Changing Demographics article review

policearticle.pdf

Home >Law homework help >Changing Demographics article review

International Journal of Police Science & Management Volume 16 Number 3

Page 205

International Journal of Police Science and Management, Vol. 16 No. 3, 2014, pp. 205–220. DOI: 10.1350/ijps.2014.16.3.340

Workforce diversity: assessing the impact of minority integration on intra-workgroup interaction

Salomon Alcocer Guajardo Department of Public Management, John Jay College of Criminal Justice, The City University of New York, 445 West 59th Street, Room 3536, New York, NY 10019, USA. Tel: +1 646 557 4783; email: [email protected]

Submitted 12 March 2013; accepted 30 May 2014 Keywords: disparity index, path analysis, Simpson index, social interaction, workforce diversity

Salomon Alcocer Guajardo is associate profes- sor of public management at John Jay College of Criminal Justice, The City University of New York. He serves as the coordinator for the undergradu- ate Criminal Justice Management Program. His research on workforce diversity has appeared in Public Administration Review, Public Personnel Management and the Review of Public Person- nel Administration.

AbstrAct Research on workforce diversity and minority inte- gration in police agencies has focused primarily on the extent to which these agencies have hired men and women of different ethnic and racial back- grounds. Some studies have examined the impact of minority integration on police management and operations. However, the research has neglected to examine whether the integration of ethnic and racial minorities has an impact on intra-workgroup interactions in which White and minority police officers and civilians communicate and interact among each other. This study undertook a casual (or path) analysis of contextual organisational factors and their impact on intra-workgroup interaction. The findings show that contextual organisational factors have direct and indirect effects on the extent to which White and minorities within a workgroup are likely to interact.

INTRODUCTION Research on minority and female employ- ment integration in police agencies has focused primarily on the extent to which these agencies hire women and men of dif- ferent ethnic and racial backgrounds. Studies that have researched the integration of minor- ity men and women have discussed the gen- eral trend to which police agencies have hired Blacks, Hispanics and Asians (Harrington & Lonsway, 2004; Hochstedler, Regoli, & Poole, 1984; Martin, 2004; Schulz, 2004; Stokes & Scott, 1996; Walker, 1983, 1989; Winters, 1991). These studies, however, have ignored to assess whether workforce diversity has an impact on intra-workgroup interactions within police agencies and police ranks. This issue is important when group dynamics is assessed in police agencies because the inte- gration of ethnic and racial minorities may have the unintended consequence of restrict- ing inter- and intra-workgroup interactions as well as reinforcing employee isolation as more minorities join the police force and are promoted to higher ranks.

In contrast to past research on workforce diversity in police agencies, this study exam- ines whether contextual organisational fac- tors (eg, type of workforce, workgroup size

Workforce diversity

Page 206

and position or rank status) are determinants of workforce diversity and intra-workgroup interaction. This study also addresses whether contextual organisational factors have direct or indirect effects on intra-workgroup inter- action. This is accomplished through the use of path analysis. In so doing, the following questions are addressed:

●● Are there direct and indirect effects of contextual organisational factors on work- force diversity and the interactions among White and minority employees?

●● Does the level of diversity within police and civilian workforces enhance inter- actions among White and minority employees?

●● Does the size of the workgroup influence the level of interaction among White and minorities employees?

●● Does the status of the police rank or civil- ian job classification influence the level of interaction among White and minority employees?

These questions are important to address because it has been asserted that the integra- tion of minorities within police departments should enhance the ‘interaction between minority and White police officers’ (McClus- key & McCluskey, 2004, p. 71; Smith, 2003).

LITERATURE REVIEW The research addressing minority recruit- ment and integration into police departments has primarily examined the rate to which minorities have been hired. Hochstedler et al. (1984) found that Black officers were under- represented in police departments, despite achieving employment gains from 1967 to 1981. Walker (1989) found that police depart- ments in the 50 largest cities had made uneven progress in hiring Blacks and Hispanics from 1983 to 1988. With regard to the employment of Hispanic police officers, McCluskey and McCluskey (2004) Winters (1991), Walker and

Turner (1992) and other studies have found that police departments had achieved some success in hiring Hispanics since the 1980s. In some cases, police departments have been suc- cessful in hiring Hispanics in comparison with Blacks (Walker & Turner, 1992). Although police departments have hired more Blacks and Hispanics, the extent to which these minority hires have made significant contri- butions to increasing the overall workforce diversity of the agencies is not fully known.

The research also shows that police agen- cies have made some progress in hiring Asian police officers (Stokes & Scott, 1996; Walker, 1983; Winters, 1991). According to Winters, the Chicago Police Department increased the number of Asian police officers between 1985 and 1987. Stokes and Scott found that some police agencies had Asian police officers by 1990, although other agencies did not. Despite their efforts to increase significantly the num- ber of Asian police officers, police departments have had a difficult time doing so (Stokes & Scott, 1996; Walker, 1989; Winters, 1991).

Similarly, research on minority and White female integration into police agencies has focused on the rate at which females are hired by police departments (Hochstedler et al., 1984; Martin, 2004; Winters, 1991). Hochstedler et al. found that the New York Police Department and Washington, DC Police Department had increased the num- ber of female police officers from 1974 to 1981. Winters (1991) found that the percent- age of White and Black female officers had increased in the Chicago Police Department from 1985; however, Winters noticed that Hispanic female officers were almost non- existent. In addition, Martin (2004) noted that in municipal police agencies, female officers continue to compose fewer than 10 per cent of the sworn officers. Although these studies have found that police agencies have hired some female officers over time, the extent to which White and minority female officers add to the overall diversity within police departments is unclear.

Guajardo

Page 207

Few studies have examined the organi- sational factors that contribute to minority integration into public and police agencies (Choi, 2010; Kellough, 1990; Kim, 1993; McCluskey & McCluskey, 2004). In exam- ining the contextual factors that contribute to the employment of Hispanic police offic- ers, McCluskey and McCluskey (p. 69) used agency size as an independent variable to assess ‘the changes in diversity of law enforce- ment agencies between 1990 and 2000. . .’. They hypothesised that larger police agen- cies are more likely to hire more minorities than smaller agencies. The logistic regression results indicate that agency size was not a significant predictor at α = 0.05. Because the dependent variable was dichotomised, a score of workforce diversity was not calculated or used in the regression model; consequently, the contribution that Hispanic offices make to the overall level of workforce diversity was not measured.

Recent studies on workforce diversity have begun to examine the impact of work- force diversity and management on the per- formance of public organisations. Pitts (2005) examined whether the level of workforce diversity in Texas public schools influenced school performance. In this study, Pitts found mixed results in which the diversity of the teaching staff was positively correlated with dropout rates and SAT scores, but negatively correlated with Texas Assessment of Aca- demic Skills (TAAS) scores. Pitts (2005) also found that the diversity among management staff had a non-significant statistical result with the dropout rate, SAT scores and TAAS scores. More recently, Pitts (2009) exam- ined whether the management of workforce diversity in organisations led to higher group performance. Pitts (2009) found that the effective management of workforce diver- sity influenced the perception of group per- formance. This type of research, however, has not yet been conducted with respect to whether an increase in minority repre- sentation and diversity in police agencies

leads to effective law enforcement services or whether it encourages more social inter- actions among White and minority police officers. In a police department, studying the impact of workforce diversity and its man- agement is important because it provides insight regarding how minority integration impacts the organisational structure, race relations among police officers and the deliv- ery of law enforcement services to citizens in the community.

Although the studies cited above con- tribute to our understanding of the extent to which police agencies hire ethnic and racial minorities, there are a number of issues that have not been addressed adequately by researchers. For example, because diversity indices have not been used in prior studies to benchmark the degree of homogeneity or heterogeneity that exists within police agencies, it is unclear whether the addi- tional hiring of ethnic and racial minority police officers increases the level of work- force diversity in the police department and within each police rank significantly. In addition, it is unclear whether the integra- tion of ethnic and minority police officers into a police department increases the rate of interaction among White and minority officers. Further, it is unclear whether con- textual organisational factors and the level of workforce diversity within each police rank have direct and indirect effects on White and minority police officer interactions. These issues are important to address because they may have a significant impact on race rela- tions and organisational performance within police departments.

As an exploratory study on workforce diversity and its impact on intra-workgroup interaction in a police agency, this study attempts to address the questions and issues enumerated above. In so doing, this study uses the Simpson Diversity Index to meas- ure the degree of homogeneity that exists within the police ranks and job classifica- tions of the New York City Department of

Workforce diversity

Page 208

Police (NYPD). The diversity scores are then used to examine whether workforce diver- sity has a direct effect on the level of intra- workgroup interaction in addition to serving as a mediating variable. This study also uses an interaction index to assess the probabil- ity that a White police officer within each rank will interact with a minority officer. Finally, this study proposes a conceptual model illustrating the direct and indirect effects of contextual organisational factors on intra-workgroup interactions in order to test whether there is a causal relationship among the contextual factors, workforce diversity and intra-workgroup interaction.

CONCEPTUAL FRAMEWORK OF DIVERSITY AND INTRA-WORKGROUP INTERACTION Because past research suggests that contex- tual organisational factors influence the inte- gration of minorities and workforce diversity within police agencies, Figure 1 shows the theoretical relationships among workgroup size, type of workgroup, position status, minority employment disparity, workforce diversity and intra-workgroup interaction. As indicated in the literature review, organisa- tional size is hypothesised to be a significant predictor of minority integration (Kellough, 1990; Kim, 1993; McCluskey & McCluskey, 2004). The findings of past studies suggest that larger workforces have lower levels of minority employment disparity and higher

levels minority integration (Kellough, 1990; Kim, 1993; McCluskey & McCluskey, 2004). However, this may not hold true when examining the size of workgroups and its impact on minority employment disparity and diversity within police agencies. When assessing the impact of workgroup size on minority employment disparity and diver- sity, small or mid-size workgroups may have lower levels of employment disparity and higher levels of diversity in comparison with larger workgroups. This may result from the need of smaller groups to integrate more eth- nic and racial minorities because they may be perceived as being elitist or discriminatory in accepting ethnic or racial minorities. Large workgroups within police agencies may be less integrated than smaller workgroups because they have more opportunities to select more Whites and fewer minorities. In selecting fewer ethnic and racial minorities and more Whites, larger workgroups within police agencies may give the perception that they are inclusive when in reality they are not and thereby have no impact on the exist- ing level of employment disparity and diver- sity within the workgroup. In addition, the size of the workgroup is likely to affect the interactions of members within the group. In large workgroups within police agen- cies, Whites may not need to interact with minority group members, whereas in smaller workgroups, Whites and minorities may be more likely to interact more frequently with each other out of necessity.

Figure 1 Conceptual model for minority employment

disparity, diversity and intra-workgroup interaction

Guajardo

Page 209

The nature of the workgroup may also affect the level of employment disparity among White and minority officers, the level of work- force diversity, and the interactions amongst workgroup members within police agencies. In large police departments, there are two types of workgroups: police officers and civilians. Traditionally, police forces have been White male dominated and relatively homogeneous, requiring little interaction with minorities. The nature of police work also reinforces the lack of interactions among White and minor- ity police officers because police officers gen- erally work alone or in groups of two, which limit the interactions among police officers. By contrast, civilian workforces within police departments tend to have generally higher lev- els of gender, ethnic and racial integration and lower levels of minority employment disparity. The nature of the civilian jobs also requires that Whites and minorities interact more fre- quently as a result of collaboration and link- ages among and within functions and tasks. In addition, civilian-based and -centred functions may encourage more inter- and intra-work- group interactions because of the workflow and coordination associated with processing documents and reports.

Job position or rank status may also have an impact on the extent to which work- groups within police agencies are homo- geneous (or heterogeneous) and the extent to which group members interact with each other. Positions that are non-super- visory or management-based are likely to be more ethnically and racially integrated than management-based positions, generally because non-management-based positions require less education than management- based positions. Within police departments, police ranks (eg, lieutenant, captain, inspec- tor) that require high levels of education and experience tend to be composed primarily of White males, whereas police ranks that are non-supervisory (eg, police officer and detective) tend to be composed of Whites and minorities. Because management-based

positions tend to be White male dominated, these workgroups are likely to have high lev- els of minority employment disparity and low levels of ethnic and racial diversity. Within police departments, management-based posi- tions also are likely to foster less frequent inter- and intra-workgroup interactions with police officers and civilians because of the nature of police work. In addition, management-based positions that are located within the upper police ranks or civilian workforce may influ- ence personnel to have less frequent interac- tions with lower-ranking police officers and civilian personnel, respectively.

Historical levels of minority employment disparity within workgroups may also impact how quickly they become ethnically and racially integrated and increase their level of workgroup diversity. Workgroups that have been historically White male dominated may have more difficulty integrating and interact- ing with minorities than workgroups which have been more apt to integrate ethnic and racial minorities in the past. One reason is that White-dominated workgroups are more likely to resist the integration of minorities. It is also plausible that few minorities are likely to seek integration into workgroups that have been perceived to be exclusively White. By contrast, workgroups with historically low levels of minority employment disparity may be more likely to have higher levels of integration and diversity. These relatively het- erogeneous workgroups are also more likely to accept men and women of different ethnic and racial backgrounds and encourage intra- workgroup interactions among White and minority members.

Finally, the level of workforce diversity that exits within a workgroup may mediate the level of interaction within and across work- groups. Workforces and workgroups with historically low levels of diversity may have high levels of interaction among their mem- bers because they are ethnically and racially homogeneous. When minorities are intro- duced into these relatively homogeneous

Workforce diversity

Page 210

workgroups, the level of interaction may decrease because the workgroup becomes more heterogeneous as a result of the minor- ity integration that has taken place. In police agencies, this may be the case within mid- and high-level police ranks that have his- torically been composed primarily of White males. This may also be the case among civil- ian workgroups within police agencies.

PATH ANALYSIS EQUATIONS Based on the conceptual framework pre- sented in Figure 1, the path analysis for intra-workgroup interaction (WI) consists of three regression equations. The first assesses whether workgroup size (WS), position sta- tus (PS) and type of workgroup (WT) have direct effects on minority employment dis- parity (MED) and is as follows: ŷ

MED = α +

β 1 WS + β

2 PS + β

3 WT + ε

ij . The second equa-

tion assesses whether WS, PS, WT and MED have direct and indirect effects on workforce

diversity (WD) and is as follows: ŷ WD

= α + β

1 WS + β

2 PS + β

3 WT + β

4 MED + ε

ij. The

third regression equation assesses whether WS, PS, WT, MED and WD have direct and indirect effects on intra-workgroup interac- tion (WI) and is as follows: ŷ

WI = α + β

1 WS

+ β 2 PS + β

3 WT + β

4 MED + β

5 WD + ε

ij . As

discussed in the section ‘Data and Method’, forward selection is used to identify the inde- pendent and mediating variables which are significant predictors of MED, WD and WI.

STATISTICAL HYPOTHESES Table 1 summarises the null and alternative hypotheses of the three regression equa- tions for intra-workgroup interaction. As presented in Figure 1, the first equation in the causal model uses workgroup size (WS), type of workgroup (WT) and position status (PS) as the independent variables (IVs). Each beta (β) is tested at α = 0.01. For the sec- ond equation in the model, each β is tested

Table 1: Null and alternative hypotheses for intra-workgroup interaction

Dependent variables

Independent variables Minority Employment Disparity (MED)

Workforce Diversity (WD) Intra-workgroup Interaction (WI)

Intercept Ho: α = 0 at α = 0.01 Ho: α = 0 at α = 0.01 Ho: α = 0 at α = 0.01

Ha: α ≠ 0 at α = 0.01 Ha: α ≠ 0 at α = 0.01 Ha: α ≠ 0 at α = 0.01

Workgroup Size (WS) Ho1: β1 = 0 at α = 0.01 Ho1: β1 ≠ 0 at α = 0.01 Ho1: β1 = 0 at α = 0.01

Ha1: β1 ≠ 0 at α = 0.01 Ha1: β1 ≠ 0 at α = 0.01 Ha1: β1 ≠ 0 at α = 0.01

Workgroup Type (WT) Ho2: β2 = 0 at α = 0.01 Ho2: β2 = 0 at α = 0.01 Ho2: β2 = 0 at α = 0.01

Ha2: β2 ≠ 0 at α = 0.01 Ha2: β2 ≠ 0 at α = 0.01 Ha2: β2 ≠ 0 at α = 0.01

Position Status (PS) Ho3: β3 = 0 at α = 0.01 Ho3: β3 = 0 at α = 0.01 Ho3: β3 = 0 at α = 0.01

Ha3: β3 ≠ 0 at α = 0.01 Ha3: β3 ≠ 0 at α = 0.01 Ha3: β3 ≠ 0 at α = 0.01

Mediating Variables

MED – Ho4: β4 = 0 at α = 0.01 Ho4: β4 = 0 at α = 0.01

Ha4: β4 ≠ 0 at α = 0.01 Ha4: β4 ≠ 0 at α = 0.01

WD – – Ho5: β5 = 0 at α = 0.01

Ha5: β5 ≠ 0 at α = 0.01

Guajardo

Page 211

T ab

le 2

: E th

n ic

a n d r

ac ia

l d is

tr ib

u ti o n o

f N

ew Y

o rk

C it y P

o li ce

p er

so n n el

b y g

en d er

a n d p

o li ce

r an

k f o r

2 0 1 2

P ol

ic e

R an

k W

hi te

B la

ck H

is pa

ni c

A si an

N at

iv e

A m

er ic an

T ot

al T ot

M in

or ity

M in

or ity

is pa

ri ty

I nd

ex S im

ps on

iv er

si ty

I nd

ex In

tr a-

w or

kg ro

In te

ra ct io

n In

de x

M al

e po

lic e

of fic

er s

B u re

au c

h ie

fs 5

1 1

0 0

7 2

2. 50

0. 44

90 0.

00 01

A ss

is ta

n t

ch ie

fs 16

2 2

1 0

21 5

3. 20

0. 39

91 0.

00 02

D ep

u ty

c h ie

f 42

4 3

0 0

49 7

6. 00

0. 25

49 0.

00 04

In sp

ec to

r 89

4 7

1 0

10 1

12 7.

42 0.

21 70

0. 00

D ep

u ty

i n sp

ec to

r 13

7 5

12 2

0 15

6 19

7. 21

0. 22

16 0.

00 10

C ap

ta in

31 2

22 38

8 0

38 0

68 4.

59 0.

31 21

0. 00

L ie

u te

n an

t C

o m

m an

di n g

O ffi

ce r

(C O

) 10

7 2

17 2

0 12

8 21

5. 10

0. 28

31 0.

00 11

L ie

u te

n an

t sp

ec ia

l as

si gn

m en

t 69

10 11

0 0

90 21

3. 29

0. 38

49 0.

00 10

L ie

u te

n an

t 99

8 99

19 5

45 3

1, 34

0 34

2 2.

92 0.

41 75

0. 01

Se rg

ea n t

su pe

rv is

o r

de te

ct iv

e 20

8 25

41 11

0 28

5 77

2. 70

0. 43

75 0.

00 35

Se rg

ea n t

sp ec

ia l as

si gn

m en

t 14

6 15

24 5

0 19

0 44

3. 32

0. 38

66 0.

00 21

Se rg

ea n t

2, 11

1 39

8 65

1 24

3 7

3, 41

0 1,

29 9

1. 63

0. 56

16 0.

04 95

D et

ec ti ve

fi rs

t gr

ad e

21 0

55 56

2 1

32 4

11 4

1. 84

0. 52

12 0.

00 45

D et

ec ti ve

s ec

o n d

gr ad

e 54

6 11

8 18

2 21

3 87

0 32

4 1.

69 0.

54 34

0. 01

D et

ec ti ve

i nv

es ti ga

to r

1, 39

9 34

7 67

3 76

3 2,

49 8

1, 09

9 1.

27 0.

59 35

0. 03

D et

ec ti ve

s pe

ci al

is t

29 8

54 80

4 0

43 6

13 8

2. 16

0. 48

38 0.

00 58

P o lic

e o ffi

ce r

9, 55

1 2,

51 8

4, 90

3 1,

32 2

18 18

,3 12

8, 76

1 1.

09 0.

63 22

0. 28

T o ta

l: A

ll ra

n ks

16 ,2

44 3,

67 9

6, 89

6 1,

74 3

35 28

,5 97

12 ,3

53 1.

31 0.

59 89

57 %

13 %

24 %

6% 0.

1% 10

0% 43

(C on

tin ue

d )

Workforce diversity

Page 212

T ab

le 2

: (C

on ti

n u ed

)

P ol

ic e

R an

k W

hi te

B la

ck H

is pa

ni c

A si an

N at

iv e

A m

er ic an

T ot

al T ot

M in

or ity

M in

or ity

is pa

ri ty

I nd

ex S im

ps on

iv er

si ty

I nd

ex In

tr a-

w or

kg ro

In te

ra ct io

n In

de x

F em

al e

P ol

ic e

O ffi

ce s

B u re

au c

h ie

fs 1

0 0

0 1

0 0.

00 0.

00 00

0. 00

A ss

is ta

n t

ch ie

fs 1

0 0

0 1

0 0.

00 0.

00 00

0. 00

D ep

u ty

c h ie

f 3

0 0

0 3

0 0.

00 0.

00 00

0. 00

In sp

ec to

r 2

4 1

0 8

6 0.

33 0.

65 63

0. 00

D ep

u ty

i n sp

ec to

r 6

2 2

0 10

4 1.

50 0.

56 00

0. 00

C ap

ta in

17 4

0 0

27 10

1. 70

0. 53

22 0.

00 37

L ie

u te

n an

t C

o m

m an

di n g

O ffi

ce r

(C O

) 2

2 1

0 5

3 0.

67 0.

64 00

0. 00

L ie

u te

n an

t sp

ec ia

l as

si gn

m en

t 5

3 2

0 10

5 1.

00 0.

62 00

0. 00

L ie

u te

n an

t 74

44 39

0 16

4 90

0. 82

0. 66

60 0.

02 38

Se rg

ea n t

su pe

rv is

o r

de te

ct iv

e 10

10 6

0 26

16 0.

63 0.

65 09

0. 00

Se rg

ea n t

sp ec

ia l as

si gn

m en

t 7

10 5

0 23

16 0.

44 0.

66 92

0. 00

Se rg

ea n t

23 8

26 1

20 7

2 3

2 73

1 49

3 0.

48 0.

68 53

0. 09

D et

ec ti ve

fi rs

t gr

ad e

11 12

1 0

35 24

0. 46

0. 68

41 0.

00 44

D et

ec ti ve

s ec

o n d

gr ad

e 35

41 27

0 10

5 70

0. 50

0. 66

99 0.

01 37

D et

ec ti ve

i nv

es ti ga

to r

14 8

99 13

3 0

38 0

23 2

0. 64

0. 66

33 0.

05 30

D et

ec ti ve

s pe

ci al

is t

28 24

0 0

72 44

0. 64

0. 66

05 0.

01 00

P o lic

e o ffi

ce r

1, 11

6 1,

36 1

1, 63

7 10

3 4

4, 22

1 3,

10 5

0. 36

0. 67

51 0.

48 18

T o ta

l: A

ll ra

n ks

1, 70

4 1,

87 7

2, 09

4 14

1 6

5, 82

2 4,

11 8

0. 41

0. 68

29 %

32 %

36 %

2% 0.

1% 10

0% 71

Guajardo

Page 213

at α = 0.01 with the β for minority employ- ment disparity (MED), the mediating vari- able, tested at α = 0.01. In the third equation of the models, the β of the independent and mediating variables are tested at α = 0.01. Although the intercepts in each regression equation are not used in the path analysis, they are tested at α = 0.01 as well and are included for completeness.

DATA AND METHOD For this study, Equal Employment Oppor- tunity Commission (EEOC) employment data for police and civilian employees of the NYPD for 2012 are used. Table 2 summarises the data for the NYPD police workforce by gender by ethnicity (or race) by rank. The police force employment data facilitate the calculation of a minority employment dis parity (MED) ratio, as well as Simpson Diversity Index scores for each police rank. These data also facilitate the calculation of interaction index scores for each police rank and civilian job classification.

In addition, the data for the civilian work- force provide sufficient information to assess the employment disparity among White and minority employees for each workgroup, as well as measuring the level of diversity within each workgroup (Table 3). The civilian-based data also allow for assessing the degree of interaction within each workgroup. For this study, the data are sufficient to assess whether the independent and mediating variables have direct and indirect effects on intra- workgroup interaction.

Position status was operationalised by classifying ranks or job classifications in the police and civilian workgroups as either supervisory/management based or non- supervisory. In the police workforce, ranks of sergeant and above were classified as super- visory/management based, and the ranks of police officer to detective first grade were classified as non-supervisory. In the civil- ian workforce, the job classifications of

professionals and below were classified as non-supervisory, and the job classification of officials and administrators was classi- fied as supervisory/management-based. The supervisory/ management-based positions were dummy coded as 1 and the non-supervisory positions were dummy coded as 0.

Type of workgroup was operationalised as either police or civilian. Police-based work- groups were coded as 1, and civilian-based workgroups were coded as 0. This classifica- tion method was based on the NYPD EEOC report where separate employment data are reported for its police and civilian work- forces.

Size of workgroup (or workgroup size) was based on the total number of employees in each male and female police and civilian workgroup. Each workgroup is of a differ- ent size, which allows us to assess whether larger workgroups have higher rates of intra- workgroup interaction in comparison with smaller groups. Because workgroup size is used by a number of the indices that serve as independent or mediating variables, the size of the workgroup was converted to a rank. The smallest workgroup within the police workforce was assigned the rank of 1 and the largest assigned a rank of 34. For the civil- ian workforce, the smallest workgroup was assigned a rank of 1 and the largest assigned a rank of 16.

To measure the degree of minority employment disparity that exists within each police rank and civilian job classification (ie, workgroup), the number of White employees was divided by the total number of ethnic and racial minorities within each workgroup. This ratio produces the number of White employees per minority employee in the workgroup. A ratio > 1.0 indicates that there are more White employees per minority employee; conversely, a ratio < 1.0 indicates that there are more minority employees per White employee (Fridell, 2004). A ratio of 1.0 indicates that there are an equal number of White and minority employees. As shown

Workforce diversity

Page 214

T ab

le 3

: E th

n ic

a n d r

ac ia

l d is

tr ib

u ti o n o

f N

ew Y

o rk

C it y P

o li ce

p er

so n n el

b y c

iv il ia

n c

la ss

ifi ca

ti o n f o r

2 0 1 2

C iv

ili an

c la

ss ifi

ca tio

n W

hi te

B la

ck H

is pa

ni c

A si an

N at

iv e

A m

er ic an

T ot

al T ot

M in

or ity

M in

or ity

D is pa

ri ty

de x

S im

ps on

iv er

si ty

I nd

ex In

tr a-

w or

kg ro

In te

ra ct io

n In

de x

M al

e ci vi

lia n

w or

kf or

O ffi

ci al

s/ ad

m in

is tr

at o rs

13 3

73 41

2 4

1 27

2 13

9 0.

96 0.

65 84

0. 05

P ro

fe ss

io n al

s 14

0 40

18 2

6 0

22 4

84 1.

67 0.

55 76

0. 04

T ec

h n ic

ia n s

10 9

64 35

3 8

0 24

6 13

7 0.

80 0.

69 19

0. 05

P ro

te ct

iv e

se rv

ic es

30 3

1, 59

7 75

1 60

8 6

3, 26

5 2,

96 2

0. 10

0. 66

46 0.

23 26

P ar

a- pr

o fe

ss io

n al

s 5

4 0

0 11

6 0.

83 0.

62 81

0. 00

O ffi

ce a

n d

cl er

ic al

83 15

2 40

5 4

1 33

0 24

7 0.

34 0.

68 31

0. 05

Sk ill

ed c

ra ft

s 36

2 60

60 4

9 4

53 5

17 3

2. 09

0. 50

86 0.

09 90

Se rv

ic e

an d

m ai

n te

n an

ce 47

10 1

74 1

7 2

24 1

19 4

0. 24

0. 68

70 0.

03 20

T o ta

l: A

ll ra

n ks

1, 18

2 2,

09 1

1, 01

9 81

8 14

5, 12

4 3,

94 2

0. 30

0. 71

23 %

41 %

20 %

16 %

0% 10

0% 77

F em

al e

ci vi

lia n

w or

kf or

O ffi

ci al

s/ ad

m in

is tr

at o rs

18 6

31 5

68 3

8 0

60 7

42 1

0. 44

0. 62

03 0.

07 86

P ro

fe ss

io n al

s 14

1 86

35 4

2 1

30 5

16 4

0. 86

0. 67

46 0.

04 62

T ec

h n ic

ia n s

41 20

3 33

9 0

58 6

54 5

0. 08

0. 55

19 0.

02 32

P ro

te ct

iv e

se rv

ic es

86 8

4, 40

5 1,

83 3

14 2

21 7,

26 9

6, 40

1 0.

14 0.

55 45

0. 46

P ar

a- pr

o fe

ss io

n al

s 2

4 1

0 9

7 0.

29 0.

69 14

0. 00

O ffi

ce a

n d

cl er

ic al

39 6

2, 41

7 51

1 9

2 4

3, 42

0 3,

02 4

0. 13

0. 46

41 0.

21 32

Sk ill

ed c

ra ft

s 0

0 0

0 0.

00 0.

00 00

0. 00

Se rv

ic e

an d

m ai

n te

n an

ce 8

91 79

0 18

2 17

4 0.

05 0.

55 92

0. 00

T o ta

l: A

ll ra

n ks

1, 64

2 7,

52 1

2, 86

0 32

9 26

12 ,3

78 10

,7 36

0. 15

0. 55

13 %

61 %

23 %

3% 0%

10 0%

87 %

Guajardo

Page 215

in Tables 2 and 3, MED ratios were calculated for both the male and female workgroups in the police and civilian workforces.

The Simpson Diversity Index (1949) (D = 1 – Σp2) was used to measure the degree of homogeneity within each police and civil- ian workgroup. As stated by Simpson (1949, p. 688), the index obtains ‘the probability that two individuals chosen at random and inde- pendently from the population will be found to belong to the same group’. Mathemati- cally, the index provides a measurement of

Ethnicity or race Frequency (n) n2 p p2

White 17,949 322,166,601 0.521 0.272

Black 5,556 30,869,136 0.161 0.026

Hispanic 8,990 80,820,100 0.261 0.068

Asian 1,884 3,549,456 0.055 0.003

Native American 41 1,681 0.001 0.000

Total (Σ) 34,420 437,406,974 1.000 0.369

= − ∑

1D n

2 D = 1 – Σp2

D = −1 437,406,974

1,184,736,400

D = 1 – 0.369 D = 1 – 0.369

D = 0.631 D = 0.631

The level of intra-workgroup interaction (WI) within each male and female police and civilian workgroup was obtained by using the interaction index which is represented symbolically as follows:

W I = No. majority in workgroup

No. majority in workforce

No. minority in workgroup ×

Total employees in workgroup

As McClauley, Plummer, Moskalenko, and Mordkoff (2001) state, the index can be used to assess interaction between two or more

the level of homogeneity (or diversity) that exists within a workgroup or organisation. As developed by Simpson, the index has a range from 0 to 1, where 0 indicates the absence of diversity (ie, maximum homogeneity) and 1 indicates maximum heterogeneity (ie, diver- sity) (Blau, 1977; Guajardo, 1999; Lieber- son, 1969, Pitts, 2005; Simpson, 1999; Starks, 2009). To obtain a coefficient of diversity for each male and female workgroup within the police and civilian workforce of the NYPD, the Simpson index was applied as follows:

distinguishable groups defined in terms of age, gender, occupation, socio-economic status, etc. McClauley et al. (2004) also state that the index can be extended to social units includ- ing workgroups. With regard to applying the index to public agencies, the interaction index provides a measure of the extent to which there is interaction among Whites and minorities and whether more diverse workgroups have higher rates of interaction among different ethnic groups in comparison to workgroups with lower levels of ethnic or racial diversity.

When the interaction index was applied to the data for the male police officer

Workforce diversity

Page 216

workgroup, it produced a score of ~ 0.28, indicating that the minority percentage in the workgroup of a typical White male police officer is 28 per cent. To obtain this intra- workgroup interaction score for the male police officer workgroup, the following cal- culations were performed:

W I 9,551 White police o�cers

16,244 total White police o�cers =

8,761 minority police o�cers

18,312 male police o�cers ×

WI = 0.587971 × 0.478429 = 0.2813 This method was applied to each police

rank and civilian job classification by gender. Mathematically, the index produces either

a minority- or majority-weighted aver- age of majority or minority proportion in a workgroup or unit when it is applied to an organisation such as a police agency. In the example above, the interaction index yields the majority-weighted average because the data for White police officers were used. According to McClauley et al. (2001, p. 324) the ‘result is the proportion of [minority] seen by the average [majority] member look- ing around in his or her [workgroup]. . .’.

Because the index scores were calculated using some of the same data elements (eg, total employees in the workgroup), they were converted to ranks separately for the police and civilian workforces to reduce the level of serial correlation and collinearity among the indices serving as mediating variables, as well as to reduce the possibility that any signifi- cant findings are the result of statistical arte- facts rather than the direct or indirect effects of the independent and mediating variables. The index scores for MED, WD and WI based on the police EEOC data were ranked from 1 (the smallest score) to 34 (the high- est score). For the civilian workgroups, the lowest score was assigned a rank of 1 and the highest a rank of 16 on the same independ- ent variables. The same process was followed for workgroup size.

Multivariate linear regression with forward selection was used to regress the independent and mediating variables on intra-workgroup interaction. This approach was chosen to identify the significant predictors of intra- workgroup interaction. In addition, forward selection was chosen to reduce the causal model by identifying direct and indirect effects associated only with the significant predictors. By using forward selection to identify significant and non-significant pre- dictors of intra-workgroup interaction, the effects associated with statistical artefacts are reduced in addition to reducing indirect and spurious effects associated with variables lacking explanatory power.

FINDINGS Table 4 summarises the multivariate regres- sion results for intra-workgroup interaction. With regard to the three-equation model for intra-workgroup interaction, the first regres- sion equation shows that workgroup size is a significant predictor of minority employ- ment disparity (β = 0.461/0.461, p < 0.001). In this case, higher rates of minority employ- ment disparity are associated with larger workgroups within the NYPD’s police and civilian workforce. Position status also is a sig- nificant predictor of minority employment disparity (β = 9.817/0.571, p < 0.001), indi- cating that there are fewer minorities than Whites in supervisory/management posi- tions within the NYPD’s police and civilian workforces. The results also show that type of workgroup (ie, police and civilian) does not predict the level of disparity among minori- ties and Whites within the NYPD, suggesting that both the civilian and police workforces are just as likely to have low or high levels of minority employment disparity.

With regard to the level of ethnic and racial diversity within NYPD workgroups, the results show that type of workgroup and position status are significant predictors of workforce diversity. With regard to type of

Guajardo

Page 217

workgroup, police workgroups have higher levels of workforce diversity than civilian workgroups (β

WS = 15.375/0.757, p < 0.001).

However, the position status of the work- group has an inverse relationship with work- force diversity (β

PS = –10.384/–0.547, p <

0.001). Specifically, workgroups composed of supervisory or management-based positions are less diverse than those that are not. Con- trary to prior research, workgroup size does not influence diversity in either the police or civilian workforces. Table 4 also shows that the level of minority employment disparity within police and civilian workgroups does not influence the level of diversity.

With respect to intra-workgroup inter- action, Table 4 shows that workgroup size is a significant predictor, whereas minor- ity employment disparity and workforce diversity are significant mediating variables. Specifically, workgroup size is positively related to intra-workgroup interaction (β

= 0.795/0.795, p < 0.001), indicating that larger workgroups tend to encourage inter- actions among White and minority employ- ees. Intragroup interaction is also influenced positively by the level of diversity within the

workgroup (β WD

= 0.376/0.376, p < 0.001). Conversely, the level of minority employment disparity that exists within the workgroup is inversely related with intra-workgroup inter- action (β

MED = –0.154/–0.154, p < 0.01).

Figure 2 summarises the direct and indirect effects of the independent and mediating vari- ables on intra-workgroup interaction. Minor- ity employment disparity has a negative direct effect on intra-workgroup interaction (β

MED

= –0.154/–0.154, p < .01), whereas work- force diversity has a positive effect on intra- workgroup interaction (β

WD = 0.376/0.376,

p < 0.001). The size of the workgroup also has a positive direct effect on intra-workgroup interaction (β

WS = 0.795/0.795, p <0.001). In

addition, Figure 2 shows a negative indirect effect of workgroup size and minority employ- ment disparity (–0.071 = β

WS = 0.461 × β

MED

= –0.154), suggesting that larger workgroups with higher levels of employment disparity among Whites and minorities have lower lev- els of intra-workgroup interactions. Position status and minority employment disparity also have a negative indirect effect on intra-work- group interaction (–1.512 = β

PS = 9.817 ×

β MED

= –0.154), suggesting that supervisory or

Table 4: Regression results for intra-workgroup interaction

Dependent variables

Independent and moderating variables

Minority Employment Disparity (MED)

Workforce Diversity (WD) Intra-workgroup Interaction (WI)

Intercept 3.163 ns 9.149** –0.251 ns

Workgroup Size (WS) 0.461/0.461** 0.795/0.795**

Workgroup Type (WT) 15.375/0.757**

Position Status (PS) 9.817/0.517** –10.384/–0.547**

MED –0.154/–0.154*

WD 0.376/0.376**

R2 0.44 0.40 0.93

Adjusted R2 0.42 0.37 0.92

N 50 50 50

ns, not significant at α = 0.01; * significant at α = 0.01; ** significant at α = 0.001.

Workforce diversity

Page 218

management-based positions with higher lev- els of employment disparity have lower levels of intra-workgroup interaction. Figure 2 also shows a positive indirect effect among type of workgroup and workforce diversity on intra-workgroup interaction (5.781 = β

WT =

15.375 × β WD

= 0.376), suggesting that police workgroups with higher levels of diversity have higher levels of intra-workgroup interaction. However, position status and diversity have a negative indirect effect on intra-workgroup interaction (–3.904 = β

PS = –10.384 × β

WD =

0.376). The path analysis shows that minority employment disparity and workforce diversity have a direct effect on the intra-workgroup interaction and serve as mediating variables for workgroup size, position status, or type of workgroup.

DISCUSSION The findings suggest that the level of intra- workgroup interaction within the NYPD is a function of workforce diversity, minor- ity employment disparity and the size of the workgroup. Whereas workforce diversity and the size of the workgroup have a posi- tive effect on intra-workgroup interaction, the disparity among the number of Whites and minorities within a workgroup reduces the level of interaction among the employ- ees. This may result from White and minority group members distancing themselves from

each other. This may also result from a work environment in which Whites and minorities do not have a need to interact in order to perform duties.

The findings also suggest that workforce diversity is a mediating variable for type of workgroup and the status of the position within the NYPD organisational structure. As a mediating variable variable, workforce diversity lessens the negative effect that management-based positions have on intra- workgroup interaction. This suggests that management-based positions with higher levels of diversity have higher levels of inter- action among Whites and minorities than those with lower levels of diversity. When minority employment disparity serves as a mediating variable, it has a negative effect on the size of the workgroup and status of the position on intra-workgroup interaction because the ethnic or racial composition of the workgroup would limit the opportunities for Whites and minority members to com- municate and interact among each other.

Similar to McCluskey and McCluskey (2004), the results produced a non-significant finding for size of the workgroup and work- force diversity for the NYPD. This suggests that the size of the police and civilian work- group does not influence or predict whether it will have low or high levels of ethnic or racial homogeneity. Accordingly, small and large police and civilian workgroups are just

* Significant at α = 0.01 ** Significant at α = 0.001

15.375/0.757 **

–10.384/–0.547 **

0.461/0.461 **

0.795/0.795 **

9.817/0.517 **

Workgroup Size

Workgroup Type

Minority Employment

Disparity

Workforce Ethnic and Racial Diversity

Intra-workgroup Interaction

Position Status

0.376/0.376 **

–0.154/–0.154 *

Figure 2 Casual model for minority

employment disparity, diversity and intra-workgroup

interaction

Guajardo

Page 219

as likely to have similar levels of minority integration in which the degree of homo- geneity is not affected. The size of the work- group, however, is likely to impact the level of interaction among members within a workgroup.

CONCLUSION Although this study assessed intra-work- group interaction within the NYPD and its police and civilian workgroups, a larger study of police agencies is needed to validate the model because several of the independent variables produced non-significant results. A larger study might also identify additional variables that may influence intra-workgroup interactions among police officers and civil- ians so that the casual model is refined and expanded. In addition, a larger study may address the statistical issues confronted in this study more adequately because the analysis is likely to be conducted at an aggregate level rather than focusing on workgroups within a single police agency. In order to understand the direct and indirect effects of minor- ity integration and diversity on inter- and intra-workgroup interaction within police agencies more fully, future research needs to sample police agencies at local and state levels.

references Blau, P. M. (1977). Inequality and heterogeneity.

New York, NY: Free Press. Choi, S. (2010). Diversity in the U.S. federal

government: antecedents and correlates of diversity in federal agencies. Review of Public Personnel Administration, 30, 301–321.

Fridell, L. A. (2004). By the numbers: A guide for analyzing race data from vehicle stops. Washington, DC: Police Executive Research Forum.

Guajardo, S. A. (1999). Workforce diversity: monitoring employment trends in public organizations. Public Personnel Management, 28, 63–85.

Harrington, P., & Lonsway, K. A. (2004). Current barriers and future promise for women in policing. In B. R. Price & N. J. Sokoloff (Eds.), The criminal justice system and women (pp. 495–510). New York, NY: McGraw-Hill.

Hochstedler, E., Regoli R. M. & Poole, E. D. (1984). Changing of the guard in American cities: a current empirical assessment of integration in twenty municipal police departments. Criminal Justice Review, 9, 8–14.

Kellough, J. E. (1990). Integration in the public workplace: determinants of minority and female employment in federal agencies. Public Administration Review, 50, 557–564.

Kim, P. S. (1993). Racial integration in the American federal government with special reference to Asian Americans. Review of Public Personnel Administration, 13, 52–66.

Lieberson, S. (1969). Measuring population diversity. American Sociological Review, 34, 850–862.

Martin, S. E. (2004). The interactive effects of race and sex on women police officers. In B. R. Price & N. J. Sokoloff (Eds.), The criminal justice system and women (pp. 527–542). New York, NY: McGraw-Hill.

McClauley, C., Plummer, M., Moskalenko, S., & Mordkoff, J. T. (2001). The exposure index: a measure of intergroup contact. Journal of Peace Psychology, 7, 321–336.

McCluskey, C. P., & McCluskey, J. D. (2004). Diversity in policing: Latino representation in law enforcement. Journal of Ethnicity and Criminal Justice, 2(3), 67–81.

Pitts, D. W. (2005). Diversity, representation, and performance: evidence about race and ethnicity in public organizations. Journal of Public Administration Research and Theory, 15, 615–631.

Pitts, D. W. (2009). Diversity management, job satisfaction, and performance: evidence from U.S. federal agencies. Public Administration Review, 62, 328–338.

Schulz, D. M. (2004). Invisible no more: a social history of women in U.S. policing. In B. R. Price & N. J. Sokoloff (Eds.), The criminal justice system and women (pp. 483–494). New York, NY: McGraw-Hill.

Simpson, E. H. (1949). Measuring diversity. Nature, 163.

Workforce diversity

Page 220

Smith, B. W. (2003). The impact of police officer diversity on police-caused homicides. Policy Studies Journal, 31, 147–162.

Starks, G. L. (2009). Minority representation in senior positions in U.S. federal agencies: a paradox of underrepresentation. Public Personnel Management, 38, 79–90.

Stokes, L. D., & Scott, J. F. (1996). Affirmative action and selected minority groups in law enforcement. Journal of Criminal Justice, 24, 29–38.

Walker, S. (1983). Employment of Black and Hispanic police officers. Review of Applied Urban Research, 5, 1–6.

Walker, S. (1989). Employment of Black and Hispanic police officers, 1983–1988: A follow-up study. Omaha: University of Nebraska at Omaha.

Walker, S., & Turner, K. B. (1992). A decade of modest progress: Employment of Black and Hispanic police employment in the 50 largest cities, 1983–1992. Omaha: University of Nebraska at Omaha.

Winters, C. A. (1991). Hispanics and policing in Chicago and Cook County, Illinois. The Police Journal, 64, 71–76.

Copyright of International Journal of Police Science & Management is the property of Vathek Publishing Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.