Google

Using learning as a construct to measure civilian review board impact on the police: The Philippine experience

Melchor C de Guzman, James Frank. Policing. Bradford: 2004. Vol. 27, Iss. 2; pg. 166

Abstract (Summary)
Determining the impact of civilian review board on the police is a challenging process. The task is complicated due to the absence of baseline data that will account for observed changes in citizen complaints, especially if the concept is a novelty in a particular jurisdiction. Likewise, using traditional measures of impact such as the number of complaints or conviction rates is problematic due to a variety of confounding factors. This study examines the perceptions of complainants and officers concerning the impact of civilian review boards. Using data collected through surveys of police officers and complainants in a metropolitan area in the Philippines, the study focuses on “learning” as a viable construct to measure the impact of civilian review boards and the perceived deterrent effects of these boards. The research found that civilian review boards have a significant impact on police officer perceptions as well as on the police department. The study also shows that learning may be a viable measure for studying the impact of civilian review boards.

Introduction
In democratic societies, controlling the abuse of powers of the police (Smit and Botha, 1990) and controlling the behavior of police officers have been daunting tasks for police managers (Crank and Caldero, 2001). A manifestation of this concern is the multiplicity of mechanisms that simultaneously attempt to control the police. The result has been that the police are “perhaps the most scrutinized occupation – after the Presidency of the United States” (Crank and Caldero, 2001, pp. 21-22). Despite the abundance of different checks on the power of the police, civilian review boards are being advocated by the public and police scholars as agencies for handling citizen complaints, as control mechanisms for police misconduct, and as a means for holding the police accountable to the public (American Civil Liberties Union, 1991; Armstrong and Wood, 1991; Carillo, 1991; Walker, 2001).

Further, the popularity of civilian review boards is related to the community policing movement and its emphasis on police accountability. Integral to this movement is the expanded role of the public and the development of police-community partnerships (Bayley, 1994; Frank et al, 1996; Grinc, 1994; Kelling, 1995; Kerley and Benson, 2000; Maguire and Mastrofski, 2000). Civilian review of complaints involving police misconduct provides a mechanism for consumers of police services to be involved in the oversight of decisions by officers and agencies (Walker, 2001). Through these decisions, civilians may influence future police agency policies and officer behavior on the street.

The Philippines recently instituted a control mechanism that is synonymous to civilian review called the People’s Law Enforcement Board (PLEB). The PLEB is a civilian board that receives, investigates, adjudicates, and decides cases involving members of the Philippine National Police, the national police force of the Philippines. The jurisdiction of each PLEB is local and there is at least one PLEB for each municipality, although the larger cities (i.e. Manila and Quezon) have one PLEB for each legislative district.

Though civilian review boards have become quite popular in both United States and Philippines, only a limited number of systematic evaluations have been conducted of review board operations. This is especially true with assessments of the impact of review boards on police officer behavior and agency policies (de Guzman, 2001). This shortcoming may not necessarily be due to a lack of interest about the subject. Instead, it may more likely be the result of data and methodological difficulties involved in assessing the impact of review boards (Walker, 2001).

More specifically, in a recent review of the US civilian review boards, Walker (2001) identified the shortcomings and pitfalls of traditional methods for assessing the performance of civilian review boards. He noted the complexities, as well as the potential inadequacies, of evaluations that relied solely on official performance indicators such as sustain rates and the number of complaints filed. In particular, he argued that researchers using official data had difficulty with causal interpretations because one could hardly attribute observed changes in police behavior to the existence and performance of the board without other data. Instead, Walker (2001) suggested that Perez’s (1978) methodology using “learning” as a construct for measuring the impact of civilian review boards was an innovative attempt at measuring this elusive construct.

This research endeavors to utilize “learning” as a construct and attempts to introduce a methodology for operationalizing this construct. In the end, it is hoped that this study will shed light on alternative measures for assessing the impact of civilian review boards. In other words, we present a construct that has the potential for more validly assessing the utility of civilian review boards as mechanisms for controlling and changing police misbehavior.

Problems interpreting official data
As noted by Walker and Bumphus (1992, p. 10), “the interplay of complaint procedures and the volume of and rate of complaints is, in short, highly complex and complicates any attempt at evaluation.” For instance, an increase in the number of complaints may be susceptible to a number of contradictory explanations. First, it may be interpreted as an indicator that the board is inspiring public confidence in the complaint procedure and individuals are coming forward to report alleged instances of improper police behavior, even though the actual number of complaints has remained stable. second, an increase in the number of filed complaints may also merely be a manifestation of the heightened mood for litigation by the public. Third, a contrary interpretation may be inferred from the increase of complaints. The antithesis that may be advanced is that the board may be ineffective in deterring police officer misconduct, consequently, resulting in a substantial number of incidents of police misbehavior and a greater number of complaints against the police.

Likewise, a decrease in the number or rate of complaints may also give rise to plausible contradictory interpretations. The observed reduction in the filings of complaints may result when complainants are merely apathetic, apprehensive, fatalistic, public spirited, or simply unaware (Russell, 1978; Wagner and Decker, 1993). Thus, the actual number of misconduct instances may be increasing while the public is unwilling to file complaints. In contrast, the decrease in complaints may occur because of an actual reduction in instances of improper police behavior. In other words, for every argument that board actions have a beneficial effect on police misbehavior, there is an equally plausible argument that board actions have either an equally damaging effect or no effect.

Theory and concepts
The adoption of a valid measure of impact is always a challenging task (Rossi and Freeman, 1989). In most existing evaluations of review boards, official data are utilized to assess impact. With respect to civilian review boards, however, Walker and Bumphus (1992) and, more recently, Walker (2001) contend that official data are inadequate measures of the impact of civilian review boards because of problems specifying causal relationships if there is an increase or decrease in the number of complaints, the incidences of reported police misconduct, or official disciplinary acts (resolution rates and patterns).

The aforementioned problems with official data were the catalyst for Perez’s (1978) [1] development of “learning” as a construct that may be used as an alternative measure of the board’s impact on the police. He conceptualized learning as the extent to which police departments and officers are influenced by the operations of the board. In other words, police “learn” what is proper behavior from decisions of the boards and this learning impacts on their actions when they interact with citizens. In addition, the mere existence of review boards may influence officers and agencies because they “learn” that their actions will be reviewed and they will be held accountable for their behavior. Learning may be manifested through changes in department policies, police officer behavior, and levels of supervision each of which may be undertaken to ensure conformity with board decisions. In addition, officer perceptions of the perceived deterrent effect of civilian review on police misconduct may indicate a belief that the police have learned from board actions.

Learning and deterrence theories, both derived from criminology, lend support to the use of “learning” as a construct. As applied to the present study, if officers believe their decisions to engage in certain behaviors are influenced by the existence and prior decisions of review boards, and they think that these boards influence agency decisions to implement specific policies, then the street-level behavior of officers is likely to be influenced. Then, perceptions will provide a measure of learning that may be used to assess impact. Admittedly, such perceptions may not perfectly reflect the reality of PLEB operations or any civilian review board. Nevertheless, these perceptions are often times the background serves as the catalyst for the demise of civilian review boards in democratic societies. Besides, civilian review boards, as with other government bureaucracies, exist not because of their “actual” utility, but because of their “perceived” utility in society. As such, it is possible and viable to use learning as a proxy for measuring the impact of civilian review boards. We argue that the concept of learning may be a more valid and reliable indicator of review board impact than inferences derived from official data.

Extant research on review board impact
In a series of studies, Perez (1978; 1992; 1994) concentrated on the behavioral aspect of the civilian review boards. He concluded that civilian review has only limited impact, if any, on the behavior of police officers. In fact, he commented that “behavioral impact is the Waterloo of civilian review” (Perez, 1992, p. 7). Perez (1994, p. 159) reported that of the Berkeley police officers he surveyed in 1978, “12 percent responded that they felt the PRC’s operation had a positive effect on behavior on the street; 37 percent of the group felt that the PRC had no impact; and 38 percent believed it to be counterproductive (12 percent did not know about its impact)”. However, he qualified his findings by stating that structural factors and other qualities of the board members may be explaining this shortcoming of the boards in deterring police officers. Perez (1994, p. 162) reasoned that:

.. .learning does not develop for essentially defensive reasons because review systems are too lenient. Learning is also limited because civilian systems are seen to be run by and for people who know nothing about police work and therefore, need not be listened to. And finally because they are formalized and removed from the individual police officer’s street experience, civilian systems may limit and even inhibit informal learning channels from changing police behavior patterns.

Unfortunately, Perez (1978, p. 1994) only examined the perceived relationship between board decisions and officer behavior, and not the influence of civilian review on policies, supervision, and deterrence.

Studies of the impact of civilian review boards on officer behavior across various jurisdictions have produced inconsistent findings. In some instances, the existence and decisions of civilian boards constrained officer actions and produced detrimental effects, while in other jurisdictions the effects are more positive. For example, the negative effect of the board was manifested in Philadelphia where police officers exhibited a degree of reluctance to make arrests during a riot because of the board’s presence (Halpern, 1974) [2]. This is precisely the type of dysfunctional behavior that police officers and opponents of civilian review predict will happen with the presence of an external review board.

In contrast, other research has suggested that civilian review boards may have a positive impact on the police officer’s learning experience. For example, Terrill (1988) contended that Detroit’s Board of Police Commissioners (BPC) was an example where civilian oversight was one of the significant factors that contributed to improvements in police conduct. His findings confirmed those of Littlejohn (1981) who earlier found that the BPC produced a deterrent effect on officer misconduct in Detroit [3].

Spencer Coxe, the Executive Director of the Philadelphia Branch of the American Civil Liberties Union (ACLU), believed that the board in Philadelphia “contributed to the enlightenment of members of the force as to the limits of their authority, and has helped reveal and correct objectionable departmental practices and policies” (Coxe cited in Terrill, 1988, p. 84). Gellhorn (1966, p. 9) suggested that complaints could serve as feedback on the appropriateness of police conduct. He said that “police officers learned that they need fresh guidance (underscoring ours) on when or when not to use handcuffs at the time of arrest”, so as to not aggravate an existing injury or to prevent causing an injury to the offender. Existing research has shown that police behavior may be influenced by policies of the department (Fyfe, 1985; Sherman, 1992; Wilson, 1968). Thus, if decisions of civilian review boards influence department policies, these policies, in turn, may produce significant changes in the behavior of individual officers. Policy changes maybe, therefore, a critical evaluation measure when assessing the impact of civilian review boards.

With respect to the deterrent effects of civilian review boards, Halpern (1974) noted that the boards have a dysfunctional effect on the police. He claimed that the lack of initiative on the part of the police officers was due to their fears of becoming a party to a civilian review board action. However, he failed to note that such reluctance might be an indication of a deterrent effect produced by the board on police behavior. Some observers might interpret such reluctance as acts of temerity. Others could view the same facts as circumspection – an indication that police officers are more careful due to the operations and presence of the board. In the latter situation, review boards have a beneficial impact.

Without additional data, it cannot be definitively stated that the absence of misconduct is a direct function of a deterrent stimulus as there are a host of factors that can be responsible for an individual not engaging in a deviant act or misconduct (Brown et al, 2001). However, debriefing respondents about why they failed to act may provide the necessary information to assess more conclusively whether the presence of the review board had a deterrent effect on their decisions to forego misconduct. In any case, if officers believe they were deterred, this information more directly addresses the impact of civilian review than inferring such from the official data.

Method
The present study employed a survey methodology and measured the PLEB clients, both officers and civilians, in the Philippines regarding the impact of the PLEBs on police officers and agencies. While numerous reservations had been voiced about the use of surveys and respondent perceptions, other data collection procedures were not possible because of time and financial constraints, and would not allow for the collection of data on a substantial number of cases. For instance, data would need to be collected concerning the history and reasons why department policies were enacted to see if they were implemented in response to review board actions. Officers would need to be debriefed using protocol analysis on a regular basis to determine whether the existence of review boards influenced their daily street-level decisions. The costs associated with these types of data collection processes would be prohibitive.

At the same time, using beliefs and perceptions to measure impact are not without their detractors. The underlying concerns are that respondents may forget, filter, or misinterpret events and they may provide socially desirable responses to questions. In other words, they may reconstruct their own reality concerning an issue about which they are being questioned. However, a substantial body of research in the field of both policing and criminology indicates that perceptions are important determinants of actions (Gove et al, 1985; Paternoster, 1987; Sherman, 1978; 1983; 1990; 1992; Sherman and Weisburd, 1995) as people’s act is based on the information that is at their disposal. Manning (1978) commented that in the field of policing, perceptions and manipulation of appearances are important factors that often times become the basis for action.

Study site
The National Capital Region of the Philippines is the site of the study. The region is composed of 11 cities and six municipalities. The country’s capital, Manila, is located in the region and has a population of approximately two million people, while the entire region has a population close to eight million. It is a culturally diverse population as most of the inhabitants have migrated to the area from provinces throughout the Philippines. The National Capital Region is the banking, financial, industrial and commercial center of the country. Therefore, it is similar to large urban population centers where the police are required to interact with culturally diverse populations residing in communities that range from very affluent to economically distressed.

The Philippines has a national police force. The study site is policed by the National Capital Region Police Office (NCRPO) and is divided into five districts. Manila and Quezon, the two major cities in the region, account for two of the districts while the remaining cities and municipalities within the region form the other three districts. While a director is the chief officer in charge of all five districts, each city and municipality has a police station managed by a chief of police. Police chiefs in these city and municipal police stations implement policies that are developed at the national police headquarters [4].

Philippine law (section 27, Republic Act 6975) mandates that each legislative district has a PLEB office. Within the National Capital Region, there are 28 PLEB jurisdictions. Manila has six different PLEB offices since it is divided into six legislative districts. Likewise, Quezon City has five PLEB offices and Caloocan City has two. The remaining cities and municipalities have one board each. As a rule, complaints are to be filed in the city or municipality where the officer is assigned (Rule IV, sec. 1, Memorandum Circular No. 91002). If an officer is assigned to an area outside of a city or municipality (i.e. he or she is assigned in the provincial, regional or national offices), the complaint shall be filed in the PLEB office where the act took place (Rule III, sec. 3, Memorandum Circular No. 91002).

Survey instrument
Three survey instruments were prepared – one for police officers who have been the subject of complaints, another survey for police officers who have never been a party to a complaint, and another for complainants. The questionnaires consisted of two parts. The first part of each questionnaire collected demographic information common to all respondents such as the respondents’ age, gender, marital status, level of education, and monthly household income. For the police officer samples, questions were also asked about the officers’ rank, while complainants were asked about their employment status during the period when their complaint was being processed.

Expanding on the work of Perez (1994), the second portion of the survey instrument contained three items aimed at soliciting responses concerning the perceived impact of review boards on police officer and agency behavior. The first question was a dichotomized measure (yes/no) asking respondents about perceived policy changes in the department that may have resulted because of the existence of the PLEB. Specifically, respondents were asked to indicate whether they noticed policy changes in their departments due to the existence or operations of the PLEB. The second question asked respondents if they have observed changes in the behavior of police officers due to the existence and operation of the PLEB. The response options were “no change”, “a positive change” or a “negative change”. The third question requested respondents to indicate whether they noticed changes in the extent of supervision of officers due to PLEB operations. A dichotomized response option of “yes” or “no” was provided for this last question.

In addition, police officers were asked to respond to three statements about the deterrent effects of the PLEBs. A four-point Likert-scale response option was provided with each statement. The response options were “strongly agree,” “agree,” “disagree,” and “strongly disagree”. The three statements were:

(1) when I stop a citizen, one of the things that go through my mind is the potential of receiving a complaint before the PLEB,

(2) the potential of being prosecuted before the PLEB makes police more careful when dealing with citizens, and

(3) the potential of receiving complaints before the PLEB stops police from violating an individual’s rights.

The surveys were administered to the respondents in various forms. For complainants, the questionnaire was administered either through a structured interview or through a self-administered survey, depending on the preference of the respondent. This dual method was necessary to accommodate those respondents (primarily complainants) with difficulties reading or understanding the survey items. The interview method also allowed for explanations of specific questionnaire items and for surveys to be translated in the respondent’s native language. For police officers, the questionnaires were self-administered.

Before administering the police officer surveys, the prepared questionnaires were pre-tested with ten police officers to identify any problematic survey items. These individuals were provided the opportunity to make comments concerning the questionnaire particularly as to the language and content of the questions. Since these ten officers did not voice any significant concerns, the questionnaires were reproduced with slight modifications for distribution to the targeted respondents.

Selection of study samples
Three samples were selected for the study. The first sample consisted of complainants who instituted actions before a PLEB in the National Capital Region. The second sample consisted of police officers who had complaints filed against them and had their cases processed by the board from 1991 to 2000. The third group consisted of police officers who have never been a party to a complaint filed before a PLEB and are assigned to one of the various police stations of the National Capital Region. The details of the sample selection process for each group are discussed below.

Complainants. A list of complainants with their addresses was requested from the local PLEB offices through the National Police Commission (NAPOLCOM). The list provided to the first author covered the years 1991-2000 and included the names and addresses of the complainant, the names and ranks of the involved officers, the type of complaint filed, the date the complaint was filed, and the resolution of the case for all cases handled in the National Capital Region during this period. This list was further categorized by PLEB jurisdiction thereby facilitating the stratified selection of respondents. Employing the aid of trained researchers [5], complainants were traced using the lists provided by the National Police Commission.

In order to select the study sample, the complainant list was first divided into the five districts within the region and researchers were assigned to cover specific jurisdictions within each police district. Data collection personnel were instructed to select every tenth name on their list starting from the top of the list. If the chosen respondent could not be traced, the researchers were instructed to select the next tenth respondent. This procedure was carried out until 150 participants per district were selected for a total sample of 750 complainants.

After obtaining the endorsement from the local political official [6], researchers personally contacted the complainants. A letter was presented to the complainants requesting their participation in the study. If necessary, the content of the letter was explained to the respondents orally and in the vernacular. The letter and the oral explanation emphasized that the study was independent from any government agency, that participation was voluntary, and that all responses would be kept confidential. Some of the respondents requested that the questionnaires be retrieved later, and the researchers complied with the requests. Data collection personnel were able to contact 692 (92.3 percent) of the targeted survey respondents and 515 of these individuals completed surveys in sufficient detail to be of value to the study, leaving a useable response rate of 68.7 percent.

Police officers with cases. As noted earlier, officer names were provided on the complainant list. This list was used to select police respondents involved in cases. Two hundred police officers per district (or a total of 1,000 officers) were randomly selected using the same method that was used to select the complainants. Officers were tracked through the chiefs of police in each district after a letter requesting permission to conduct the survey was presented to the chief of police in the area where the officer worked. Unfortunately, it was possible to locate only 289 (28.9 percent) of the officers.

This low response rate resulted due to a variety of unexpected reasons. A substantial number of officers were either transferred to another jurisdiction outside of the region, retired from police work, died or were physically and mentally incapable of responding to the survey. It must be remembered that the list consisted of officers and cases that were possibly 10 years old. It is, therefore, highly likely that officers who received complaints were not available for the study [7]. Also, due to political turmoil during the administration of the survey, many of the first author’s police contacts were removed as chiefs of police in the study site. The result was that police cooperation was not at the level expected.

The first author personally distributed the questionnaires to the selected officers and explained the purpose and mechanics of the survey. It was explained to the officers that the research will be treated confidentially and that the researchers are not in any way connected with any government agency. It was also explained that the survey instrument contained no identifiers that will link them to the sampling list. Out of the 289 respondents, 206 (71.2 percent) completed the survey instrument. Respondents (n = 83) who failed to respond to the critical aspects of the survey were excluded.

Police officers without cases. The second sample of police officers consisted of individuals who were never a party to a complaint before the PLEB. These respondents were randomly chosen from the roster of each district. One thousand officers were targeted, stratified by district. Thus, 200 questionnaires were distributed per district. Six hundred and thirty-five (635) surveys were retrieved representing a response rate of 63.5 percent. Five hundred and fourteen of the surveys were completed in sufficient detail to be useable for a final response rate of 51.4 percent. The problems encountered with the officers with cases sample were not evident with this sample because the initial sampling lists only contained the names of active officers. The survey administration procedure used with officers with cases was followed with these officers.

Sample characteristics
Table I gives the demographic characteristics of the three groups of respondents. Age and average monthly income were collected as continuous data, but were categorized after examining the distributions of the raw data. The raw data for income reflects the household income per month in Philippine pesos. These amounts were transformed into US dollars using the current exchange rate of 50 pesos for every dollar. The categories in the table reflect the economic situation in the Philippines. Individuals earning below 400 dollars are the upper-lower class, those earning more than 400 but not more than 800 are the middle-class, those earning more than 800 but less than 1,200 are the upper-middle class, and those earning more than 1,200 are the upper class. The findings for the other variables in the table reflect the actual response categories used in the survey.

Using learning1
Table I.
Respondents’ demographic information

Complainants. A majority of the complainants are male (67 percent) and their mean age is 33.56 years. Almost 70 percent (69.4 percent) are employed and have a mean monthly household income of $341.39. Over half of the complainants are at least college graduates (58.4 percent) while a few of the complainants hold graduate degrees (0.6 percent). Most of the complainants are either married (49.3 percent) or single (41.6 percent).

Police with cases. Police who had cases with the PLEB are predominantly male (94.8 percent). They are the oldest among the three samples with a mean age of 39 (mean = 39.55). Almost half of these officers (49.5 percent) are between the ages of 46-55 and an additional 27.5 percent are 25-35. A slight majority (50.9 percent) of these police officers would be considered middle class earning a monthly household income of $400-800. At the same time, almost 40 percent (39.8 percent) of these officers are in the upper-lower class with a monthly household income of less than $400. Over three-fourths of the officers with cases (77.1 percent) are college graduates and an additional 7.8 percent hold graduate degrees. As regards their civil status, 90.2 percent are married.

Police without cases. With a mean age of 35 (mean = 35.29), police without cases are relatively younger than the police with cases. Most of the respondents from this group are within the age of 25-35. They are predominantly male (83.5 percent) and the modal income category for this group is between $400 and $800 per month (42.7 percent). Compared to police with cases, more of these respondents have income that is greater than $1,200 (19.7 percent) per month. Most of these sample respondents are college graduates (84.2 percent) with almost 10 percent possessing graduate degrees (9.5 percent). Finally, close to 70 percent are married (69.7 percent).

Results
Frequencies and proportions were used to summarize the data on the different variables. Means, ranges, and standard deviations were reported for the continuous data in the study. Inferential statistics were used to determine the relationships of various variables in the study. The data collected on the learning variables were analyzed using chi-square and t-test statistics with a < 0.05 level of significance as the critical region.

Perceived PLEB impact
Table II displays the responses of all three samples to the three survey items that examined the perceived PLEB impact on the police. Several patterns appear across the responses to the survey items. First, for each item police officers with cases perceived the greatest positive impact from the PLEB actions, followed by the officers without cases. Also for each item (supervision, attitude change, and policy change) complainants were least likely to believe that the review board had positively impacted officer and agency behaviors. second, the greatest proportion of respondents for each sample voiced perceptions indicating positive change for each learning measure. While this is most evident with the two police samples, complainants tended to split their responses among the available response options. Third, the findings in Table II indicate that there are significant differences in the responses across the three groups.

Using learning2
Table II.
Frequencies and percentages of the respondents’ perceived impact of the board on the police and chi-square tests of significance

More specifically, the findings show that a majority (54.6 percent) of the complainants perceived that stricter supervision of the police resulted because of the operations of the PLEB. Also, a majority of each of the police samples – police with cases (67.8 percent) and police without cases (51.8 percent) – believed that strict supervision of the police occurred due to the PLEB actions. The police respondents (police with cases, 85.9 percent and police without cases, 64.6 percent) also saw positive changes in the attitudes of police officers and believed that certain policy changes have been enacted due to the operations of the PLEB (police with cases, 85.9 percent and police without cases, 67.8 percent). In contrast, a majority of complainant responses indicated that there has been either no change (38.2 percent) or a negative change (21.1 percent) in police attitudes. Similarly, less than half (43.1 percent) of these complainants believe that PLEB actions have resulted in policy changes.

Chi-square statistics were used to determine if there were differences across the groups concerning their perceptions of the impact of the civilian review boards. The findings reported in Table II indicate that there are significant differences among the groups on each of the three measures of learning. However, an examination of the raw frequencies indicated that there were marked differences in the responses of police with cases on all of the measures and, thus, one might question whether the significant difference detected among the three groups was accounted for by the police officers with cases sample. To check for this possibility, three analyses were performed comparing the responses of each sample to one another. First, the responses of officers with and without cases were compared and the chi-square scores were statistically significant for each of the three survey measures. second, the responses of police with cases and the complainants were examined and the differences between the two groups on all measures were again statistically significant. Finally, the responses of police without cases and the complainants were compared on all three learning measures. These results were mixed in that the supervision and attitude change measures were not significantly different, while perceptions of policy changes were significantly different with complainants being more likely than police without cases to perceive that the PLEB had no impact on department policies. The general finding, however, is that police with cases have significantly different perceptions from all the other two groups.

Deterrent effect of PLEB

Three survey items asked police respondents about their perceptions of the effect of PLEBs on their street-level behavior. The data in Table III show that police officers believe their behaviors are influenced by the operations of the review boards. The descriptive data indicate that a relatively large percentage of the police respondents, both with and without cases, tend to consider the repercussions of their actions when they deal with citizens.

Specifically, for the first two measures reported in Table III (the potential of receiving a case before the PLEB and the potential of being prosecuted), police with cases tend to perceive greater deterrent effects of PLEBs than do police without cases. Not only were police with cases were more likely to agree with the survey statements than were police without cases, but their responses were also significantly different from one another. In contrast, a larger proportion of police without cases than police with cases believe that review boards stopped police from violating the rights of citizens (72-55.1 percent). With this survey item the police officers’ perceptions, however, were not significantly different (p = 0.861) from one another.

Using learning3
Table III.
Police respondents’ perceived deterrent effects of the PLEB on officers (police with cases and without cases)

Discussion
Overall, the study findings indicate that the Philippine review boards are perceived as having an effect on officer behavior as indicated by the learning that the police derive from the operations of the PLEB. Contrary to what Ferez (1992) found in his study of the California Review Boards, the PLEB are believed to exert a considerable degree of influence over department policies and officer behavior. However, it should be noted that the observed differences in the study findings and those reported earlier by Perez (1992) might result because of the differences in the structure and power of the two review systems. In as much as the PLEB is a completely civilian board with investigative, adjudicative, and punitive powers, PLEB features may be significantly different from the boards considered by Perez. The influence of these board features on the perceptions of their clients remains an empirical question.

The present findings also suggest that complainants have a more reserved or conservative evaluation of the impact of the PLEB. Complainants were less likely than both groups of officers to report positive changes in police policies and attitudes across all of the measures. There are several reasons why this may have been observed. First, complainants’ perceptions may be due simply to a lack of information concerning police policy changes. They are not likely to be privy to every policy implemented by the police on account of PLEB operations. Thus, they tend to perceive less of a PLEB impact. second, and related, changes in police policies and behaviors that result due to the actions of the PLEB may not be noticeable unless the changes are drastic or publicized. Third, complainants’ evaluations of PLEB impact may be diminished by the failure of the review boards to sustain their allegations, thus their negative case outcome may be causing the observed perceptions. A prior analysis using the same data indicated that case outcome was positively related to complainant attitudes toward the police generally (de Guzman, 2001). This overall attitude toward the police may merely be extended to the PLEB process.

Despite these possible explanations, complainant perceptions about the board’s lack of impact should not be dismissed. Complainants are consumers of police services. Departments concerned about being accountable to their constituents should consider their perceptions. In addition, police interact with the public on a regular basis and improving public perceptions may positively influence the ability of the police to perform their tasks and may help foster a more cooperative relationship with the public.

Both samples of police respondents indicate that the PLEB has a positive impact on officer and behavior actions, though police with cases voiced more positive attitudes. Police with cases tend to perceive PLEB as deterring police misconduct and at the same time having a significant impact on police behavior. A smaller proportion of police without cases also see positive changes, though they voice positive responses overall. Perhaps, this phenomenon can be explained by the fact that police officers who have never had cases before the boards are unfamiliar with the process. As mentioned above, PLEB decisions usually find that the police acted properly and therefore, officers with cases may see the boards as performing an important police function. In addition, as Paternoster (1987) explained, the deterrent effect of punishment is usually diminished by lack of experience with being caught as well as being punished swiftly and severely. Similarly, perceptions of review board effectiveness as far as the police are concerned, may depend on officer interactions with the civilian review process.

In any case, the perception that PLEBs matter as far as police policy making and officer behavior are concerned appears to contradict the common impression that civilian review boards have limited impact on police activity (Perez, 1992). Whether PLEB decisions actually impact policies or not remains an empirical question, though this study finds that officers believe review board decisions do have an impact.

The assessment of learning that occurs due to the operations of the PLEB may be useful for police departments. The study findings may assist police managers in planning and implementing police training and supervision initiatives in response to problems of police behavior. Schwartz (1985) and Kerstetter (1985) share the belief that civilian review board operations may efficiently disclose problems pertaining to policies, especially those that relate to police work in the streets (Fyfe, 1985). If disclosed problems are acted upon, then review boards will be seen as a positive mechanism for controlling the police and holding them accountable to the public.

In conclusion, learning is a viable and important concept that may be used to evaluate the impact of civilian review boards. The generalizations from this study must, however, be taken with some caution. It must be remembered that the civilian review boards considered under the present study are, to a large extent, independent from the police and the boards possess full investigative, adjudicative, and punitive powers. The United States has several variations of the civilian review boards that the findings in the study may not be necessarily be true for any or all of these variations. The next step for research, therefore, is to determine which among these varieties of civilian review boards will have the most impact on police department’s and police officers’ learning.

Notes
1. Perez’s initial study on this subject was conducted in 1978 as part of his dissertation. In 1992, an article that was published contained his findings and in 1994 a book was published. The latter publications were based on the data originally collected in 1978 and he elaborated on the issues raised in his dissertation.
2. Barton (1970) has suggested that such claims of police timidity to perform their functions due to the presence of the review boards have not been actually studied in detail.
3. Because the BPC can have operational control of the police, it cannot be directly compared to PLEB. The BPC can direct the police agency to change tactics. The PLEB does not have such influence and control over police operations.
4. A centralized and nationalized form of government requires the general conformity of local laws with the national laws.
5. At the time of data collection, the first author was the Dean of Student Affairs at Adamson University in the Philippines. Graduate students working for the Dean’s office were recruited and trained to administer the survey.
6. The researchers were instructed to secure a written permit from the mayor, barangay captain (synonymous to a neighborhood leader), or the PLEB before actually conducting citizen interviews. The endorsement of any of the above officials was necessary for the researchers to gain the confidence of the respondents. This element of confidence was critical because people, especially complainants against the police, have the tendency to be hostile toward or non-cooperative with strangers for a number of reasons. Using people familiar with respondents as referrals tends to alleviate these problems.
7. There is suspicion that a large proportion of police officers may have left police service as a result of being subject to a complaint. However, this is not likely the case. First, assignments are not a matter of choice for the officers but are service demands. second, anecdotal evidence indicates that officers who left the service did so due to causes such as death, mandatory retirement (i.e. they reached the age of 56), disability, or other opportunities. Third, the police with cases sample targeted officers charged over a 10-year period. During this period, political turmoil caused many officers to leave police service. Admittedly, we are unsure of the exact reasons why we were unable to contact a greater number of these officers, though we are confident that most officers did not leave because a complaint was filed against them or because of actions of the PLEBs as most citizen complaints were not sustained.

References
American Civil Liberties Union (1991), “Both federal and local reforms are needed to stop police brutality”, in Dudley (Ed.), Police Brutality, Greenhaven Press, Inc., San Diego, CA, pp. 130-9.
Armstrong, S. and Wood, D.B. (1991), “Curbing police brutality: an overview”, in Dudley (Ed.), Police Brutality, Greenhaven Press, Inc., San Diego, CA, pp. 103-6.
Barton, P.G. (1970), “Civilian review boards and the handling of complaints against the police”, University of Toronto Law Journal, Vol. 20, pp. 448-68.
Bayley, D.H. (1994), Police the Future, Oxford University Press, New York, NY.
Brown, S., Esbensen, F. and Geis, G. (2001), Criminology: Explaining Crime and Its Context, Anderson Publishing, Cincinnati, OH.
Carillo, D. (1991), “Civilian review boards can reduce police brutality”, in Dudley (Ed.), Police Brutality, Greenhaven Press, Inc., San Diego, CA, pp. 124-5.
Crank, J. and Caldero, M. (2001), Police Ethics: The Corruption of Noble Cause, Anderson Publishing, Cincinnati, OH.
Frank, J., Brandi, S.G. and Worden, R.E. (1996), “Citizen involvement in the coproduction of police outputs”, Journal of Crime and Justice, Vol. 19 No. 2, pp. 1-30.
Fyfe, JJ. (1985), “Reviewing citizens’ complaints against police”, in Fyfe, JJ. (Ed.), Police Management Today: Issues and case Studies, International City Management Association, Washington, DC, pp. 77-87.
Gellhorn, W. (1966), “Police review boards: hoax or hope”, Columbia University Forum, pp. 5-10.
Gove, W.R., Hughes, M.H. and Geerken, M. (1985), “Are uniform crime reports a valid indicator of the index crime rates?”, Criminology, Vol. 23 No. 3, pp. 451-501.
Grinc, R.M. (1994), “Angels in marble: problems in stimulating community involvement in community policing”, Crime and Delinquency, Vol. 40 No. 3, pp. 437-68.
de Guzman, M.C. (2001), “Integrity, legitimacy, efficiency, and impact: do all these matter in the civilian review of the police”, unpublished Dissertation, University of Cincinnati, Cincinnati, OH.
Halpern, S.C. (1974), “Police employee organizations and accountability procedures in three cities: some reflections on police policy-making”, Law and Society Review, Vol. 8 No. 4, pp. 561-82.
Kelling, G.L. (1995), “Measuring what happens”, paper for ‘Measuring What Matters”, 28 November 1995, the Police Research Institute, US Department of Justice, The National Institute of Justice and Office of the Community Oriented Policing Services, Washington, DC.
Kerley, K.R. and Benson, M.L. (2000), “Does community-oriented policing help build stronger communities?”, Police Quarterly, Vol. 3 No. 1, pp. 46-69.
Kerstetter, W.A. (1985), “Controlling the police – the art of the possible: a rejoinder to Napper, Pomeroy, and Schwartz”, in Geller (Ed.), Police Leadership in America: Crisis and Opportunity, American Bar Foundation and Praeger, Chicago, IL, pp. 196-8.
Littlejohn, EJ. (1981), “Civilian police commission: deterrent to police misconduct”, University of Detroit Journal of Urban Law, Vol. 59 No. 1, pp. 5-62.
Maguire, E.R. and Mastrofski, S.D. (2000), “Patterns of community policing in the United States”, Police Quarterly, Vol. 3 No. 1, pp. 4-45.
Manning, P.K. (1978), “Lying, secrecy and social control”, in Manning and Van Maanen (Eds), Policing: A View from the Street, Goodyear Publishing Co., Santa Monica, CA, pp. 238-55.
Memorandum Circular No. 91002, National Police Commission, Makati City, Philippines.
Paternoster, R. (1987), “The deterrent effect of perceived certainty and severity of punishment: a review of the evidence and issues”, Justice Quarterly, Vol. 4 No. 2, pp. 173-217.
Perez, D. (1978), “Police accountability: a question of balance”, PhD dissertation, University of California, Berkeley.
Perez, D. (1992), “Police review systems”, MIS Report, Vol. 24 No. 8, ICMA.
Perez, D. (1994), Common Sense About Police Review, Temple University Press, Philadelphia, PA.
Republic Act (RA) No. 6975.
Rossi, P.H. and Freeman, H.E. (1989), Evaluation: A Systematic Approach, Sage Publications, New York, NY.
Russell, K. V. (1978), “Complaints against the police: an international perspective”, Police Journal, Vol. 52 No. 1, pp. 34-44.
Schwartz, A. (1985), “Reaching systemic police abuses – the need for civilian investigation of misconduct: a response to Wayne Kerstteter”, in Geller (Ed.), Police Leadership in America: Crisis and Opportunity, Praeger, New York, NY, pp. 187-92.
Sherman, L.W. (1978), Scandal and Reform: Controlling Police Corruption, University of California Press, Berkeley, CA.
Sherman, L.W. (1983), “Reducing police gun use: critical events, administrative policy, and organizational change”, in Punch, M. (Ed.), Control in the Police Organization, MIT Press, Cambridge, MA.
Sherman, L.W. (1990), “The police crackdowns: initial and residual deterrence”, in Tonry, M. and Morris, N. (Eds), Crime and Justice: A Review of Research, Vol. 12, University of Chicago Press, Chicago, IL, pp. 1-48.
Sherman, L.W. (1992), Policing Domestic Violence, Free Press, New York, NY.
Sherman and Weisburd (1995), “General deterrent effects of police patrol in crime ‘hot spots’: a randomized trial experiment”, Justice Quarterly, Vol. 12 No. 4, pp. 625-48.
Smit, B.F. and Botha, CJ. (1990), “Democracy and policing: an introduction to paradox”, South African Journal of Criminology, Vol. 3 No. 1, pp. 36-42.
Terrill, RJ. (1988), “Police accountability in Philadelphia: retrospect and prospects”, American Journal of Police, Vol. 7 No. 2, pp. 79-99.
Wagner, A.E. and Decker, S.H. (1993), “Evaluating citizen complaints against the police”, in Dunham and Alpert (Eds), Critical Issues in Policing: Contemporary Readings, 2nd ed., Waveland Press, Inc., Prospect Heights, IL, pp. 275-91.
Walker, SL. (2001), Police Accountability: The Role of Citizen Oversight, Wadsworth, Belmont, CA.
Walker, S. and Bumphus, V.W. (1992), “Effectiveness of civilian review: observations on recent trends and new issues the civilian review of the police”, American Journal of Police, Vol. 11 No. 4, pp. 1-16.
Wilson, J.Q. (1968), Varieties of Police Behavior, Harvard University Press, Cambridge, MA.

Author Affiliation
Melchor C. de Guzman
School of Public and Environmental Affairs, Indiana University South Bend, South Bend, Indiana, USA
James Frank
Division of Criminal Justice, University of Cincinnati, Cincinnati, Ohio, USA