38). If the gaps are closed, the customer will perceive a high level of service quality. These provider gaps and the customer gap are listed in table 1.
In this setting, provider gap one is the discrepancy between what the management team of the radiology department thinks the referring physicians expect and what they really expect. Provider gap two is the difference between the management team’s ideas about what the referring physicians expect and the translation into service standards and specifications. Next, provider gap three is the discrepancy between these standards and the actual performance by radiologists and other staff members. Provider gap four is the incongruence between the communication of the members of the radiology department about their service delivery and the performance perceived by the referring physicians. Finally, the customer gap in this setting is the discrepancy between the expectations and the perceptions of the referring physicians on the performance of the radiology department.
The structure of the gap model demonstrates where the roots of quality problems may be found and how problems may accumulate. The model also shows which steps have to be considered when analysing and planning service quality. At the same time, it indicates what types of corrective actions may be needed if the perceived quality is to be improved (Blois and Grönroos 2000). This makes the model very useful for the purpose of this thesis; assessing the perceived level of service quality and exploring the causes of this level.
Table 1: Gaps of the service quality model. Four provider gaps influence the customer gap.
Provider gap 1:
Not knowing what the customer expects
The knowledge gap
Provider gap 2:
Not selecting the right service designs and standards
The service design and standards gap
Provider gap 3:
Not delivering to service standards
The service performance gap
Provider gap 4:
Not matching performance to promises
The communication gap
Gap 5:
The difference between the expectations and perceptions of the customer on the service performance
Frost and Kumar (2000) modified the gap model to an internal service setting. However, this model is not completely valid because it focuses on a lower aggregation level: the relation between support and front line staff, without management intervention or communication. A better fit to the aim of this thesis is the original model. In earlier published articles, Parasuraman et al. already argued that the gap model can be used by every service organization. They claimed that it can be used by departments within an organization to ascertain the perceived level of service quality they provide to other departments (1990: 180). In addition, other scholars (Lings and Brooks 1998: 325; Wisner and Stanley 1999: 27) also believe that the gap model can be transferred from the external to the internal customer and Reynoso and Moores (1995: 80) even validated this in a hospital setting. The wide appliance of the model of Parasuraman, Zeithaml and Berry in comparing settings confirms the applicability to this study’s setting. Parasuraman et al. (1990: 180) and Large and König (2009: 25) nevertheless note that the wording in the measurement instrument should be modified to be administered to a sample of internal customers.
3.2 Extended gap model
In 1988, Parasuraman et al. wrote an article where they proposed the ‘extended gap model’ (Zeithaml et al. 1988). In this article they propose a set of antecedents to the provider gaps, based on marketing literature, organizational behaviour literature, and their own exploratory research. They argue that the process of closing the ‘customer gap’ (gap 5 in figure 2) can be subdivided in closing these four ‘provider gaps’. To assess the magnitude of the provider gaps, a researcher needs to study the antecedents (Zeithaml and Bitner 1996: 38). In the years thereafter, the extended gap model and its antecedents have been modified (Zeithaml and Bitner 1996; Nel and Boshoff 1997; Wilson et al. 2008). Some scholars used the extended gap model on a departmental level (Chaston 1993; Pitt et al. 1988), indicating the applicability of the model to this study’s setting. The antecedents are drawn in figure 3.
These dimensions are a representation of criteria that customers employ in evaluating perceived service quality. The instrument the authors developed is called SERVQUAL and measures the difference between expectations and perceptions of customers on these dimensions (Parasuraman et al. 1985: 42). The standardized SERVQUAL-survey consists of 22 questions regarding expectations, 22 questions concerning perceptions, and questions to assess the importance of the dimensions. In 1993 Parasuraman et al. called for the validation of the dimensions in specific industries (1993: 144). Nevertheless, Alderson (2000: 319) believes that these five dimensions are “fundamental service principles that can easily be related to radiology practice”.
Figure 3: Extended gap model (Zeithaml et al. 1988)
Legend: On the left side the antecedents are shown. The boxes on the right side are the provider gaps, and as is shown in the model, they all influence the perceived level of service quality (gap 5). For example, the items in the first box on the left are antecedents to provider gap one, which influences the perceived level of service quality.
Inadequate market research orientation Lack of upward communication
Insufficient relationship focus
Poor service design
Absence of customer-defined standards Inappropriate physical evidence
Deficiencies in human resource policies Customers not fulfilling roles
Ineffective alignment with service intermediaries Failure to match supply and demand
Inadequate service recovery
Inadequate horizontal communication Lack of integrated service communication
Inefficient management of customer expectations Over-promising
Gap 2: Not selecting the right service designs and standards
Gap 3: Not delivering to service standards
Gap 4: Not matching performance to promises
Gap 5: Perceived level of service quality
Gap 1: Not knowing what the customer expects
3.3 Proposition to link provider gaps with dimensions
In this paragraph a proposition is made. The five dimensions assigned to the SERVQUAL instrument can be linked to the provider gaps in the extended gap model. Notwithstanding the fact that both models originate from the same authors, the linkage between both models is hypothetically. As a result, the validity of this proposition is unknown.
The first dimension ‘empathy’ is the individualized attention to customers. This can be linked to provider gap one; the radiology department should know the preferences of the customers to provide customized services. The second relationship is more disputable. But an antecedent to provider gap two is the ‘appropriateness of physical evidence’ which can be linked to ‘tangibles’. The third provider gap can be matched to ‘assurance’ (as skilled staff members should meet certain specifications) and ‘responsiveness’ (as the service should be delivered according to (time) standards). A clear link can also be made between the antecedent ‘match of supply and demand’ with ‘responsiveness’, as an aligned supply with demand will enable the radiology department to provide their services promptly. Lastly, ‘reliability’ means that the radiology department should keep its promises and communicate an integrated message to all customers. This matches provider gap four. The proposed links between the dimensions and the provider gaps can be seen in table 2.
Table 2: Proposed link between dimensions and provider gaps. Dimensions Provider gap
4. Hypotheses
Based on the gap model, the extended gap model, literature, and three pilot interviews with staff members of the radiology department, four hypotheses are set. These hypotheses reflect the most important aspects to be researched, both from an academic as from a practical standpoint.
4.1 Hypothesis one
The first hypothesis deals with the consequences of service quality improvement of the radiology department. Lau argued that quality in a radiology practice is interlinked with workload, accuracy, safety, and turn-around time (2007:3). But besides these benefits for the radiology department and benefits to referring physicians, a higher perceived level of service quality would perhaps also benefit the external customers (patients). The importance of internal service provision is underlined by Brooks et al. (1999: 64) as they claim that the main driver for internal service quality appears to be the external customer, who makes demands on the customer contact personnel. The internal service provider should therefore support the external service provider. However, despite the argumentation of Alderson (2000: 319), who believes that radiology services are essential to the care of patients, Strifle et al. (2007: 773) argue that the measurement of the impact of the activities of a radiology department on clinical care is still problematic.
Departments in the UMCG hospital which receive services from the radiology department have little or no freedom of choice between service providers. In other words, the internal supplier has a monopoly status (Stauss 1995: 69) and the internal customers are captive (Nagel and Cilliers 1990 in Gremler et al. 1993: 38).
In addition, Reynoso and Moores (1995) conducted a study on the perceived level of internal service quality in a hospital, claiming that improving internal service quality would benefit the service quality to the end customer. Therefore, hypothesis one proposes that the perceived level of service quality by the radiology department is essential for the level of service quality a customer department provides an external customer. In a hospital this will mean that the quality of patient care and cure.
H1: The perceived level of service quality delivered by the radiology department influences the level of service quality referring physicians deliver to their patients.
4.2 Hypothesis two
The customer gap in the gap model of Parasuraman et al. is a function of the magnitude and direction of the gap between expected service quality and perceived service quality (Parasuraman et al. 1985:46). The second hypothesis deals with the variety among departments in their perceptions on the service quality the radiology department provides.
The difference in perception can be based on personal characteristics or departmental origin. The age of a referring physician can be relevant, as Farner (2001: 356) found that an older respondent will probably perceive the level of service quality lower compared to younger respondents. Other issues are found by Hoe, as he notes (2007: 645) that a physicians’ referral decision for a radiology service is usually complex and influenced by professional (high quality), practical (in a timely fashion), patient related (satisfactory for the patient), and personal (convenience) issues. The professional issues can accurately be assessed by a referring physician, as they are educated in their medical specialism at their department.
Three findings originating from healthcare management literature support the previous made notions, as Joseph (1996:57) argues that “cultural diversity is a reality in many hospitals and this poses special problems in service delivery”. Zabada et al. (1998: 57) further elaborate on this general remark as they argue that the variety of powerful subcultures in healthcare impede management control about quality. Each subculture or department in healthcare has its own perspective of what quality should be, which will influence their expectations and perceptions of the service the radiology department delivers. Lastly, Stauss argues that organization variables like hierarchy and departmental origin influence quality expectations and perceptions greatly (1995:75).
Based on their own knowledge and departmental culture, each customer department will have their own perception on the level of service quality the radiology department provides.
H2: There is a high variety among departments in their perception on the level of service quality the radiology department provides.
4.3 Hypothesis three
To measure the customer gap (gap 5 in the gap model) Parasuraman, Zeithaml and Berry (1991) listed five dimensions that influence customers’ assessment of service quality. The scores on these dimensions would represent the perceived level of service quality from a customer’s perspective. The third hypothesis deals with the dimensions on which customers assess the perceived level of service quality. The question is which dimension is most important in the eye of the customer. These five dimensions (reliability, assurance, tangibles, empathy, and responsiveness) in the measurement model are discussed before. Usually, customers weigh the dimension ‘reliability’ as the most important (Zeithaml et al. 1990: 28). But internal customers in a hospital setting, like referring physicians, will probably judge service quality differently. The dimension ‘tangibles’ is probably not important to them, based on literature. When Brooks et al. (1999: 58) and Farner et al. (2001: 335) found in preliminary interviews that ‘tangibles’ was not a relevant factor to internal customers they excluded it from their measurement instrument.
An experienced physician specialized in cardiology will probably interpret a heart image better or comparable to a specialized radiologist. Nevertheless, some scholars found in their studies that the quality of diagnoses in reports and consultation is important to a referring physician (Marasco and Linton 1990: 193; Alderson 2000: 322), leading to the statement that ‘assurance’ is an important dimension.
But based on other studies on the performance of radiology departments, ‘responsiveness’ is more important than ‘assurance’. Seltzer et al. (1992) conducted market research on the services of an academic radiology department and identified three key issues for referring doctors: waiting time to get an appointment, the scheduling processes and procedures required to get an appointment, and the communication of findings (especially the timeliness). The same results were found in studies of Lopiano et al. (1990: 1328), Marasco and Linton (1990: 193), Alderson (2000: 322), and Johnson et al. (2006: 548). Next, Mozumbar et al. (2003: 908) found in their study that in institutional hospitals 33% percent of the referring physicians would change radiology service provider if expectations of scheduling ease are not met.
All these findings seem to resemble ‘responsiveness’ as the most important dimension, by Berry et al. (1991) defined as “the willingness to help customers and to provide prompt service”. In addition, Strifle et al. (2007: 772) argue that the main goal of a radiology department should be to provide diagnostic interpretation in a timely fashion. Often radiology departments fail to reach this goal as Hoe (2007: 646) claims that “those long waiting times for radiology appointments are simply not acceptable, even at institutions”.
The radiology department in the UMCG already focuses on this aspect, as is agreed upon in the so-called ‘service level agreements’ (Slack et al. 2004:725). Although these agreements between the radiology departments and other departments are implicit, they are mainly focused on access times and waiting times. In a pilot interview ‘responsiveness’ was also considered the most important dimension, as a radiologist argued: “Most referring physicians want their patients to be helped directly. The image and corresponding radiology report should be available soon thereafter”.
All these statements support the notion made by Frost and Kumar (2000: 374). They studied the SERVQUAL-dimensions in an inter-organizational context and found that ‘responsiveness’ was the most important of the five dimensions.
4.4 Hypothesis four
The gap model of Parasuraman et al. assumes that all four provider gaps should be closed to close the customer gap (Zeithaml and Bitner, 1996: 38). However, not all provider gaps may have the same magnitude and influence on the perceived level of service quality. To study these two elements, Parasuraman et al. wrote an article in which they proposed the extended gap model (1988). The fourth hypothesis concerns the magnitude of the four provider gaps and their influence on the perceived level of service quality.
First, based on literature some arguments can be made about the relationship strength between the provider gaps and the perceived level of service quality. However, scholars disagree on this issue. Parasuraman et al. (1990:32) found only weak associations, but gap three had the highest correlation with the customer gap, followed by gap two and four. Provider gap one would have a minor effect on the perceived level of service quality. This is a contradiction with a study by Nel and Boshoff (1997), who found that gap one is the only latent variable significantly influencing the customer gap.
Based on literature, the department will score positive on an antecedent of gap 1. Luk and Clayton (2002: 123) found that most managers do not have a reasonable good understanding of customers’ expectations. Joseph therefore notes that senior management in healthcare should get close to their internal customers as a mean to improve the overall level of service quality (1996:58). In the radiology department, almost all members of the management team are practitioners. This would consequently mean that they have a good insight in the needs of the customer.
In summary, literature argues the probable large magnitude of provider gap 2 and 4. But these studies are not validated in a healthcare setting.
Two antecedents are irrelevant in this study. In the original study of Parasuraman et al., an important factor was the behaviour of intermediaries (Wilson et al. 2006: 110). In this setting, no intermediaries (gap 3) are involved in the provision of service to the customers. The second modification is the absence of price as a mean of communication. In the Dutch healthcare system, prices between departments are non-negotiable and stable (gap 4). In a study done by Hoe (2007: 645) he also concluded that price is not a major factor for referring specialists as to decide to which radiology centre they will send a patient. This is because the costs are paid by third-parties as insurance companies and the government. Therefore both these antecedent do not need to be studied in this setting.
Table 3: Probable importance of provider gaps based on arguments found in literature and pilot interviews. Three pilot interviews were held with members of the radiology department. Based on the arguments found in literature and pilot interviews the researcher assigned the most importance to provider gap three.
Influence on customer gap Magnitude Provider
gaps Literature Literature Pilot interviews
(radiology dep.)
Importance (assigned by researcher)
1. small / large small - small
2. moderate large - moderate
3. large moderate large large
4. moderate large - moderate
An overview of all the arguments is drawn in table 3. Although literature indicates the magnitude of gap four and the influence on the perceived level of service quality of gap one, provider gap three is probably the most important gap. The rationale is that Parasuraman et al. (1990) argued that gap three has the largest influence on the perceived level of service quality and in pilot interviews items concerning this gap were mainly exposed. The researcher assigned more weight to the pilot interviews, because these findings are grounded in the specific setting of this study.
H4: Provider gap three is the most important gap, because of its magnitude and influence on the perceived level of service quality.
4.5 The relationships between the four hypotheses
The four hypotheses all reflect an important piece of the puzzle that needs to be solved to improve the perceived level of service quality of the radiology department in the UMCG. With hypothesis one the importance of an improvement of the perceived level of service quality can be understood. A high level of perceived service quality does not only benefit the radiology department or the referring physician, but also the patient. Hypothesis two gives more insight into the extent to which degree departments differ in their perceptions on the level of service quality the radiology department provides.
Hypothesis three deepens this issue and supposes that referring physicians focus mainly on the ‘responsiveness’ dimension when they assess the service quality of the radiology department. The rejection or non-rejection of this hypothesis combined with the results of hypothesis two will give the department the most important areas for improvement of their perceived level of service quality. The model of Parasuraman, Berry and Zeithaml namely indicates what types of corrective actions may be needed if the perceived quality is to be improved (Blois and Grönroos 2000). This will depend on the magnitude of the four provider gaps and their influence on the perceived level of service quality. So, hypothesis four will result in an analysis of the problem areas inside the radiology departments which influence the perceived level of service quality negatively.
Starting with the end-result to the patient, next researching what the perceived level of service quality is, how the perception on service quality varies between departments, studying which aspects of the service provision are the most important to referring physicians, and finally ending with an internal analysis to improve the perceived level of service quality, these steps will result in a research answering the main question: “How can
5. Methodology
Parasuraman et al. (1988: 40) argue it is desirable to couple quantitative and qualitative service quality research. One form of research without the other would be insufficient, because qualitative research provides information on issues that require attention and quantitative research gives managers priority order for action (Balantyne et al. 1995: 16). A survey will be used as a quantitative research tool, as Wisner and Stanley (1999: 28) argue that subjective ratings are a reliable alternative to actual performance data. The usage of the concept of perceived service quality also consequently means the usage of subjective ratings. So, to answer the four hypotheses, both quantitative and qualitative research methods were used.
As a sample of all the different customers, the management team of the radiology department identified a selection. These departments have a variety of size, number of requests, and needs. These departments are internal medicine, orthopedics, intensive care, children’s clinic, and neurology. In their study on the relation between a radiology department and their customers, Mozumbar (2003: 910) also made a selection which included internists and orthopaedics. All hypotheses were assessed with the help of a survey, interviews, and observation. An overview is presented in table 4.
Table 4: Method of data collection (H= hypothesis. M: = method)
H1 The perceived level of service quality delivered by the radiology department influences the level of service quality referring physicians deliver to their patients.
M: First phase: survey
Second phase: interviews with referring physicians
H2 There is a high variety among departments in their perception on the level of service quality the radiology department provides.
M: First phase: survey
Second phase: interviews with referring physicians
H3 Responsiveness is the most important dimension on which referring physicians assess the perceived level of service quality of the radiology department.
M: First phase: survey
Second phase: interviews with referring physicians
H4 Provider gap three is the most important gap, because of its magnitude and influence on the perceived level of service quality.
M: First phase: survey and observation
The rationale behind the choice of the items in the survey and the method of analysis is elaborated in a data analysis plan, which can be found in the appendix. The survey had an indicative character and was used as a point of departure for a more thorough analysis with the help of interviews. Combined with time restrictions, a small sample size was chosen. The disadvantage of using a small sample for one specific case study is that it prevents any generalization of the results to a larger population (Brooks et al. 1999:65).
However, for this study the survey was a starting point for interviews which provided more in-depth understanding. To collect the surveys, several specialists were contacted though e-mail to make an appointment. The considerations for this method were practical as physicians are restricted to a tight time schedule, and often not willing to answer a mail or electronic survey. For example, Beeksma (2009) conducted her research in the UMCG by surveying nurses, but had a response of only 30%. Another advantage of this method was the possibility for the respondents to ask questions for clarification to the researcher and vice versa.
By administrating the survey at six physicians per customer department, the sample size was aimed at n = 30. A survey in a study by Mozumbar et al. (2003: 910) aimed at a radiology department was also eventually filled in by 33 referring physicians. This was therefore a comparable sample size to strife for, in addition to the argument that the results of the survey were used as an indication. To reach this number of respondents a request to cooperate was sent to a larger number of physicians, because of the assumed non-response.
The main scoring method was a Likert scale, just like in the original SERVQUAL instrument (Zeithaml et al. 1990: 175). Respondents had to assign a score from one till seven on diverse statements related to the four hypotheses. A score of ‘one’ means that they totally do not agree with the positively stated statement, while ‘seven’ means they fully agree. The numerical score reflects the degree of attitudinal favourableness to the statement. The Likert scale produces interval data (Cooper and Schindler 2003: 253).
5.1 Hypothesis one
The extent to which internal service quality benefits external service quality was assessed with a survey and interviews. In the survey respondents were asked to assess to which extent they think that high-quality service of the radiology department enables them to improve the quality of their job. Next, they were asked to which extent they need clear images, excellent reports, and regular consultation with radiologists to perform their job. By describing the mean and standard deviation information was gathered to know if physicians are dependent on the perceived service quality the radiology department provides. The interviews provided more in-depth information and were held with several referring physicians. The number of physicians interviewed depended on the satisfactory principle (Boeije 2005: 52). This means when no new information was revealed, no new interview was planned. Eventually, fifteen referring physicians were interviewed for answering both hypothesis one, two, three, and four. An overview of the amount of interviewees can be seen in table 5. One referring physician was not willing to share his opinion about the perceived level of service quality of the radiology department in an interview.
5.2 Hypotheses two
Second, although the SERVQUAL instrument is widely used, Cronin and Taylor (1992) developed the SERVPERF-instrument where only customer perceptions on the five dimensions are measured. Diverse scholars argue the superiority of simple performance-based measures of service quality (Bolton and Drew 1991: 381; Brown et al. 1993: 138; Mehta and Durvasula 1998:49; Lee et al. 2000: 217; Wilson et al. 2008: 135). These scholars in favour of SERFPERF argue that performance perceptions are the result of customers’ comparison of the expected and actual service (Babakus and Boller 1992: 264; Boulding et al. 1993:24; Carrilat et al. 2007: 476).
Service improvement can thus be done by altering the expectations or improving the actual service delivery. In addition, Abdullah (2006:32) believes that “customers’ assessment of continuously provided services may depend only on performance”. This applies to the setting in the thesis, where customer departments continuously receive services from the radiology department. Hence, SERVPERF was the instrument to measure hypothesis two. But the survey was modified, as was recommended by scholars (Carman 1990: 35; Brown et al. 1993: 139; Bolton and Drew 1991: 384). With the help of three referring physicians the survey was configured to increase the validity of the instrument. The survey (in Dutch) is displayed in the appendix.
5.3 Hypothesis three
Hypothesis three assessed the relative importance of the dimensions in the SERVPERF-instrument. Based on literature it was assumed that referring physicians will weigh ‘responsiveness’ as the most important facet, compared to the other dimensions ‘reliability’, ’assurance’, ‘tangibles’, and ‘empathy’. In earlier studies the relative importance of these dimensions was measured by a weighing factor (Zeithaml et al. 1990: 177; Mehta and Durvsala 1998: 43), so in the survey the same kind of question was used. Respondents were asked to rank order the dimensions, by assigning points to the dimensions. The ordinal data was analysed with a Friedman test, assessing if the medians of the rank ordered dimensions are equal (Cooper and Schindler 2003: 556; Huizingh 2004:331). Packed with this information interviews were conducted with referring physicians afterwards, to acquaint more in-depth knowledge to answer hypothesis three.
5.5 Hypothesis four
To assess the importance of the provider gaps, both the magnitude and the influence on the perceived level of service quality were analysed.
5.5.1 Magnitude of provider gaps
First, in their book, Zeithaml et al. (1990) used quantitative methods to describe and measure the magnitude of the provider gaps. Some of these questions were the foundation of the survey items which were applied to issues like the quality of communication, the role of the customer, customer needs, and the internal organization of the radiology department.
By describing the means and standard deviations of these variables, information was gathered to get a preliminary insight into the most problematic factors related to the provider gaps. In interviews was elaborated on specific issues that yielded a particular low score or a high level of variance. These insights were the starting point for a series of structured interviews. This is because of the nature of the extended gap model, which required a relative complex research design, as data needed to be gathered from several different actors: customers, customer-contact service employees, and managers (Wetzels 1998: 51).
Most scholars therefore use interviews to asses the magnitude of the provider gaps in the extended gap model (Samson and Parker 1994; Dann 2008). The diversity of interviewees and the potential variety of interpretations made interviewing also better applicable to assess the antecedents to the provider gaps.
Table 5: Number of interviewees and hypotheses. Staff members included radiologists, laboratory technicians, and administrative staff.
Interviewees Amount of interviewees Hypotheses
Referring physicians 15 H1,2,3,4
Managers of customer departments 5 H4 Management team of radiology department 4 H4 Staff members of radiology department 8 H4
To assess if and how customers fulfil their role and responsibility (gap 3), earlier studies were used and the researcher was planning to observe the amount of incorrect requests during one day. But because the researcher was not skilled enough to assess the content of the medical requests, two physician assistants were willing to help and looked into a sample of ten requests from the internal medicine department. Despite the reliability of a small sample, the insights were used in follow-up interviews.
5.5.2 Influence of provider gaps on the customer gap
Second, the influence of the provider gaps on the perceived level of service quality was more difficult to analyse. However, three approaches were used. First, in interviews with the management team of the radiology department was asked which antecedents they think are the most important for their perceived service quality.
Next, a correlation matrix was drawn from the variables that give insights in the provider gaps and the perceived level of service quality on a functional and technical dimension. Because the Likert scale produces interval data, a bivariate correlation was drawn (Cooper and Schindler 2003: 571). The linkage between the variables in the survey and the provider gaps is listed in table 6.
Table 6: Variables in the survey that are linked to the provider gaps.
Question Gap
7 The department is aware of my expectations 1 8 I regularly meet with a staff member of the radiology
department to express my needs
1
12 The internal organization of the radiology department is excellent
2 and 3
9 The communication with a staff member of the radiology department is always excellent
4
10 Staff members of the radiology department never over promise
4
Because provider gaps two and three mainly involve the internal organization of the radiology department, these gaps were difficult to assess by the referring physicians. For that reason a general question about the perception of the internal organization was asked (question 12), which can be assigned to provider gap two (the service design and standards gap) and provider gap three (the service performance gap).
Third, the results found in hypothesis three gave an insight in the importance of the provider gaps. In hypothesis three the relative importance of the five functional dimensions of service delivery was measured. These dimensions can be linked to the provider gaps, thereby weighing importance to the provider gaps. In a way, this method let referring physicians assign weight to the provider gaps. The proposition of this linkage is elaborated in the theory section of this thesis.
6. Results
Although 46 requests were sent to referring physicians, the survey was eventually filled in by 33 respondents. Five physicians did not answer the e-mail request for an appointment, three were on holiday, two were currently not working in the hospital, two were not involved in the clinical work at the moment, and three respondents were not willing to fill in the survey. This resulted in a response rate of 70 %. Two departments were willing to distribute the survey among their specialists, both for the intensive care and the orthopaedics department this resulted in a high number of surveys collected. The higher response of physicians working in the children’s clinic originates from their willingness to cooperate, resulting in a low non-response. The number of respondents per department is shown in table 7.
Like mentioned before, the main scoring method was a Likert scale, as respondents had to assign a score from one till seven on diverse statements related to the four hypotheses. A score of ‘one’ means that they totally did not agreed with the positively stated statement, while ‘seven’ means they fully agreed. The numerical score reflects the degree of attitudinal favourableness to the statement (Cooper and Schindler 2003: 253). Descriptive statistics of all the variables can be found in table 23 in the appendix.
Table 7: Number of respondents of every department.
Table 8: Descriptive statistics of the technical and functional dimensions of the perceived level of service quality.
* A reliability analysis was performed on the five functional dimensions. When the value of Cronbach’s alpha was lower than 0.7, the variable with the largest negative influence was deleted to increase the reliability of the dimension.
Dimensions Questions Cronbach’s alpha * Mean
Clarity of images Q1 - 5.2 Content of reports Q2 - 4.0 Responsiveness Q13 till Q16 0.73 3.7 Q22 till Q25 0.51 Tangibles Q22 excluded 0.71 4.6 Reliability Q17 till Q21 0.74 4.2 Empathy Q26 till Q30 0.86 4.0 Q31 till Q33 0.61 Assurance Q32 excluded 0.79 4.4
Alderson (2000: 319) argued that the SERVQUAL dimensions could easily be applied to a radiology practice. The reliability analysis confirms that statement. Next to the results of the reliability analysis, the factor analysis did also identify the five functional dimensions in somewhat the same factors. The factor analysis can be found in the appendix.
6.1 Hypothesis one
The perceived level of service quality delivered by the radiology department influences the level of service quality referring physicians deliver to their patients.
The first hypothesis was tested in the survey with four questions. The first question referred to the degree in which more accurate, reliable, and better radiology diagnostics would improve the level of patient care referring physicians deliver. Respondents thought this was the case, as the mean (on a Likert scale from one till seven) was 5.65 with a standard deviation of 1.27. Other questions were aimed at looking to which extent this can be applied to the clarity of the images, the content of the radiology reports, or regular interaction with radiologists. The factor mean of these three variables (‘usefulness of radiology services’) was high with a mean of 5.86.
Clear imaging was the most important (6.34) and the content in the reports (5.52) the least important to the respondents. Considering the perceived performance on these aspects: the clarity of images scores relatively high (5.19), but improvements can be made on the content of the reports (4.03). Consultation with radiologists (5.59) is also important to referring physicians. In interviews with referring physicians the dimensions ‘assurance’ and ‘responsiveness’ were identified as the most relevant regarding consultation with a radiologist. Looking at the perceived performance on these dimensions, most improvement should be made on the dimension ‘responsiveness’ (3.68). The descriptive statistics are in table 9.
Table 9: Descriptive statistics for hypothesis one.
Lastly, in interviews respondents were asked to which extent they use radiology diagnostics. Internists claim they need radiology diagnoses in about 30% of their cases, although one claimed he used radiology diagnostics only “once a week”. For neurology this percentage increases to approximately 90% percent. A specialist explains: “Everything we do is based on imaging. The only exceptions are nerves, but we do not conduct a lot of research on this field in neurology in this hospital”.
In the children’s clinic the usage of radiology diagnostics varies a lot (10% - 90%), depending on the sub specialism of the physician. An intensive user group is the children’s oncologists, as a physician explains: “I use radiology diagnostics for every patient to locate and measure the growth or decline of tumours in a patient’s body”. Two other departments where almost every patient is diagnosed with the help of imaging are orthopaedics and intensive care. For example, radiology imaging in the latter department was used every day to see if a ventilation pump in a patient’s lungs is still in right spot.
Based on the results of the survey and the interviews with referring physicians, the perceived level of service quality provided by the radiology department positively influences the level of service quality physicians deliver to their patients, where clear images are the most important to them. Hypothesis one is not rejected.
These findings, combined with the extensive usage of the diagnostics, underline the relevance of the radiology department and its perceived level of service quality.
Variables and factors Mean Stand. dev.
Q3 improvement consequences for job performance respondent 5.65 1.27 Factor ‘Usefulness of radiology services’ 5.86 0.78
Q4 need for clear imaging to perform job 6.34 0.83
Q1 clarity of images 5.19 1.20
Q5 need for excellent content in report to perform job 5.52 1.26 Q2 content of radiology reports 4.03 1.45
Q6 need for consultation with radiologist to perform job 5.59 0.85 Dimension ‘Responsiveness’ 3.68 1.00
6.2 Hypothesis two
There is a high variety among departments in their perception on the level of service quality the radiology department provides.
The second hypothesis is measured with the help of interviews and a survey. To see whether the perceptions differ based on the departmental origin of the respondent, a Kruskal-Wallis analysis was performed. The Kruskal-Wallis analysis revealed that the departments do not differ in their assessment of the perceived level of service quality, except on the content of the radiology reports (p is 0.06). These results can be seen in table 10.
Looking at the descriptive statistics in table 11, it revealed that respondents from the neurology department and especially the orthopaedics departments perceive a very poor performance of the radiology department on the content of the reports. But because of the small sample size, one respondent could have caused a major distortion of the results. For that reason a more thorough analysis was conducted with the help of interviewing. But also here it was found that respondents of the same department have differing perceptions on the service quality of the radiology department, while especially the neurologists and orthopaedics were dissatisfied with the content in the radiology reports.
In interviews the orthopaedics and neurologists told “we often do not need a report, because we are also educated to interpret a radiology image”. An earlier made statement by a neurologist (“almost everything we do is based on imaging”) supports this statement, as he did not talk about “radiology”, but only about “imaging”.
Because this aspect was also assessed in the survey (Q5), it was possible to statistically test this assumption. A Kruskal-Wallis analysis identified the differing opinions among departments on the necessity of a radiology report for the referring physicians (p is 0.02). The descriptive statistics also show that neurologists and orthopaedics are in less need of radiology reports in comparison with their colleagues at other departments. In interviews orthopaedic surgeons also identified “regular disagreement with radiologists about their interpretation of images” and lacked “the amount of knowledge about skeleton radiology in this hospital”. Not surprisingly, the dimension ‘assurance’ yielded the lowest score by referring physicians working in that department (3.8).
Table 10: Kruskal-Wallis analysis. Because alpha (0.1) exceeds the p-value (0.059), the departments seem to have differing perceptions on the content of the radiology reports. The same applies to the necessity of radiology reports (p is 0.023).
Test Statisticsa,b
Responsiveness Tangibles Reliability Empathy Assurance
Q1 (clarity images) Q2 (content reports) Q5 Necessity of report Chi-Square 2,784 ,831 5,993 1,576 4,160 2,306 9,080 11,317 df 4 4 4 4 4 4 4 4 Asymp. Sig. ,595 ,934 ,200 ,813 ,385 ,680 ,059 ,023
a. Kruskal Wallis Test
b. Grouping Variable: department
Table 11: Perception of the five departments on the level of service quality provided by the radiology department. Departments Intensive care Internal medicine Children
clinic Neurology Orthopedics
Mean Mean Mean Mean Mean
Clarity of images 5,4 5,4 5,5 5,0 4,6 Assurance 4,2 4,9 5,1 4,3 3,8 Responsiveness 3,8 3,8 3,7 2,7 3,8 Tangibles 4,6 4,6 4,8 4,7 4,3 Reliability 4,1 4,7 4,7 3,8 3,6 Empathy 3,7 4,0 4,2 3,9 4,1 Content of reports 4,6 4,4 4,6 3,7 2,6 Necessity of report 5,5 6,2 6,4 4,3 4,8
6.3 Hypothesis three
Responsiveness is the most important dimension on which referring physicians assess the perceived level of service quality of the radiology department.
For the radiology department is it important to know on which aspects referring physicians assess the perceived level of service quality. In this study the aspects are derived from the gap model; namely reliability, assurance, tangibles, empathy, and responsiveness. Based on literature, hypothesis three assumes that referring physicians assign the most weight to responsiveness.
In the survey the respondents were asked to assign hundred points to the five dimensions of SERVQUAL (question 33). Contrary to the expectations, a statistical analysis with the Friedman test revealed that ‘assurance’, ‘responsiveness’ and ‘reliability’ are the three most important dimensions. The results of this analysis can be found in table 12. In interviews most importance was allocated to the technical dimension (“The images and reports should be good”), followed by assurance (“It all depends on personnel with expertise”), reliability (“Their service provision should work flawless”), and responsiveness (“I need the radiology diagnosis soon”). The technical dimension is of course interlinked with the functional dimension assurance: staff members with expertise can make clear images and excellent reports. The dimensions ‘empathy’ and ‘tangibles’ were of minor relevance to the referring physicians. A neurologist was very straightforward in his argumentation: “I never look on their website. And individual attention is also totally irrelevant”. It was therefore not surprising that these dimensions yielded a low score in the Friedman analysis. In the survey the question aimed at the perceived modernity of the website of the radiology department also yielded a very high amount of missing values (13 out of 33).
Table 12: Friedman test. A higher score means the dimension is more important to the respondent. The results of this analysis are significant, because p (0.00) < alpha (0.1).
Table 13: Perceived performance on dimensions. Important dimensions are marked grew.
Dimensions Mean Responsiveness 3.7 Empathy 4.0 Reliability 4.2 Assurance 4.4 Tangibles 4.6
The perceived performance on the five functional dimensions of service quality in combination with the findings of the Friedman analysis gives an insight into the improvement areas for the radiology department. The perceived performances on the dimensions are shown in table 13, where the most important dimensions—according to the referring physicians—are in the grey rows. The perceived performance on assurance yield a high score (4.4), responsiveness a low score (3.7), while the other important dimension–-reliability–-scores around average (4.2).
6.4 Hypothesis four
Provider gap three is the most important gap, because of its magnitude and influence on the perceived level of service quality.
Based on literature and preliminary research hypothesis four was formulated that provider gap three is the most important gap, because of its influence on the perceived level of service quality and its magnitude.
6.4.1 Magnitude of provider gaps
To assess the magnitude of the provider gaps, the researcher needed to study the antecedents (Zeithaml and Bitner 1996: 38). These antecedents are shown in figure three. Based on the data collected in the survey and by obtaining more in-depth knowledge in following interviews the researcher was able to assign a score to the antecedents and thereby estimate the magnitude of the provider gap. The scores are a + (meaning that the department performs well on this aspect), a 0 (the department performs average on this aspect), and a – (the department does not perform well on this aspect). At the end of each paragraph the score on the antecedents are summarized.
6.4.2 Provider gap one ‘Not knowing what the customer expects’
Three antecedents influence provider gap one; relationship focus, upward communication, and market research orientation. The first is relationship focus, an important antecedent as Kennegy et al. (1999: 664) point out that “leadership must forge dynamic relationships with like-minded, service oriented partners”. Most managers at customer departments were moderately satisfied with the relationship focus of the radiology department. But they recommended that improvements in this field would yield more benefits: “By regular interaction, a better understanding between the different specialities will develop”.
