Correlates of the endorsement of two competing policing styles among police cadets in China

ABSTRACT

The purpose of this study is to investigate police cadets’ endorsement of the law enforcement style and the community policing style of policing at a time when both styles of police practices have received significant attention in China. The data used in the analysis were collected from a national police university in China featuring a large sample of 900+ cadets. The first research objective concerns the dimensionality of sentiments toward the two styles of policing among police cadets. The second is the identification of key factors associated with these styles. The principal findings are that police cadets do see the law enforcement style and the community policing style as two distinctive conceptual dimensions. Moreover, different factors are associated with endorsement of each of the two styles of policing. Policy implications of these findings are discussed in some detail.

KEYWORDS:

• Policing
• cadets
• law enforcement
• community policing
• construct coherence

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. This comprehensive guideline regarding the implementation of community policing is considered as an official document issued by the Ministry of Public Security (Zhong, 2009).

2. High return rates of self-report surveys are commonly seen in the prior school-based studies conducted in China. For example, in their comparative study on adolescent youth behavior in China and the USA, Jessor et al. (2003) reported that the response rate of 98 percent obtained from the Chinese sample was much higher than that of the US sample (74 percent). Similar high response rates were also reported by Greenberger, Chen, Beam, Whang, and Dong (2000) and Webb, Ren, Zhao, He, and Marshall (2011).

3. Hinton, Brownlow, McMurray & Cozens (2004, p. 364) noted, ‘There is much debate among researchers as to where the appropriate cut-off points for reliability. A good guide is ….0.70–0.90 shows high reliability, 0.70–0.50 shows moderate reliability, and 0.50 and below shows low reliability’ (also see: DeVillis, 1991; Tavakol & Dennick, 2011). In addition, inter-item correlation can be used to assess the integrity of a factor. In his Meta-analysis of alpha levels, Peterson found that an average inter-item correlation of .31 is equivalent to an average alpha level of .80 based on a factor of 9 items (Peterson, 1994). More specifically, the commonly accepted range of inter-class correlations is from .15 to .50 (Clark & Watson, 1995) or more strict standards from .30 to .50. The mean of inter-item correlations of the law enforcement factor is .30, which is within the range of acceptable level of scale coherence.

4. The seriousness of dating violence incidents depends on a number of factors. For example, if one of the parties is not injured (e.g., verbal arguments), police officers usually handle the incident informally. However, if one of the parties is seriously injured, it can constitute a crime incident. Police officers will make an arrest of the attacker. A prosecutor is likely to charge him for aggravated assault. Rape can occasionally happen in dating violence if both parties know each other online and the male has sex with the woman without her consent (e.g., using force or drugs).

5. The debate on the definition of the community policing style vs. the law enforcement style started from the early days of American police reform in the 1980s (Cordner, 1998). Today, there is lack of clear definition in the U.S. as well as in China (Chen, 2004). Not surprisingly, it can be argued that some items may not fall under the umbrella of a universal concept. In this study we have relied on the factor analysis to differentiate one style from the other style, and we have assumed further that items loaded onto one factor reflect the theoretical concept among our survey respondents.

6. Multicollinearity is a potentially serious problem with ordinary least squares (OLS) regression analysis. Accordingly, the variance inflation factor (VIF) often is used to detect whether high collinearity exists between any variables. Some researchers use a VIF score of 4 or greater as an indication of severe multicollinearity (Fisher & Mason, 1981; Judge, Hill, Griffiths, Lutkepohl, & Lee, 1988). Our collinearity statistics showed that none of the VIF values exceeded 1.5; accordingly, we concluded that multicollinearity was not a problem in our analysis.

Additional information

Notes on contributors

Jia Qu

Jia Qu is an assistant professor at Guizhou University of Finance and Economics. Her current research interests include policing, law and social research.

Xinting Wang

Xinting Wang is a Ph.D student in the Department of Criminal Justice and Criminology at Sam Houston State University. Her research interests include policing, law enforcement, police-citizen relationship, police legitimacy, and organizational culture.

Jihong (Solomon) Zhao

Jihong (Solomon) Zhao is a professor in the Department of Criminal Justice and Criminology at Sam Houston State University. He was formerly an associated professor in the Department of Criminal Justice, University of Nebraska at Omaha. His current research interests include police organization, with a particular emphasis on organizational change, culture, and individual behavior and values.