All of the most pressing problems executives face involve a lapse in reliability (Weick and Sutcliff, 2001). Naevestad (2009) has suggested a major challenge for leaders is the ability to integrate reliability at all organizational all levels (macro, micro, and meso). Creating high reliability requires an in-depth understanding of safety culture. Guldenmund (2007) has argued that research of safety culture is fragmented into functional, interpretive, and HRO perspectives. The functional perspective, which predominates in governmental agencies, focuses on critical macro level variables of safety and reliability. Several safety scholars argue that capture of critical micro level variables requires an interpretive understanding of shared communication patterns and shared identity Richter and Koch, 2004; Alvesson, 2002; Gheradi and Nicolini, 2002; Glendon and Stanton, 2000; Pidgeon and O'Leary, 2000). And Naevestad (2009) argues that important meso levels have been completely left out of the safety culture discussion. He suggests a comparative analysis of HROs must take place at the group level. This of HROs must take place at the group level. This research plans to study and integrate all three levels of reliability.

Safety culture is a relatively new and diverse field of inquiry. Defining, measuring, and assessing it are difficult because of the misalignment of the three important research perspectives. Scholars approach safety culture from functionalist, interpretive, and high reliability orientations. The purpose of this research is to develop an understanding of safety culture that incorporates technology, culture, and structure through organizational communication by studying a high reliability emergency management system – Gulf Coast City Fire Fighters.

Gulf Coast City Fire Fighters (a pseudonym) was officially organized in 1888 making it one of the oldest fire departments in the United States. Today the department is staffed by over 450 uniformed men and women. Their mission is to provide the highest quality of fire, rescue, and emergency medical services to citizens in their city. Their safety culture includes comprehensive training, emergency planning, code enforcements and innovative safety awareness programs. The Gulf Coast City Fire Fighters (GCC FF) is a multi-functional emergency management system with the responsibility of providing a broad range of emergency services to their community.

Professional firefighters (FF) have a stressful and dangerous occupation that ranks fifth in occupational mortality in the United States (Beaton, Murphy, Pike, Corneil, 1997). Evidence suggests that most FF mortality and morbidity is related either directly or indirectly to the stressful nature of their work (Beaton and Murphy, 1993). In addition FF spend long hours together at the firehouse. FF eat and sleep together creating an occupational culture which is unique when compared to other emergency workers (Corneil, 1993).

In 2010 the International Association of Fire Fighters unanimously passed a resolution to censure GCC Fire Department Chief Alpha (a pseudonym). Three major issues were present during the tenure of Fire Department Chief Alpha and resulted in the censure: an Insurance Service Office (ISO) rating downgrade, discriminatory hiring practices, and low employee moral that resulted from a benefits reduction (Teske, 2010).

The overall objective of this study is to evaluate GCC FF communication, and provide information and recommendations that may help to improve its overall organizational communication and have a positive effect on its safety culture (Jones, 2002). Specific objectives include determining the amount of information under or overload, and evaluating information sources and the quality of relationships between upper management and the rank-and-file.

Naevestad (2009) states that safety culture is a crucial factor for the safety of organizations but it lacks proper conceptualization of the relationship between technology, culture, and structure. A major problem with most of the research on safety culture is that it is a complex intersection of changing technology, highly trained people, and organizational constraints which make it difficult to study holistically (Guldenmund, 2007; Richter and Koch, 2004; Glendon and Stranton, 2000; Pidgeon, 1998. Consequently safety culture is often broken into its Consequently safety culture is often broken into its component parts and studied as individual phenomenon leaving important relationships unexamined. This research may bring together otherwise neglected phenomenon and highlight important relationships.

The field of safety culture research is divided by three competing perspectives: functionalist, interpretive, and high reliability (HRO) frameworks. This diversity explains the fragmentary and contradictory approaches to safety culture and how to gather information about it Glendon and Stanton, 2000. Comparing functionalist, interpretive, and HRO approaches demonstrates the different levels of abstraction: functionalists focus on macro organizational levels, interpretives focus on the situated micro levels, and HRO on the meso level. This research series employs functionalist, interpretive, and HRO frameworks and may provide new linkage to these previous competing perspectives.

Naevestad (2009) states that safety culture is treated as a noncontextual phenomenon that is detached from technology, culture, and structure. This research proposes to utilize all three perspectives in a series of three articles to build a contextual understanding of this complex phenomenon. This research studies the component parts as part of a comprehensive whole in an effort to examine and understand important relationships.

This article focused on the functionalist perspective. Functionalist, as the name implies, examines the functions and tasks in organizational environments. The functionalist perspective dominates the field of safety research and is the prevailing perspective at GCC FF. At the fire house, task is understood to mean activities and time spent “on a run” whereas time in the fire house is the “down time” and consequently unexamined. The reason that it is so important to distinguish between run and down time at the fire house is because the great majority of communication interactions occur in the down time. Moreover, run time conversations are highly stylized and reflect hours of crisis training, whereas down time conversations are largely unexamined.

According to Naevestad (2009) functionalist safety culture views culture as shared patterns of behavior. A functionalist approach to culture views it as a critical variable that influences technology and reliability outcomes. Functionalist safety culture research has mainly been conducted in conformity with traditions of social and organizational psychology. Operating from an organizational psychology perspective, von Thaden and Gibbon (2000) offer a definition of safety culture that is strongly influenced by upper management and includes personal responsibility, learning from mistakes, and communication. They define safety culture as:

1) The enduring prioritization of the value of worker and public safety in every level of the organization.

2) Every organizational member commits to personal responsibility for safety.

3) Every organizational member strives to actively learn and adapt.

4) Communication is key.

Safety culture is typically examined through a questionnaire survey. While several studies of safety culture in organizations demonstrate that this approach can only provide a superficial conceptualization of culture, this research builds from the strengths that functionalist perspectives can add to a comprehensive approach to understanding the combination of three cultural levels (micro, meso, and macro). Organizational communication provides a natural bridge to all three levels.

Jones (2002) quoting Putnam (1982) has suggested the formulation (numbered below) to distinguish between functionalism and interpretivism in organizational communication. The value of making such a binary split in this essay is to highlight functionalism, which permeates the literature of both safety culture and the communication audit. The functionalist framework for organizational communication typically works as follow:

1. A sender-receiver model of communication, relying on…

2. A positivist and mechanical model of organization as objective, machine-like entities with communicative channels and barriers; organized through top- down management;

3. An agenda for inquiry that is set by senior managers via strategic discourses; and

4. An approach to organizational change based on identifying and solving problems defined by senior managers.

From a functionalist perspective, communication does not constitute organizational life itself but can be defined as a series of specific behaviors, a vital binding ingredient, holding together all the working parts of the organization (Hargie and Tourish, 2000).

There is frequent slippage from communication to information, especially via the familiar engineering metaphor of message transmission through channels. As an alternative, this study proposes to follow Jones' (2002) model of a communication auditor who can listen with a trained interpretive ear to a range of organizational voices, provide feedback offered in an open and tentative spirit, and collaborate with members of the FF to frame inquiries, carry out investigations, and generate knowledge about communication practices.

Jones' (2002) model is based on collaboration language in the workplace. This model focuses on reflective processes Schon, 1983 and draws on the philosophies of the learning organization Senge, 1992 and of appreciative inquiry Srivastva and Cooperrider, 1990. This approach avoids looking for communication problems or applying packaged solutions to individual communication skill problems. Rather it sees the organization holistically as a communicative system. It seeks to involve the whole organization in a reflexive learning community that builds on existing strengths and enables a focus on organizational rather than individual competencies.

This research treats this Fire Fighting (FF) community as an organizational communication system Jeffres, 2008. Jeffres (2008) suggests the International Communication Association Communication Audit (ICA CA) as a tool for collecting and evaluating an organization's communication system. DeWine (2001) states that the Organizational Communication Division of ICA devoted more than five years to the development of the Communication Audit. This research employs the survey portion of the Communication Audit to begin to evaluate the Gulf Coast City Fire Fighters (GCC FF) communication, and provide information and recommendations that may help to improve its overall safety culture by evaluating communication processes Jones, 2002. Specific objectives include determining the amount of information under or overload, evaluating information sources and the quality of relationships between upper management and the rank-and-file. This study's objectives are to profile communication behaviors and suggest a set of recommendations for new communication practices that may improve the safety culture.

The key research question is whether or not there is a difference between upper management (Executive Fire Chiefs) and the rank-and-file (District Fire Chiefs, Captains, and below) in terms of responses to various dimensions of the ICA CA. To this end, the authors examine the responses to the ICA CA of both groups along the following dimensions:

R1 Receiving information from others,

R2 Sending information to other,

R3 Follow-up information sent,

R4 Interpersonal communication relationships, and

R5 Organizational outcomes

And because of the unanimous resolution to censure GCC Fire Department Chief Alpha this study expects to find significant issues along all five dimensions.

This section has two parts, a description of the instrument and information about the sampling procedure. A communication audit is one of many diagnostic tools used by the communication discipline. The International Communication Association Communication Audit (ICA CA) is the most well known and widely used (DeWine, 2001).

The International Communication Association (ICA) developed a multi-instrument approach for auditing the communication behavior of organizations known as the ICA Communication Audit Goldhaber, 2002. The theoretical objective of the ICA CA was to develop a standardized procedure for assessing organizational communication systems. Although the ICA CA ceased to exist as an official ICA sponsored project, the audit methodology consisting of surveys, interviews, diaries, and network analysis became standard research tools due to the ease of development, administration, and interpretation. Researchers and consultants use the ICA CA instruments independently or in various combinations depending on the needs of the organization.

As the theory and practice of the communication audit has proliferated, the ICA CA has become a sophisticated precision instrument generating volumes of data that can guide an organization though the intricacies of organizational change (Shaffer, 1993). This research employs the ICA CA sur vey for auditing the communication behavior of GCC FF. At a practical level the ICA CA may offer factual insight into the organization's communication structure Jeffres, 2008. This research employed the ICA CA survey which consists of 120 communication questions.

Data were obtained from GCC FF surveys. The surveys were administered to a representative sample at each bureaucratic level. Three tiers of respondents completed the ICA CA survey (N = 228). The first tier is composed of200 FF at the rank of Captain and below. The second tier is made up of 15 District Fire Chiefs. Finally, the third tier contains 10 Executive Fire Chiefs.

FF do shift work and are on a rotating schedule, so appointments were scheduled with random crews throughout the GCC domain to collect the first tier of 200 FF information. Additional appointments were scheduled at the main firehouse to collect 10 of the 13 Executive Chiefs. Finally, all 15 of the District Chiefs were surveyed and interviewed at an offsite location. All interviews were tape-recorded and transcribed.

The analysis proceeds in three phases. First the alpha coefficients are examined to ensure that the scales are reliable. Second, the two groups (1 – Executive Fire Chiefs, hereafter to referred to as E, and 2 – District Fire Chiefs and below, hereafter referred to as CB) are examined along each question item by either creating a difference score between what they say they needed and what they received for each item, or reporting mean scores. To calculate the difference scores the mean score for what E say they receive in terms of item Q1a “how well am I doing my job” is subtracted from the mean score for what E says they need in terms of item Q1a. The same is done for CB and the two difference scores are compared. Positive difference scores indicate areas where needs exceed what is received and thereby describes a negative situation; conversely, negative difference scores indicate areas where what is received exceeds what the group feels it needs. The scenario is positive to the extent that there is no gap between what the group receives and what it needs but may also be negative to the extent that resources are devoted to unneeded items. Third and finally, a series of bivariate regressions are conducted to determine if there are statistically significant differences between E and CB along the five dimensions.

First the scales are examined to see if they are reliable. Table 1 presents information on each of the scales. The first scale is q1_1 “Receiving information from others – how much received.” This scale combines items 1a through 1i from the ICA CA questionnaire regarding how much information from others they actually receive. The alpha for this scale is 0.6888 which is right on the cusp of the 0.7 standard suggesting good inter-item reliability. It's complementary scale is a1_2 “Receiving information from other – how much needed.” This scale also combines items 1a through 1i but in terms of how much Fire Fighters feel they need. The alpha for this scale is 0.7933 and well above the .7 standard.

Table 1. Summary of ICA-CA Scales

The second scale is q2_1 relates to “Sending information to others – how much received.” This scale combines 2a through 2g regarding how much Fire Fighters receive. The alpha is 0.7271, close to but still above the .7 standard indicating good reliability. It's complement, scale q2_2 relates to “Sending information to others – How much needed,” combines the same items (2a-2g) but as it related to how much Fire Fighters feel the need. The alpha is 0.7910, indicating good inter-item reliability.

Scale q3_1 combines items 3a through 3e and looks at “Following-up information sent – how much received.” It's complementary scale, q3_2 looks at the same set of items but as they relate to how much Fire Fighters need. The alphas are respectively 0.6810 and 0.8232. While the alpha for q3_1 is on the cusp, it satisfies the standard for good reliability. The alpha for q3_2 is well above the standard.

Q4 represents the scale for “ Inter personal communication relationships.” It combines items 4a though 4i. The alpha for Q4 is 0.8203 and indicates very good inter-item reliability. Finally, Q5 represents the scale for “Organizational degree of inter-item reliability.

Here all the individual items are examined as they relate to the responses of the Executive Fire Chief (E) and District Fire Chiefs and below (DB). In order to better summarize the information difference scores between items 1a through 3e are presented rather than presenting each individual item by both how much they receive and how much they need. The difference score is created by subtracting what is received from how much is needed. This score not only helps to demonstrate differences between E and CB but also shows where are the gaps between what is received and what is needed (positive values) and where needs are over met based on how much is received (negative values).

Chart 1 looks at items 1a through 1i that are related to receiving information from others. In terms of 1a “How well I am doing my job” and 1b “My job duties” both E and CB say that their needs are exceeded by how much they receive; although in the case of 1a, what CB's receive is not very much more than what they need, while E's are receiving much more than they need. In 1c “Organizational policies” both groups again receive more than they need but for CB's this is only slightly more while for E's it is considerably more than what they need. Item 1d “Pay and benefits” presents the starkest difference between E and CB. E's say that they receive quite a bit more than they need while CB's say exactly the opposite, they need considerably more than what they receive. For items 1e “How technological changes affect my job,” 1f “Mistake and failures of my organization,” 1g “Promotion and advancement opportunities,” 1h “New program, product or service developments,” and 1i “Specific problems faced by management” both E and CB say that they need more than they receive. For each of these items though there are clear differences between E and CB. For items 1e and 1f the needs gap is considerable for E while somewhat more modest for CB. In items 1g, 1h, and 1i this trend is reversed. Overall the gaps are more modest but for these items CB need much more than they receive relative to E.

In Chart 2 the difference scores for items 2a thru 2g related to sending information to others are presented. In item 2a “Reporting what I am doing in my job” CB say that they are doing more of this than they need to, while E say that they need to do more. CB in item 2b “Reporting what think my job requires me to do” are at an equilibrium saying that what they receive matches what they need; however, E say that they need more than they receive. Items 2c “Reporting job-related problems,” 2d “Complaining about my job/work conditions,” 2e “Requesting information necessary to do my job,” 2f “Evaluating the performance of my immediate supervisor,” and 2g “Asking for clear work instructions” all show that needs exceed what is received for both E and CB. For items 2c and 2e the gaps for both E and CB are similar. However, for items 2d and 2g the gap between what is needed and what is received is considerably greater for E than for CB. This is reversed in item 2f where the gap between needs and what is received is much greater for CB than for E.

The difference scores in Chart 3 examine items 3a thru 3e that deal with sending follow-up information. In items 3a “Subordinates,” 3b “Co-workers,” and 3c “Immediate supervisor” there are major differences between E and CB. In all three items E say that they need more than they receive (quite a bit more in the case of items 3a and 3c). In contrast, CB say in all three items that what they receive is more than they need, albeit their needs are not vastly exceeded. In item 3d “Middle management” both E and CB say that they need more than they receive with E suggesting a larger gap than for CB. Similarly for item 3e “Top management” both groups again say that they need more than they receive but here CB say that they need much more than they are receiving while the gap for E is much more modest.

Chart's 4 and 5 present mean values rather than difference scores. Response categories for items in both charts ranged from 1 “very little” to 5 “very great.” Thus higher mean values indicate a more positive response to the item while lower mean values indicate a more negative response.

Chart 4 looks at interpersonal communication relationships in items 4a through 4i. Items 4a “I trust my coworkers,” 4b “My co-workers get along with each other,” 4c “My relationship with my co-workers is satisfying,” 4d “My immediate supervisor is honest with me,” 4e “My immediate supervisor listens to me,” and 4f “My relationship with my immediate supervisor is satisfying” all show relatively strong positive reactions to these statements across both E and CB. Major differences emerge though in items 4g “I trust top management,” 4h “top management is sincere in its communication efforts,” and 4i “My relationship with top management is satisfying.” For each of these items E generally provides strong positive reactions; however, CB is considerably more negative in their evaluation of each of these statements.

Finally Chart 5 examines items 5a thru 5k regarding organizational outcomes. In item 5a “My job” both E and CB provide similarly strong positive evaluations. Item 5b “My pay” reflects the greatest difference between E and CB among these items with E giving a strong positive evaluation and CB much more negative. Item 5d “My chances for getting ahead” is the only item where CB provides a stronger positive evaluation than E; however, the difference between the two groups is not very great. In the remaining items 5c “My progress in the organization now,” 5e “My opportunity to make a difference,” 5f “Organizations systems for recognizing and rewarding performance,” 5g “Organizations overall communicative effort,” 5h “My organization compared to similar organizations,” 5i “Organizations overall efficiency of operation,” 5j “Overall quality of my organizations product or service,” and 5k “My organization's achievement of its goals and objectives” evaluations are all moderate to strongly positive but E generally have mean ratings around a full point higher than CB on each.

Table 2 presents the results of a series of bivariate regression equations. Each of the eight scales (Q1_1, Q1_2, Q2_1, Q2_2, Q3_1, Q3_2, Q4, and Q5) were treated as dependent variables. The independent variable for each of the eight different models was the dichotomous variable related to the type of position the respondent held: either 1) Executives (E, N=7), or 2) Chiefs and below (CB, N=171). There is a very low N size for E reducing expectations for statistically significant relationships; therefore, when such relationships are found it should magnifies that differences really exist between E and CB. Regression models including the scales Q1_2 “Receiving information from others – how much needed”, Q2_1 “Sending information to others – how much receive”, and Q2_2 “Sending information to others – how much needed” did not attain significance and therefore are not presented in the findings.

Table 2. Summary of Bivariate Regressions of each Scale by Executives vs. Chiefs and Below.

The scale Q1_1 “Receiving information from others – how much receive” is included even though technically it does not attain statistical significance at the minimum threshold of .05; however, at .0504 it is close enough to warrant inclusion and consideration with the results of the remaining significant models. Models for Q3_1 “Followup information sent – how much receive,” Q3_2 “Followup information sent – how much needed,” and Q4 “Interpersonal communication relationships” were all statistically significant at the .05 level. The model for Q5 “Organizational outcomes” was statistically significant at the .01 level.

None of the significant models explain much of the variation in the scale (dv). The R2 values showing how much variation is explained range from a low of 2.2% to a high of only 5.3%. However the more interesting aspect of the findings is that in each case the b demonstrates a positive slope. As an independent variable, type of position held (either E or CB) was introduced as a dichotomous variable with CB = 0 and E = 1. Therefore the b or unstandardized regression coefficient shows information relevant to E relative to the information of CB which is contained in a (the constant). Thus, the positive slopes indicate that in each of these scales E score higher on the scales than do CB. The a or constant values show the average score on the scale for CB while the b value show the E position relative to CB. For example then in the Q3_1 scale, the average position for CB is 9.48 (remember from Table 1 the range for Q3_1 is 1-19). Since the b is 3.23, these results suggest that the average position for E is 12.71 (9.48 + 3.23). What this demonstrates then is that on five of the eight scales from the ICA-CA, there are real and systematic differences between E and CB among the Fire Fighters examined.

This section revisits the objective of this study to evaluate Gulf Coast City Fire Fighters (GCC FF) communication and provide understandings of safety culture. The preceding section demonstrated the significance in information load, evaluating information sources, and the quality of relationships between upper management and the rankand- file. This section highlights and expands specific findings in an effort to shed light on the complex intersection between technology, people, and organizational structure.

The concept of safety culture is used to explain the lack of knowledge and understanding of risk by the employees and organization which contribute to disaster (Reason, 1998; Schein, 1992; Pidgeon, 1991). From a functionalist perspective, communication does not constitute organizational life itself but can be defined as a series of specific behaviors, a vital binding ingredient, holding together all the working parts of the organization Hargie and Tourish, 2000. Organizations with a positive safety culture are characterized by communication based in mutual trust and shared perception of the importance of safety and the efficacy of preventive measures Glendon and Stanton, 2000.

At GCC FF the lack of understanding contributing to the risk of disaster is not related to the task portions of their jobs, but to all the pressing and unresolved social relationships about their jobs. While functionalist safety culture views culture as shared patterns of behavior, its focus is possible external threats. GCC FF has serious internal conditions which have smoldered undetected and present a real threat to the efficacy of preventive measures. The scale development demonstrates that groups with in the organization are attempting to exchange information at significant levels. So it would appear that if transmission were the only necessary factor for communication each group believes it is doing an adequate job. Functionalist, as the name implies, focuses on the tasks in organizational environments and within this perspective, the task of information transfer is being completed. Since the functionalist perspective dominates the field of safety research and is the prevailing perspective at GCC FF organizational members focus on the function of the information transfer and believe that the job of communicating is complete. However the dissatisfaction with the content of information transferred can be seen in the charts.

Evidence in the chart 1 reveals a stark difference between E (Executive Fire Chiefs) and CB (District Chiefs and below). E's say that they receive quite a bit more than they need while CB's say exactly the opposite, they need considerably more than what they receive. This problem is especially evident in item 1d pay and benefits. Clearly there is a serious disconnect between upper management and the ranks. An examination of chart 2 shows the inverse of the problems reported in chart 1. In item 2a reporting what I am doing in my job, CB say that they are sending more information than they need to while E say that they need more information. Then, in items 2d and 2g the gap between what is needed and what is received is reversed and considerably greater for E than for CB. This is again reversed in item 2f where the gap between needs and what is received is much greater for CB than for E.

An examination of the organizational hierarchy chart offers a clue to this puzzle of the restricted information flow presented in charts 1 & 2. Chief Charlie is a gatekeeper of all of the information moving down from the E and up from the CB. Interview data which is reported in later accounts supports the identification of Chief Charlie as a serious bottleneck.

This bottleneck can also be seen in chart 3 sending followup information. In all three items E say that they need more than they receive (quite a bit more in the case of items 3a and 3c). In item 3d both E and CB say that they need more than they receive with E suggesting a larger gap than for CB. Similarly for item 3e both groups again say that they need more than they receive but here CB say that they need much more than they are receiving while the gap for E is much more modest. The lynch pin for these two groups E and CB is Chief Charlie.

While safety research suggests a major problem with safety culture is the complex intersection of technology, people, and organizational structure, GCC FF's organizational chart indicates that, at least, their information transfer problems, may well be related to one person, Chief Charlie. Of course this does not excuse upper management or the ranks from the responsibilities of their jobs, but it may help to explain the information flow problems.

The organizational chart reveals a gate keeper, Chief Charlie, who may be responsible for this bottleneck presented in charts 1, 2, and 3, but it may be organizational structure which is, at least, partly responsible for the mean values in charts 4 and 5. GCC FF operates from a functionalist tradition of a positivist and mechanical model of top-down management and a sender-receiver model of communication. But the lack of trust and respect from the ranks represented in 4g, 4h, 4i signals a deeper problem. In the functionalist tradition strategic discourse is set by senior managers, yet chart 4 and the FF interview data does not support upper management's power in the organizational structure. So the question arises, is it the organizational structure which has removed upper management from the knowledge of the fire department or is it upper management's dictates do not create strong organizational structure? Either way CB is negative in their evaluation.

Using functionalist approach to GCC FF demonstrates that of five of the eight scales from ICA-CA there are real and systematic differences between E and CB among the 228 Fire Fighters examined. This questionnaire data points to problems in the GCC FF, but it actually raises more questions than it answers. This data signals real issues, a possible bottleneck because of a gatekeeper, and perhaps more importantly, a serious lack of mutual trust and respect. As the first of four articles, this study provides a solid starting point for building toward a comprehensive approach.

The overall objective of this study was to evaluate Gulf Coast City Fire Fighters (GCC FF) communication, and provide information and recommendations that may help to improve its overall safety culture by evaluating communication practices and organizational effectiveness. Specific objectives included determining the amount of information under or overload, evaluating information sources and the quality of relationships between upper management and the rank and file. This study's objectives were to profile communication behaviors and suggest a set of recommendation for new communication practices which may improve the safety culture. The data clearly demonstrated statistical significant and real empirical differences between upper management (Executive Fire Chiefs – E) and other groups within the rank-and-file (District Fire Chiefs and below – CB) of the organization.

While safety culture is a relatively new field of inquiry this study begins to bridge the differences in defining, measuring, and assessing the safety culture. This research utilized the functionalist perspective; however, the overall approach includes all three perspectives in a series of three articles. The next paper will investigate organizational discourse from an interpretive framework with data collected through in-depth interviews. The third and last paper in this trilogy will triangulate the survey and interview data and work from a high reliability perspective fulfilling the over all goal of developing an understanding of safety culture that incorporates technology, culture, and structure through organizational communication.

Reason (1997) states research on safety culture is fragmented and unsystematic. The main critique is that the safety culture concept is not seen as a contextual phenomenon, but as a general model without consideration of the social factors of the work-life within am given organization. This research is the first in a three part series. Future research will examine the interpretive and HRO perspectives in an effort to build a comprehensive picture of this emergency management organization.

Naevestad (2009) suggests that a fruitful understanding of safety culture should synthesize the strengths of the interpretive approach. The main strength of the interpretive approach is that it shows how safety cultures, understood as patterns of meaning, motivate and legitimize preventive measures, are related to the identity and are the basis of ascriptions of meanings for safetyrelated phenomena. This research pursues this perspective to analyze organizational discourse through an interpretive lens.

Surveys are an excellent way to diagnose common problems but they may prevent seeing new, unique, effective, or problematic forms of communication in an organization Meyer, 2002. With surveys, the research task becomes essentially one of measurement not discovery Pacanowsky and O'Donnel-Trujillo, 1982. The second and third paper of this series will specifically examine the interpretation of meaning. The interpretive approach to safety culture conceives culture as a metaphor for the organization, and the meaning aspect of all organizational phenomena. Organizational members understand culture as shared communication patterns which create meaning. Members draw on these meanings to interpret social and task behaviors and collective identities. Interpretive scholars argue that safety culture is created and re-created through organizational members' interactions and negotiation over meaning.

Naevestad (2009) states a major challenge of safety culture research is that it runs the risk of neglecting the organizational level. The third paper of this series will specifically address this level. HRO research focuses on the organizational level and the specific organizational challenges that the technological systems of HRO induce. HRO research takes the interpretive understanding of safety culture further as it prescribes collective mindfulness as a remedy for culturally induced blindness to hazards (in this case culturally induced silence) and discusses how cultural integration, differentiation and fragmentation are required to cope with technological complexity.