Organizational Permeability to Environmental Conditions

Abstract

Organizations are expected to assess and respond to environmental conditions. For police agencies in the post-9/11 and post hurricane Katrina era, the environment includes assessing the threat posed by terrorism and disasters. We use organizational contingency and institutional theories to predict the permeability of local police chiefs’ assessments of various environmental threats and what factors affect the sensing process. We use survey data from 350 police agencies to explore the dimensionality of agency assessments on disasters, accidents, and terrorism. Our findings indicate that local police chiefs view environmental threats as having three dimensions. Additionally, institutional sovereigns have a greater influence on agency assessments of threat than do contingency factors.

Keywords

terrorism police disaster contingency theory institutional theory

Modern societies expect that institutions, such as the police, are capable of sensing their relevant environments and responding to conditions or changes in their environment. The degree to which the police sense their environment is an empirical question we seek to explore in this article by discussing recent changes in American policing in the context of the larger literature on police organizations and organizational theory. Specifically, we invoke the notion of organizational permeability to describe the capacity of an organization to sense its relevant environments. Almost all organizational theories assume that organizations can sense their environments and ideally can then respond appropriately to those environments and changes (or predicted changes) in those environments.¹ Heretofore, however, the process of how police organizations perceive their environments setting has been relatively unexplored by researchers. Additionally, there are few studies of police agency permeability to environmental events that are rare and often unpredictable. We use the threat of natural disasters, large-scale accidents, and terrorist attacks to explore the nature of organizational permeability with a sample of 350 local police agencies in the United States to explore factors that predict greater or lesser degrees of permeability.²

Policing the Unexpected: Disasters, Accidents, and Terrorism

Police agencies are expected to respond to catastrophes, including natural disasters, large-scale accidents, and terrorist attacks. Historically, the police have been mandated to provide assistance and order maintenance during and after disasters and accidents which include fires, chemical spills, explosions, floods, earthquakes, and tornadoes. Likewise, police agencies are often the first responders to terrorist attacks. Disasters, accidents, and terrorist attacks represent a special kind of environmental event for police agencies, in that they are rare, their effects are far reaching (and sometimes catastrophic), and they are usually unpredictable, or at least provide little forewarning. Organizations are designed to routinize environmental inputs. Because they are rare, often unpredictable, and sometimes catastrophic, disasters, accidents, and terror attacks (hereafter environmental threats) defy attempts at routinization and thus impede the response by police agencies.

There is little doubt that, overall, American police agencies have been responsive to changes in social conditions. In response to social unrest in the 1960s, American police agencies generally adopted the tenets of community policing during the 1970s, 1980s, and 1990s (Walker, 1993). Subsequently, in response to the September 11, 2001, terrorist attacks, local police agencies increased their focus on homeland security policing (Brown, 2007; Chappell & Gibson, 2009; Gerber, Cohen, Cannon, Patterson, & Stewart, 2005; Oliver, 2009; Ortiz, Hendricks, & Sugie, 2007; Pelfrey, 2009; Randol, 2012; Thacher, 2005). The new model of policing, one that focuses on aggressive order maintenance, militarization, and the detection of, preparation for, and response to environmental threats relies on increasing coordination and control within and between police agencies (Ortiz et al., 2007). Pundits claim, however, that maintaining a community-oriented approach is also imperative, because community members’ trust in the police may motivate citizens to provide information about terrorism (Brown, 2007; Chappell & Gibson, 2009; Ortiz et al., 2007).

Studies of police agency responses to environmental threats (disasters, accidents, and terror attacks) have generally focused on agency assessments of risk or extent of organizational change (preparedness) and generally focused on terrorism. For example, police agencies in urban areas perceive greater vulnerability to threats, especially terrorist attacks (Clarke & Chenoweth, 2006; Gerber et al., 2005; Schafer, Burruss, & Giblin, 2009). Similarly, agencies serving larger populations perceive high levels of risk to attacks (Gerber et al., 2005; Randol, 2012). Small police agencies are also vulnerable to attacks despite the belief that planned attacks occur only in metropolitan and urban environments (Giblin, Schafer, & Burruss, 2009; Haynes & Giblin, 2014; Henry, 2002; Ortiz et al., 2007; Schafer et al., 2009). Collectively, recent literature has found that small police agencies perceive lower levels of risk of terrorist attacks than large police departments (Schafer et al., 2009). Overall, numerous variables such as geographical region, population size, and agency size that affect an agency’s perceived seriousness, preparedness, and response to environmental threats.

We contend that environmental threats include a range of events, encompassing disasters, large-scale accidents, and terrorism; the three horsemen, if you will. In this regard, our analysis departs from prior studies of policing, which generally investigate police responses to only one of the three horsemen. For example, agencies may prepare for terrorist attacks that involved weapons of mass disaster, such as chemical, biological, and nuclear attacks (Fricker, Jacobson, & Davis, 2002). Terrorists may attack transit systems (Boyd & Sullivan, 2000; Wilson, Jackson, Eisman, Steinberg, & Riley, 2007), and ports and other infrastructure (Stephens, 1989), among other targets. Likewise, local agencies are expected to respond to a range of natural disasters such as fires, hurricanes (Rojek & Smith, 2007), floods, tornadoes, and ice storms. Finally, police agencies may respond to large-scale accidents, such as plane crashes, train derailments, and industrial accidents involving mines, chemical plants, and nuclear power facilities. We seek to understand how local chiefs of police view these three classes of environmental threats and to predict their responses via organizational theory.

Organizational Theory and Permeability

In policing, two organizational theories have gained prominence among academic scholars: structural contingency theory and institutional theory. Our discussion focuses on the intersection where organizations sense their environments, and what these two theories posit about this sensing process. We refer to an organization’s ability to be open to or sense its environment as permeability.

Briefly, structural contingency theory presents formal organizations as residing in a rational, task-oriented environment (Donaldson, 1995). Organizations must sense and react to this rational environment, and their reactions or adaptations may take the form of changes to agency operations and behavior. In some cases, an organization’s ability to sense or react to their task environment is constrained or facilitated by elements of organizational structure.

In the policing literature when structural contingency is tested, researchers generally explore the effects of environmental factors, such as local population demographics, on police agency structures or activities (Maguire, 2003; Wilson, 2006). Most contingency models of police agencies take the form of constraint models of organizations, wherein police agencies are “… passive entities shaped or constrained by environmental forces, such as city government (Slovak, 1986; Wilson, 1968) or population complexity and turbulence (Davenport, 1999)” (King, 2009, p. 218). In reality, it is reasonable to expect that real police managers actively sense their environment and adjust agency structure and activities in response. Therefore, researchers use passive or constraint models when testing contingency models due to limitations with their data and methods (especially the complexities in disentangling causal processes) and less because they view police agencies as truly passive.

We think permeability is an important element of structural contingency theory because it predicts that some organizations will be better able to sense their contingency environments due to different organizational structural arrangements (Sutcliffe, 1994; Sutcliffe & Huber, 1998). For example, advocates of police civilianization suggest that increasing the proportion of civilian police employees will increase the diversity of inputs and information into an agency and therefore improve an agency’s responses to its environment (King & Wilson, 2014; Kostelac, 2008; Martin, 2009). We will explore the effects of different structural attributes on police agency permeability to elements of each agency’s social environment in our analysis.

Institutional theory portrays organizations as embedded within an environment, but it is an environment composed of tradition, mythology, legitimacy, and symbolism instead of functional adaptation and tangible tasks (Meyer & Rowan, 1977). Organizations change “not as rational adaptations to their contingencies in the technical-instrumentalities mode, but by reference to normative suasion, imposition, legitimation requirements, myths, beliefs, thought ways and such non-technical instrumental factors” (Donaldson, 1995, p. 80). The institutional environment is populated by powerful, external groups and entities, such as politicians, other criminal justice organizations, community groups, the media, and public opinion writ large (Crank & Langworthy, 1992). These external constituents or sovereigns control important resources for police agencies such as prestige, legitimacy, and political influence (King, 2009).

We contend that permeability to the institutional environment is also an important element of institutional theory. It is reasonable to predict that some police agencies are particularly attuned to the desires or expectations of their sovereigns, and other agencies are less permeable. Overall, we use the notion of organizational permeability to refer to the ability of an organization to sense it technical and social environment (in the case of structural contingency theory) or its symbolic environment and sovereigns (in the case of institutional theory).

Structural contingency and institutional theories are opposing concepts within the structure of police organizations. Structural contingency theory responds to environmental concerns by altering structure or operations through rational decision making (Donaldson, 1995). The latter expresses the competing view where organizational change occurs through the influence of institutional forces including sovereigns (Crank & Langworthy, 1992), rather than a focus on effectiveness or efficiency (Donaldson, 1995). The study does not fully test the theories but rather measures components of each theory.

The purpose of this study is twofold. First, we seek to determine the nature of the seriousness of disasters, accidents, and terrorist attacks as perceived by local police chiefs located in Texas. Specifically, do local chiefs of police see all three types of events as unidimensional or multidimensional in nature? Second, we also seek to predict the environmental factors, both contingency and institutional, that help predict chiefs’ assessments of the seriousness of disasters, accidents, and terrorism. Agencies that are more permeable, either to their contingency or institutional environments, should exhibit more robust relationships between these environments and their assessments of the perceived seriousness of environmental threats. Our investigation is relevant for policy makers and researchers interested in promoting efficient responses to these events. The study detailed here is also relevant for its invocation and test of two organizational theories and the notion of permeability.

Data and Methods

Our data are derived from three sources. First, we use survey data collected from 350 local police chiefs from Texas. Police chiefs were surveyed as they attended state mandated professional leadership training between October 2011 and July 2013. Paper surveys were administered to 994 chiefs of police, 926 of whom completed the survey (a 93.16% response rate). Of the 926 completed surveys, 655 were chiefs of local (municipal, town, or city) police agencies that served populations of less than 100,000; these small- and medium-sized local agencies serve as the population of interest for the current study. Chiefs from special agencies (such as college, parks, or wildlife) and school district police agencies are excluded from the present analysis. The chiefs of the 30 largest local police agencies in Texas are not included.^3,4 We excluded agencies with fewer than five full-time employees (n = 305) because they have limited variation in many of their independent variables such as population. We also excluded surveys (n = 83) in which there was insufficient variation in the dependent variable, which was indicative of respondents checking the same response for all, or most, of the questions. This process left 305 agencies.⁵ The chiefs’ survey data provide measures of individual chiefs, such as their age, race, education, and experience in law enforcement, which we include as control variables. The survey data also provide each chief’s rating (on a scale of 0–4) of the seriousness of 30 disasters, accidents, and terrorist attacks, the dependent variable. Finally, the chief’s data provide each chief’s assessment of the relative importance of 31 institutional sovereigns, such as local media, state politicians, and county prosecutors.

The second source of data is derived from the 2008 Census of State and Local Law Enforcement Agencies (CSLLEA), collected by the Bureau of Justice Statistics (BJS) (Reaves, 2011). The BJS CSLLEA is conducted periodically in order to identify and gather information from every police agency in the United States. We use the CSLLEA data to create measures of organizational structure such as organizational size, percentage of civilian employees, task scope, interagency collaboration, and community policing.

The third source of data comes from the 2010 decennial Census and the 2012 American Communities Survey, both collected by the Bureau of Census. The Census data are used to create indicators of population heterogeneity, change, and social disorganization for the communities served by the 350 police agencies.

Overall, our data may inform about local, small-, and medium-sized police agencies and their communities in the American south. The data do not, however, address the condition of special, state, or large municipal police agencies nor do they necessarily apply to agencies in other parts of the country.

Dependent Variable

The dependent variable measures police agency assessments of threat. Police chiefs’ assessed the risk or seriousness posed to their locale and agency by 30 different environmental threats (terrorist attacks, disasters, and large-scale accidents). Chiefs were provided with a list of 30 possible threats and asked to, “Rank each of the following potential crises based on its potential seriousness to your organization if they were to occur. Please do not rank incidents by how frequently they occur.” Respondents ranked each incident on a scale ranging from 0 (least serious) to 4 (most serious).

We used MPlus software to factor analyze the chiefs’ responses to the 30 environmental threats. Factor analysis lets us to simplify our dependent variable and also determine the dimensionality of how chiefs view these threats. For instance, do police chiefs view all terrorist attacks as similar, but distinct from natural disasters and serious accidents? A visual analysis of the scree plots, in conjunction with an evaluation of the Eigen values led us to conclude that a three-factor solution was most appropriate. The results also indicated that four variables were confounding the factor analysis, so the four variables were dropped and the factor analysis rerun.⁶

The three-factor solution is presented in Table 1. The results reveal considerable sophistication in how local police chiefs view environmental threats. First, chiefs do not simply classify threats as unidimensional or as three separate clusters (disasters, accidents, and terrorist attacks). Instead, chiefs view these three groups of incidents differently. The first factor, what we term routine, is composed of incidents that are somewhat commonplace, could occur in any jurisdiction, and are relatively minor in their effects. This routine threat factor is composed of six variables: Amber Alerts, Silver Alerts, Murder/Homicide, and Responses to Fires, Freezes, and Tornados. The Cronbach’s (α = .741) indicates a modestly reliable scale.

Table 1.

Disasters Descriptive Statistics and Scale Reliability.

Constructs and variables	N	Mean	Median	SD	Cronbach’s α
Routine disasters	350	2.96	3.00	.692	.741
Amber Alert
Silver Alert
Murder/Homicide
Emergency Response to a Fire
Emergency Response to a Hard Freeze/Snow
Emergency Response to a Tornado
Location-specific disasters	350	1.90	1.69	1.10	.902
TI: Port Facilities/Large Ship Attack
TI: Threats of Damage to Dams
IA: Port Facilities/Large Ship Attack
IA: Major Fire
IA: Mining Incident
IA: Derailment of Train Transporting Chemicals
IA: Derailment of Train Transporting People
IA: Plane Crash of a Commercial Airliner
Small Boat/Water Recreation Accident
Emergency Response to a Flood
Emergency Response to Damaged Dams
Emergency Response to a Hurricane
Emergency Response to an Earthquake
Major disasters/not location specific	350	3.18	3.71	.997	.927
Active Shooting
Hostage Situation
Riot Situation
TI: Truck Bomb
TI: Chemical Dispersion Attack
TI: Biological Dispersion Attack
TI: Nuclear Attack

Note. TI = terrorist incident; IA = industrial accident.

The second factor, called location-specific threats, consists of events that are more likely to occur in some jurisdictions as compared with others. For example, because there are only four nuclear power reactors in Texas, nuclear accidents, which load on this factor, are confined to a limited number of jurisdictions in Texas.⁷ Similarly, accidents involving mines, trains, dams, and hurricanes (among others) are location specific. Some police agencies are at risk for experiencing these events but other agencies are not. The 13 variables in this factor produce a reliable scale (Cronbach’s α = .902).

The third factor, called major disasters/not location specific, is composed of seven variables that encompass threats that are viewed as very serious and may occur in any jurisdiction. Major threats include active shooting incidents, hostage and riot situations, and terror attacks involving truck bombs, and terror NBC (nuclear, biological, and chemical) attacks. These seven variables are not limited to certain locations or jurisdictions. Together, they produce a reliable scale (Cronbach’s α = .927).

Not only do chiefs of police view the overall landscape of threats as having three dimensions, they also view these dimensions differentially in terms of their seriousness. Figure 1 displays the descriptive statistics for the three threat factors. Overall, local agencies view the major disasters/not location-specific threats as the most serious, followed by routine threats. Overall, location-specific threats are viewed as the least serious of all the threats.

Figure 1.

Boxplots of each of the three types of environmental threat.

Control and Independent Variables

We include two classes of control variables in our analysis. First, we include five variables, as controls, that measure the individual characteristics of the police chiefs who completed the surveys. Gender and race are operationalized as dummy variables (0 = female, 0 = non-White). Likewise, we include an indicator of whether a chief served as a chief in another police agency before becoming the chief of their present agency (0 = no) and if a chief was selected from inside their police agency (0 = no). Finally, education is measured as a seven-category, interval-level variable. Descriptive statistics for these variables are presented in Table 2.

Table 2.

Descriptive Statistics for Control and Independent Variables.

Variables	N	Mean (%)	Median	Minimum	Maximum	SD
Chief demographics
Age	350	52.2	52.00	30	75	7.647
Gender	350	.97	1.00	0	1	.175
Female	11	(3.1)
Male	339	(96.9)
Race	348	.82	1.00	0	1	.381
Non-White	61	(17.4)
White	287	(82.2)
Years of education	347	3.48	3.00	1	7	1.678
High school diploma or GED	26	(7.4)
Some college	114	(32.6)
Associates degree	38	(10.9)
Bachelors degree	90	(25.7)
Masters degree	74	(21.1)
PhD	5	(1.4)
Another agency	349	.19	.00	0	1	.392
No	283	(80.9)
Yes	66	(18.9)
Inside/outside	347	.47	.00	0	1	.500
Inside of department	183	(52.3)
Outside of department	164	(46.9)
Organizational structure
Organizational size	350	46.65	23.0	6	1,458	94.94
Log organizational size	350	3.25	3.13	1.79	7.28	.966
Civilian employees (%)	350	.91	1.00	0	1	.293
Task scope	348	21.23	22.00	4	31	3.834
Interagency collaboration	348	1.29	1.00	0	6	1.592
COP cops (0/1)	341	.54	1.00	0	1	.499
Census
Population 2010	346	19774.10	7539.50	107	649121	43989.202
Population change (2010–2012)	346	.02	.02	−.08	.24	.038
Population less than 18 years (%)	346	26.89	27.24	10.58	38.20	4.585
Hispanic (%)	345	30.96	23.20	1.17	97.79	23.88
Median income	348	52933.25	43539.00	18884	236250	27660.638
Racial heterogeneity (Gibbs Martin D)	343	.318	.306	0	.73	.184
Sovereigns
Federal/State LE organizations	348	3.44	3.50	0	5	.988
National media	347	1.78	1.50	0	5	1.39
Local LE & EMS	348	3.94	4.00	.50	5	.796
L.E.E and community organizations	344	2.15	2.00	0	5	1.142
ME and crime lab	312	3.65	4.00	0	5	1.092
Elected representatives	347	3.51	3.66	0	5	1.09
Local media and Internet	349	3.34	3.40	.20	5	.900

Note. GED = general equivalency diploma; LE = law enforcement; EMS = emergency medical services; L.E.E. = law enforcement employees; ME = medical examiner.

The second group of control variables measures the structure of the police agency. We treat structure as control variables, as opposed to indicators of either contingency or institutional theories. Among police organizational researchers, structure is sometimes treated as an independent variable and sometimes as a dependent variable. Contingency theorist contend that structure is influenced by the task or social environment of the agency, but institutional theorists likewise contend that structure is influenced by the institutional environment of the agency. We do not wish to oversimplify this argument, as there are crucial differences between the structural adaptations predicted by contingency theory as compared with those predicted by institutional theory. However, in the interest of assessing the relationship between environment, permeability, and assessments of threat, we choose to control for the effects of structure.

Our measures of organizational structure are the size of the agency, as measured by the number of full-time employees. We use the natural log for this variable to correct for the nonnormal distribution of the data. We also include a measure of civilianization, the percentage of full-time employees who are civilian (i.e., nonsworn) employees. Task scope measures the number of possible different services performed by each agency, as a count of 39 possible tasks, such as traffic enforcement and enforcing protection orders. Interagency collaboration is measured by agency participation in any of the six possible task forces (drug trafficking, gangs, human trafficking, violent crime, and antiterrorism). Community policing was measured as a dichotomous variable; whether an agency assigned one or more police officers to community police activities (1) or not (0). We initially included a measure of agency operating budget, but this variable was too correlated with agency size to be included in the multivariate models.⁸

We also include two groups of independent variables. The first group of independent variables measures the rational and social contingency environment of each agency with Census data. The prior literature generally divides data concerning an agency’s local population into indicators of social disorganization and heterogeneity (Zhao, 1995). We follow Maguire’s (2003) technique of creating these variables. Measures of social disorganization include population, percentage of population change between 2010 and 2012, percentage of population less than 18 years of age, percentage of Hispanic, and median household income. We include one measure of racial heterogeneity, the Gibbs-Martin D, which is a single-index measure of racial heterogeneity. Greater values are indicative of greater racial heterogeneity (Maguire, 2003; Wilson, 2006).

The second group of independent variables measures the perceived importance of institutional sovereigns on local agencies. Chiefs were asked to rate, on a scale from 0 to 5, the potential importance of 31 external constituents such as state or federal law enforcement organizations, national media outlets, local law enforcement agencies, emergency medical services (EMS), employee unions, and community organizations, among others. The results were factor analyzed, a process that produced a seven-factor solution, and each factor converted to a summated index representing the average importance of each group of constituents. Our methodology and findings closely replicate those of Matusiak (2016). The seven variables measure the impactfulness of different institutional sovereigns, and we argue, are indicators of the importance of different constituents to police agencies.

Results

We seek to determine how local chiefs of police in Texas view the seriousness of disasters, accidents, and terrorist attacks. Specifically, do police view these three classes of environmental threats as unidimensional or multidimensional? Additionally, we also seek to predict the effect environmental factors, both contingency and institutional environments, have on chiefs’ assessments of the seriousness of disasters, accidents, and terrorism.

The following analysis uses three ordinary least squares (OLS) regression models, one for each of the dependent variables, and the control and independent variables are entered in four blocks. Our analysis is limited to agencies with more than five full-time employees. We assessed the possibility of multicollinearity via multiple methods. Bivariate correlations were calculated for all control and independent variables, and when the bivariate correlation between two variables exceeded .66, one of the variables was dropped from further analysis. For example, we initially included a variable that measured each chief’s tenure in law enforcement, but that variable was excessively correlated with chief’s age (.744), so we dropped the tenure variable. Additionally, all tolerance levels exceeded 0.2 (Menard, 2010), with the minimum value of .328, and the variation inflation factors (VIFs) were all less than 4, with a maximum value of 3.05. Therefore, we do not see multicollinearity problems in the three models.

The first block of control variables measures police chiefs’ individual characteristics and their relationships with the three dependent variables are reported in Table 3. The measures of individual chiefs’ attributes have little relationship with the three dependent variables. None of the chiefs’ attributes are associated with the perceived seriousness of routine disasters. Only education level is positively related to chiefs’ perceptions of location-specific and nonlocation specific disasters. The adjusted R² values in all three models are extremely modest.

Table 3.

OLS Regression: Standardized Coefficients for Three Types of Environmental Threat.

	DF1	DF2	DF3	DF1	DF2	DF3	DF1	DF2	DF3	DF1	DF2	DF3
Chief demographics
Age	.002	−.078	−.017
Gender (0 = Female)	−.071	−.012	.021
Race (1 = White)	.089	.021	.037
Years of education	.042	.199**	.165**
Another agency (0 = No)	.043	.088	.076
Inside/outside (0 = Inside)	.012	−.103	−.055
Adjusted R²	.002	.033	.018
Organizational structure
Log organization size				.044	.258***	.217***	.016	.418***	.218*	.031	.203**	.107
Civilian employees (%)				.078	−.041	−.031	.116	−.066	−.049	.027	−.047	−.047
Task scope				.154**	.076	.060	.118	.023	.090	.184**	.087	.075
IA collaboration				−.107	−.104	.085	−.096	−.091	−.083	−.146**	−.106	−.086
COP Cops				.049	.006	.005	.038	.013	−.006	.008	−.033	−.002
Adjusted R²				.032	.058	.036
Census
Population 2010							.027	−.145	.038
Population change (2010–2012)							−.070	−.053	.089
Less than 18 (%)							−.002	−.034	−.059
Hispanic (%)							−.155*	−.035	−.109
Median income							.025	−.092	−.118
Racial heterogeneity							.041	−.075	.015
Adjusted R²							.034	.064	.048
Sovereigns
Federal/State LE organizations										.010	.054	.036
National media										.154*	.164*	.184*
Local LE & EMS										.228**	−.007	−.053
L.E.E & community organizations										.122	.067	−.032
ME and crime lab										.134	.065	.111
Elected representatives										−.083	.074	.141
Local media and Internet										.040	−.106	–.039
Adjusted R²										.198	.098	.085

Note. OLS = ordinary least squares; EMS = emergency medical services; DF1 = routine disasters; DF2 = location specific; DF3 = major disasters; IA = industrial accident; LE = law enforcement; EMSemergency medical services; L.E.E. = law enforcement employees; ME = medical examiner.

p < .05. **p < .01. ***p = .00.

The second block of control variables measures organizational structure and their relationships with the three dependent variables are reported in the second row of Table 3. Only task scope was significantly (and positively) related to chiefs’ perceptions of routine disasters. Organizational size was positively related to the location-specific and major disasters/not location-specific dependent variables. None of the other organizational control variables achieved significance in any of the three models and overall the three models are modest.

The third block of variables consists indicators of police agency contingency environment, including organizational structure as control variables. The three OLS regression models show that the percentage of Hispanic population has a negative influence on chiefs’ perceptions of routine disasters. None of the measures of the contingency environment are significantly related to perceptions of location-specific disasters. Only the organizational size was a significant (and positive) predictor of perceptions of seriousness for major disasters. As was seen with the two blocks of control variables previously, the overall models for contingency theory are weak; the largest adjusted R² is .064.

The fourth block of independent variables measures chiefs’ perceptions of the impactfulness of institutional sovereigns, variables that produce the most robust models. Similar to the previous model, organizational structure was included as control variables. First, task scope, interagency collaboration, national media, and local law enforcement and EMS were significant for routine disasters. Second, organizational size and national media were significant for location-specific disasters. Last, national media was the only significant variable for major disasters. National media was the only significant variables across all three threats. This was the only variable of all models that was significant across all three environmental threats. A chief’s perception of the importance of the national media is consistently related to how seriously they rate all three types of disasters. The three models using indicators of institutional sovereigns also produce the most robust models, with the best adjusted R² as .198.

Finally, we ran three combined OLS models (one for each of the dependent variables), including only variables that were significant in the stepped OLS regressions. The first combined model reveals that impactfulness of task scope, civilian employees, percentage of Hispanic, interagency collaboration, national media, and local law enforcement and EMS all continued to be significant predictors of the seriousness of routine disasters (see Table 4).

Table 4.

OLS Regression: Standardized Coefficients for Significant Variables for Each of the Three Types of Environmental Threats.

Super Model 1	DF1
Task scope	.009*
Hispanic (%)	.001***
IA collaboration	.023*
National media	.024***
Local LE and EMS	.043***
Adjusted R²	.161
Super Model 2		DF2
Years of education		.094
Log organizational size		.149*
National media		.196***
Adjusted R²		.093
Super Model 3			DF3
Years of education			.097
Log organizational size			.104
National media			.198***
Adjusted R²			.075

Note. OLS = ordinary least squares; EMS = emergency medical services; DF1 = routine disasters; DF2 = location specific; DF3 = major disasters.

p < .05. **p < .01. ***p = .00.

The second combined model predicts chiefs’ seriousness for location-specific disasters. Again, national media is significantly related. Additionally, organizational size is positively associated with the dependent variable. The third combined model used major disasters/nonlocation-specific disasters as the dependent variable. National media was the only variable that continued to be significant. Overall, the combined models improve, very lightly, our ability to predict the seriousness of the three disaster types. The adjusted R² values indicate that the independent variables are only explaining between 16.1% and 7.5% of the variance in the dependent variables. There is still a great deal of unexplained variance.

Overall, there are two key findings. First, chiefs view disasters, accidents, and terrorism as three dimensional, not unidimensional. In addition, the chiefs have rank ordered the seriousness of the environmental threats as followed: major disasters/not location specific, routine disasters, and location-specific disasters. Second, there were 14 significant relationships between variables and perceived seriousness, with institutional sovereigns as the most significant in predicting high levels of seriousness.

Discussion

The purpose of the study is to determine how local police chiefs view the seriousness of disasters, accidents, and terrorism, as well as any presence of influencing environmental factors on the chiefs’ assessments of the levels of seriousness. Survey data from 350 police agencies were used to explore the dimensionality of agency assessments of the seriousness of the three environmental threats. Our findings indicate that local police chiefs view environmental threats as having three dimensions. Additionally, institutional sovereigns have a greater influence on agency assessments of threat than do contingency factors, organizational structure, and chief demographics.

This study has demonstrated that local police chiefs view disasters, accidents, and terrorism as a three-dimensional construct, not unidimensionally. Police chiefs classify disasters, accidents, and terrorism as belonging to one of the three categories: routine disasters, location specific, or major disasters/nonlocation specific. This finding indicates that a disparate group of events (ranging from terrorist attacks involving nuclear weapons to fires) are viewed by chiefs in terms of the police work they entail and the expected agency response.

The findings revealed that police chiefs view the seriousness of the environmental threats in the following order: major disasters/not location specific, routine disasters, and location-specific disasters. The major disasters/not location specific involve serious, disastrous, or catastrophic events. For example, this factor includes terrorist attacks with nuclear, biological, and chemical weapons, as well as active shooters, riots, and hostage situations. Without a doubt, these rare events are viewed as cataclysmic by chiefs due to their possible destruction or disruption of peace and civility and due to the sudden and unexpected nature of the events. In other words, they are almost impossible to predict in advance and the outcomes are likely extremely destructive.

Routine disasters are viewed as the second most serious cluster of threats. Many of these events are serious, yet somewhat common events for most locales. Despite their not-infrequent nature, some have forewarning (such as a hard freeze/snow and the threat of tornados), while others have less catastrophic outcomes when compared with the major disasters/not location specific. Some of these events, such as homicides, are relatively common in larger agencies but uncommon in smaller agencies. Whether rare or somewhat common, all the routine disasters are still serious, but they entail relatively routine police responses and they are events that some chiefs have had prior experience with.

The location-specific disasters are rated by chiefs as the least serious of the three disaster types. These events are differential in the likelihood they will strike a particular locale. For example, hurricanes plague the Gulf Coast of Texas, but not the inland portions. Floods are likely in some areas but not others. Terrorist attacks on specific infrastructure (such as ports and large ships) and industrial accidents are more likely in areas with these facilities than in areas without. Overall, chiefs view these location-specific disasters as least serious, in part due to the differential probability of a problem.

We posit that some agencies are more permeable to environmental conditions than other agencies. Our efforts to predict the correlates of permeability, which we operationalized as an agency’s assessment of the potential seriousness of various environmental threats, invoked contingency theory and institutional/organizational theory. Our success in predicting permeability was modest. Our findings showed 14 significant relationships between control or organizational variables and one of the three dependent variables (the environmental threat factors). Two significant variables were an individual attribute of chiefs (education), seven were organizational variables (organization size, task scope, and interagency collaboration), one was Census data (percentage of Hispanic population), and four variables were indicators of the impactfulness of institutional sovereigns which again included organizational structural variables (national media and local law enforcement and EMS). The organizational structure variables listed were not only significant alone, but significant when combined with the two sets of independent variables (Census and sovereigns). Overall, chiefs who rate the potential import of institutional sovereigns, especially the national media, are more likely to rate the three environmental threat factors as being more serious.

The significance that sovereigns had over other factors, including contingency factors, on chiefs’ perceptions of environmental threats is thought provoking. The sovereigns that had significant relationships were national media, local law enforcement and EMS, and elected representatives. There are several speculations as to why these sovereigns had more of an effect than others. First, national media was the only consistent significant factor across all three environmental threats. Major disasters are very serious events that can occur in any jurisdiction and have a lasting impact. The environmental threats classified as major disasters are more common to see on national media. Therefore, this scare may trickle down the seriousness scale to create the significant relationships between national media and location-specific and routine disasters. Second, local law enforcement and EMS are significantly related with routine disasters as those activities are more common, can occur in any jurisdiction, and are relatively minor in effects. The local police chiefs’ would most likely acknowledge the opinions of local sovereigns when assessing the level of seriousness for disasters that are connected to their job performance, such as routine disasters. Finally, elected representatives, local, state, and federal representatives, have a significant relationship with major disasters as these are very serious events that can occur in any jurisdiction and have a lasting impact. Therefore, local police chiefs heed the opinions of these sovereigns since they are related to all levels of government, including the federal representatives, and all these sovereigns would be involved in response to major disasters.

Given the unexplained variance in the statistical models, our results are suggestive. It appears that institutional sovereigns play a part in affecting the permeability of local police agencies to possible environmental threats. But the findings should be considered in light of limitations involving generalizability and the interpretation of self-reported data. The data are drawn from small- and medium-sized local police agencies in Texas. The findings may not apply to large agencies, other agency types (such as sheriffs’ offices, special agencies, etc.), or agencies with fewer than five full-time employees. Additionally, regional differences (be they cultural, climate, or population specific) may mean our findings are not generalizable to other parts of the United States.

Likewise, interpretation of the results may be a potential study limitation. For example, chiefs were provided a list of possible threats and asked to rank them based on the potential seriousness to their organization. This response is based on individual perceptions which may vary from actual seriousness. Police chiefs who have little to no experience with a threat may underreport or overreport its seriousness.

In terms of policy implications, the findings suggest effective avenues for promoting attentiveness to potential environmental threats. For example, local chiefs view the national media and EMS organizations as being most impactful. Local chiefs in this study view national media outlets as being especially impactful. The fourth estate appears to be relatively effective at prompting permeability in local police agencies. State and federal agencies that seek to diffuse best practices could leverage these networks, by advertising, editorials, and promoting the importance of best practices. Likewise, chiefs view EMS agencies as especially important to their responses to threats. EMS agencies are viewed by many local chiefs as impactful for police agency responses to threats. The federal and state governments’ attempts to build resiliency at the local level might be best targeted at local EMS organizations. In turn, local EMS agencies could serve as catalysts for change with police agencies. In other words, perhaps the most efficient path to responsiveness passes from federal and state level, through EMS providers, and then onto local police agencies.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Kadee L. Brinser is a doctoral student at the College of Criminal Justice, Sam Houston State University. She has a bachelor of science degree in criminal justice and a master of arts degree in criminal justice from the Pennsylvania State University. Her research interests are police organizations and missing persons.

William R. King is a professor and associate dean of research and program development in the College of Criminal Justice at Sam Houston State University. His research publications and research include studying police organizations from a life course perspective, the criminal investigation process, and the implementation of ballistics imaging in the United States and Caribbean.

References

Boyd, A., & Sullivan, J. P. (2000). Emergency preparedness for transit terrorism. TR News, 208, 12–17,41.

Brown

(2007) Community policing in post-September 11 America: A comment on the concept of community-oriented counterterrorism 1. Police Practice and Research 8: 239–251.

Chappell

A. T.

Gibson

S. A.

(2009) Community policing and homeland security policing: Friend or foe? Criminal Justice Policy Review 20: 326–343.

Clarke

S. E.

Chenoweth

(2006) The politics of vulnerability: Constructing local performance regimes for homeland security. Review of Policy Research 23: 95–114.

Crank

J. P.

Langworthy

R. H.

(1992) An institutional perspective on policing. The Journal of Criminal Law and Criminology 83: 338–363.

Davenport

(1999) Environmental constraints and organizational outcomes: Modeling communities of municipal police departments. Police Quarterly 2: 174–200.

Donaldson

(1995) American anti-management theories of organization: A critique of paradigm proliferation Vol 25, Cambridge, MA: Cambridge University Press.

Falcone

Wells

L. E.

(1995) The county sheriff as a distinctive policing modality. American Journal of Police 14(3/4): 123–149.

Fricker

R. D.

Jacobson

J. O.

Davis

L. M.

(2002) Measuring and evaluating local preparedness for a chemical or biological terrorist attack, Santa Monica, CA: Rand.

10.

Gerber

B. J.

Cohen

D. B.

Cannon

Patterson

Stewart

(2005) On the front line American cities and the challenge of homeland security preparedness. Urban Affairs Review 41: 182–210.

11.

Giblin

M. J.

Schafer

J. A.

Burruss

G. W.

(2009) Homeland security in the heartland risk, preparedness, and organizational capacity. Criminal Justice Policy Review 20: 274–289.

12.

Hannan

M. T.

Freeman

(1989) Organizational ecology, Cambridge, MA: Harvard University Press.

13.

Haynes

M. R.

Giblin

M. J.

(2014) Homeland security risk and preparedness in police agencies: The insignificance of actual risk factors. Police Quarterly 17: 30–53.

14.

Henry

V. E.

(2002) The need for a coordinated and strategic local police approach to terrorism: A practitioner’s perspective. Police Practice and Research 3: 319–336.

15.

Jenness

Grattet

(2005) The law-in-between: The effects of organizational perviousness on the policing of hate crime. Social Problems 52: 337–359.

16.

King

W. R.

(1999) Time, constancy, and change in American municipal police organizations. Police Quarterly 2: 338–364.

17.

King

W. R.

(2009) Toward a life-course perspective of police organizations. Journal of Research in Crime and Delinquency 46: 213–244.

18.

King

W. R.

Wilson

J. M.

(2014) Integrating civilian staff into police agencies, Washington, DC: Office of Community Oriented Policing Services.

19.

Kostelac, C. A. (2008). The changing face of police organizations: Trends in civilianization (Unpublished doctoral dissertation). Phoenix: Arizona State University.

20.

Maguire

E. R.

(2003) Organizational structure in American police agencies: Context, complexity, and control, New York, NY: SUNY Press.

21.

Martin

(2009) Leadership from the outside: Civilianizing command structure in modern law enforcement. Journal of California Law Enforcement 43: 19–24.

22.

Matusiak, M. C. (2016). Dimensionality of local police chiefs’ institutional sovereigns. Policing and Society: An International Journal of Research and Policy. 7, 753–770..

23.

Menard

(2010) Logistic regression: From introductory to advanced concepts and applications, Thousand Oaks, CA: Sage.

24.

Meyer

J. W.

Rowan

(1977) Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology 83: 340–363.

25.

Oliver

W. M.

(2009) Policing for homeland security policy & research. Criminal Justice Policy Review 20: 253–260.

26.

Ortiz

C. W.

Hendricks

N. J.

Sugie

N. F.

(2007) Policing terrorism: The response of local police agencies to homeland security concerns. Criminal Justice Studies 20: 91–109.

27.

Pelfrey

W. V.

(2009) An exploratory study of local homeland security preparedness: Findings and implications for future assessments. Criminal Justice Policy Review 20: 261–273.

28.

Randol

B. M.

(2012) The organizational correlates of terrorism response preparedness in local police departments. Criminal Justice Policy Review 23: 304–326.

29.

Reaves, B. A. (2011). Census of State and Local Law Enforcement Agencies. United States Department of Justice. Office of Justice Programs. Bureau of Justice Statistics.

30.

Rojek

Smith

M. R.

(2007) Law enforcement lessons learned from Hurricane Katrina. Review of Policy Research 24: 589–608.

31.

Schafer

J. A.

Burruss

G. W.

Giblin

M. J.

(2009) Measuring homeland security innovation in small municipal agencies policing in a post-9/11 world. Police Quarterly 12: 263–288.

32.

Slovak

J. S.

(1986) Styles of urban policing: Organization, environment, and police styles in selected American cities, New York: New York University Press.

33.

Stephens

H. W.

(1989) Barriers to port security. Journal of Security Administration 12: 29–41.

34.

Stinchcombe

(1965) Social structure and organizations. In: March

J. G.

(ed.) Handbook of organizations, Chicago, IL: Rand McNally & Co, pp. 142–193.

35.

Sutcliffe

K. M.

(1994) What executives notice: Accurate perceptions in top management teams. Academy of Management Journal 37: 1360–1378.

36.

Sutcliffe

K. M.

Huber

G. P.

(1998) Firm and industry as determinants of executive perceptions of the environment. Strategic Management Journal 19: 793–807.

37.

Thacher

(2005) The local role in homeland security. Law & Society Review 39: 635–676.

38.

Walker

(1993) Does anyone remember team policing? Lessons of the team policing experience for community policing. American Journal of Police 12: 33–56.

39.

Wilson

J. Q.

(1968) Varieties of police behavior: The management of law and order in eight communities, Cambridge, MA: Harvard University Press.

40.

Wilson

J. M.

(2006) Community policing in America, New York, NY: Routledge.

41.

Wilson

J. M.

Jackson

B. A.

Eisman

Steinberg

Riley

K. J.

(2007) Securing America’s passenger-rail systems, Santa Monica, CA: Rand.

42.

Zhao

(1995) Contemporary organizational change in community-oriented policing: A contingency approach (Doctoral dissertation), Pullman: Washington State University.