A Psychometric Evaluation of the Integrity Profile 200 (IP 200) and

Grobler - 221

A Psychometric Evaluation of the Integrity Proﬁle 200 (IP 200) and the

Adaptation Thereof for Use in the South African Police Service

Anton Grobler

Renier Steyn

University of South Africa, South Africa

Abstract

Integrity is an essential element of orderly co-existance and an important facet of professional policing. The objective of this study

was to psychometrically evaluate the Integrity Proﬁle 200 (IP 200), an instrument widely used in South Africa, to assess its utility

as an integrity measure for use in the South African Police Service. Although the literature indicates that the constructs measured

with the IP 200 are sound, an exploratory investigation in the South African Police Service reveals poor reliability and unacceptable

inter-item correlations. This suggests poor factorial validity (model ﬁt). Subsequently an exploratory factor analysis (N = 1457) was

done to adapt the scale and improve the model ﬁt. Four factors were extracted and analysed, and satisfactory psychometric properties

were found for these factors, including the absence of race-based item bias. The factors are: (i) Integrity restricting orientation; (ii)

Moral conscientiousness and accountability; (iii) Organisational/management integrity, and (iv) Lie scale. The results account for a

signiﬁcant deviation and simpliﬁcation from the original instrument structure. It is recommended that the original IP 200 should not

be used in this context but that the adapted scale be used.

Introduction

Integrity in a policing context has speciﬁc importance because the relationship with the public is of

an asymmetrical nature owing to the elements of power and authority (Benson, 2010). Furthermore Walkin

(2005) is of the opinion that the people in professions that provide a service have a higher social purpose,

and that they ought to be good people. Gottschalk (2010), Lee and Vaughn (2010), Miller (2010), and Wolfe

and Piquero (2011) suggest that the public has social expectations of the police to set an example and that the

integrity of police oﬃcials should be above suspicion. Burger (1999) and the Oﬃce of Police Integrity (2009)

indicate that police integrity should be broadly deﬁned on the basis of related concepts, namely principleness,

creditworthiness, reliability, pure-mindedness, worthiness, honesty, ethics, incorruptibility, conscientiousness

and standards. Benson (2010, p. 190) deﬁnes police integrity as “the tendency of a police oﬃcial to resist the

temptation to abuse the powers and privileges of his position”.

Although high integrity may be expected of police oﬃcials, divides in integrity-deviating behaviour

often occur (Benson, 2010; Gottschalk, 2010; Huberts, Kaptein & Lasthuizen, 2007; Lee & Vaughn, 2010;

Schafer & Martinelli, 2008; Wright, 2010). Haarr (1997), Lee and Vaughn (2010), and Miller (2010) divide

integrity-deviating behaviour by police oﬃcials into four categories: work avoidance and manipulation, deviant

behaviour by employees against the organisation, employee deviance in the interests of the organisation, and

informal remuneration. Deviant behaviour also occurs in the South African Police Service (SAPS), as reﬂected

in many newspaper reports and the damming report of Holtzhauzen (2001). Benson (2010), Burger (1999) and

Grobler (2003) have identiﬁed possible reasons for a lack of integrity in SAPS members. These include inad-

equate and insuﬃcient policy and procedures, managers whose behaviour is not in line with the SAPS code of

ethics, the inability of the organisation to fulﬁl the lifestyle aspirations of it members, and inadequate selection

and recruitment.

According to Benson (2010) and Grobler (2003), the SAPS have responded to the issue of integrity-de-

viating behaviour. In an attempt to deliver a responsible, eﬀective and high quality service with honesty and

Grobler - 222

integrity, a corruption and fraud prevention plan, disciplinary regulations, as well as an ethical code were intro-

duced. Various writers (Kaptein & Avelino, 2005; Miller, 2010; Svensson, Wood & Callaghan, 2010) support

the institutionalisation of an ethical code. Arnold and Lampe (1999) and Miller (2010) maintain that this is

often regarded as the ﬁrst step in addressing integrity and ethical behaviour especially with regard to the organ-

isation’s clients. In spite of these initiatives and actions to control deviant behaviour, for example corruption in

the SAPS, the SAPS is seen as one of the least trustworthy organisations in South Africa (Benson, 2010). The

process of addressing integrity-limiting behaviour in an organisation, and thereby improving integrity in the

organisation, requires continuous attention (Arnold & Lampe, 1999; Miller, 2010; Schafer, 2010b). Self-ex-

amination is a ﬁrst step in addressing integrity in an organisational context (Kaptein & Avelino, 2005; Oﬃce

of Police Integrity, 2009; Sellbom, Fischler & Ben-Porath, et al, 2007). Self-examination will, among other

things, provide information regarding possible risk areas and individuals, and assist in the management and

development and implementation of integrity interventions. If these risk areas and individuals cannot be re-

moved from the organisation, an active reorientation with regard to the organisation’s values and norms should

be considered (Gruys, Stewart, Goodstein, Bing & Wicks, 2008; Lee & Vaughn 2010; Miller, 2010; Oﬃce of

Police Integrity, 2009). This self-examination relates to the need for a measure of integrity.

The measurement of integrity, however, is a controversial subject in organisational and industrial psychol-

ogy (Congress of the US, 1990; Huberts, et al, 2007; Sellbom, et al, 2007), mainly because there is little or no

research to support the instrumentation (Congress of the US, 1990; Murphy, 1995). It is, however, important

to have a reliable and valid instrument available for measuring integrity (Congress of the US, 1990; Kaptein &

Van Reenen, 2001; Murphy, 1995) if one wishes to predict integrity-related behaviour (Sellbom, et al, 2007).

A further controversy related to integrity measures is that these measures historically discriminate against

minority groups (Congress of the US, 1990). This could prove to be a pertinent issue in the South African

context.

Some integrity measuring instruments focus on operational integrity and are based on situational judge-

ment measurements (Gottschalk, 2010; Miller, 2010; Oﬃce of Police Integrity, 2009). The development of

such tests is very costly, and appropriate instruments are not readily available in South Africa. In order to

address the issue of the measurement of integrity in the SAPS, a more readily available instrument of measure-

ment needs to be found and used. Although many traditional psychometric tests are available, no such measur-

ing instrument for integrity has yet been standardised for use in the SAPS (Grobler, 2003). The IP 200 ques-

tionnaire, a questionnaire readily available and often used in South Africa, and the integrity model on which

it is based, have also not yet been empirically evaluated to verify that it explains integrity as a construct in the

SAPS (Grobler, 2003). Also, no other integrity measures have been standardised in the SAPS. To address this

gap, the objectives of this study are to administer the IP 200 to assess its reliability and factorial validity, and to

evaluate the instrument in terms of race bias. Factorial validity will be a proxy for the model ﬁt of the instru-

ment.

Integrity measuring instrumentation should be standardised and interpreted speciﬁcally in terms of the or-

ganisation, because what may be acceptable in one organisation or country will not necessarily be acceptable in

another (Department of Public Service and Administration, 2001; Kaptein & Van Reenen, 2001). Gottschalk

(2010) and Huberts, et al (2007) support this notion, and they suggest that, because integrity is such an import-

ant concept in policing, integrity-measuring instruments should be evaluated diﬀerently to other professions

and should be organisation-speciﬁc.

Having an eﬀective and comprehensive measure of integrity may be a forerunner to taking many correc-

tive actions (Sellbom, et al, 2007). Gottschalk (2010), Huberts, et al (2007), Kaptein and Van Reenen (2001)

and Miller (2010) support Arnold and Lampe’s (1999) view that a self-investigation by means of an integrity

measure is essential for any integrity intervention in order to develop an integrity proﬁle for every individual

and group in the organisation. Melamed and Jackson (1995) as well as the Oﬃce of Police Integrity (2009) are

of the opinion that an organisation’s values and creditworthiness can be determined by the degree to which it is

Grobler - 223

prepared to address the internal endemically, and propose that valid and reliable measuring instruments (psy-

chometric instruments) can play an important role in this process (2010).

Method

Research Design

A survey design was chosen for the research. This speciﬁc design entails a cross-sectional design, where a

sample of the population is studied in the same time period (Shaughnessy & Zechmeister, 1997). This design is

applicable when an attempt is being made to determine the relationships between the variables in a population.

According to Shaughnessy and Zechmeister (1997), this design is ideal for descriptive and predictive functions.

A survey design was used for the empirical investigation. According to de la Rey (1978), this method is

a backward look where the researcher gathers information on a particular phenomenon and then processes it

over time. On the basis of this information the course of the phenomenon is determined and the factors that

inﬂuence its course are established. The survey method will be based on the administration of questionnaires,

which, according to Dooley (1984), is the method most used in the social sciences, and which Glick (1985)

and Xenikou and Furnham (1996) propose as the best method for measuring organisational climate.

Participants

Two groups of participants were used in the study. The ﬁrst group (later called the exploratory group)

consisted of 402 participants, from two police stations, namely Alexandra and Kathlehong. The IP 200 was

administered on them. Following the unsatisfactory results achieved with this sample, a much larger sample

(N =1,776) was drawn in order to perform the required exploratory factor analysis. This involved 14 police

stations. The 14 stations were those identiﬁed by political leaders as part of a larger renewal strategy in SAPS,

which included looking at issues of integrity. The police stations consisted of large, medium and small stations.

Measuring Instrument

The Integrity Proﬁle 200 was selected as instrument by the institutional sponsor (SAPS Management) due

to its face validity appropriateness to the law enforcement environment. The test distributor granted permis-

sion to conduct further research to adapt / customise the instrument for the environment. The Integrity Proﬁle

200 (1999) is based on a model developed by scientists from various disciplines in the social sciences and the

humanities (Fick, 2001). The most important step in the development of the model was deﬁning the construct

widely enough so that all aspects of integrity were covered (Fick, 2001; Integrity Proﬁle 200, 1999). These

constructs cover the broad spectrum of integrity, from the development, correction or remediation of integrity,

to the defence mechanisms that are used to justify behaviour that violates integrity. From this Fick (2001) de-

ﬁnes nine substructures of integrity, which are represented in the Integrity Proﬁle 200. These are socialisation,

reliability, creditworthiness, work ethic, dysfunctional behaviour, integrity limiting attitude, values, manipulative

abuse of power, support-transformation, and management integrity. Numerous resemblances between Fick’s

model and literature could be found, notably socialisation (Bayram, Gursakal & Bilgel, 2009; Miller, 2010;

Schafer, 2010a; Wolfe & Piquero, 2011), reliability (Gruys, et al, 2008; Lee & Vaughn 2010; Simons, 1999;

South African Police Service, 2002), creditworthiness (Benson, 2010; Lee & Vaughn 2010; Oﬃce of Police

Integrity, 2009; Republic of South Africa, 1997; Van Aswegen & Engelbrecht, 2009), work ethic (Babović,

2000; Gruys, et al, 2008; Schafer & Martinelli, 2008; Svensson, et al, 2010), dysfunctional behaviour (Lee &

Vaughn 2010; Sellbom, et al, 2007; Trevinyo-Rodriguez , 2007; Van Duyne, Stocco, Bajovic, Milenović, &

Lojpur, 2010), integrity limiting attitude (Benson, 2010; Lee & Vaughn 2010; Miller, 2010; Trevinyo-Rodri-

guez, 2007), values (Holtzhauzen, 2001; Matei, Matei & Savulescu, 2010; Svensson, et al, 2010; Van Duyne,

et al, 2010), manipulative abuse of power (Huberts, et al, 2007; Kempen, 2001; Lee & Vaughn 2010; Schafer,

2010a), support-transformation (Burger, 1999; Hkulalwa, 2001; Van Aswegen & Engelbrecht, 2009; Wolfe

& Piquero, 2011), and management integrity (Huberts, et al, 2007; Schafer, 2010a, 2010b; Sechrest & Burns,

1992; Sellbom, et al, 2007). It is clear from the aforementioned that many authors on integrity include factors

or elements of integrity as conceptualised by the IP 200 Integrity Model (Fick, 2001) in their writings. It would

Grobler - 224

thus appear that this model is embedded in present literature on integrity, and that it explains the concept

broadly enough as to probably be of use in the SAPS.

The IP 200 is an integrated instrument consisting of nine substructures that have been speciﬁcally devel-

oped to measure the multidimensionality of integrity. Each substructure is measured with 20 items. The instru-

ment also contains a monitoring scale, consisting of a lie scale (10 items) and a consistency scale comprising

a combination of 20 items (Integrity Proﬁle 200, 1999; 2000). The respondents provided answers on a 5 point

Likert Scale. Misrepresentation is often a problem encountered in most overt integrity tests (Murphy, 1995),

and a lie scale or consistency scale is therefore important for placing the results into perspective (Fick, 2001).

Previous research conducted in the United States of America indicates that the IP 200 has acceptable

psychometric characteristics. In order to determine reliability, a previous study (N = 2 200) made use of two

techniques, namely the half-split method and the test-retest method. The reliability coeﬃcients that were re-

ported were 0.92 and 0.90 respectively (Integrity Proﬁle 200, 1999; 2000). IP 200 scores were also compared

with the “Integrity Assessment Centre” (IAC) results of 211 individuals in supervisory positions. The reported

correlation coeﬃcient of 0.76 indicates a coeﬃcient of determination of 0.58, suggesting a 58% shared vari-

ance between the measures. A further study in which 384 individuals were evaluated using the IP 200, IP 200

scores correlated the socialisation and functionality vs. dysfunctional factors (0.73 and 0.79 respectively; Integ-

rity Proﬁle 200, 2000).

Statistical Analysis

The statistical analysis was carried out using the SAS program (SAS Institute, 2000). Cronbach’s alpha

coeﬃcients, inter-item correlations, analysis of variance, as well as explorative factor analysis were calculated.

The Cronbach’s alpha coeﬃcient indicates the proportional variance error of a particular scale, while the

inter-item correlation is a measurement of its internal consistency. According to Clark and Watson (1995), the

inter-item correlation is an important index for supplying information that supplements the alpha coeﬃcients of

a scale. Decisions will be guided by cut-oﬀ scores for reliability of α > 0.70 (Nunnally & Bernstein, 1994) and

an inter-item correlation of 0.15 < r < 0.50 (Clark & Watson, 1995).

An extension of Cleary and Hilton’s (1968) method was used to determine item bias (Van de Vijver &

Leung, 1997). Item bias analysis was undertaken through the use of linear regression and analysis of variance

procedures. The assumption is that an item is unbiased if persons from diﬀerent cultures, with the same stand-

ing on the theoretical construct that underlies the instrument, have the same score on the item (Meiring, Van de

Vijver, Rothman, & Barrick 2005).

Principle component analysis was carried out to ascertain the optimal number of factors and an oblique

rotation (Promax) was applied to crystallise the solution and to assist in assigning names to the factors. This

was done by means of SAS FACTOR.

Results

In the exploratory part of the study, the IP 200 was administered to 402 police oﬃcers at the Alexandra

and Kathlehong police station. The Cronbach’s alpha coeﬃcients and the inter-item correlations of the instru-

ment were not satisfactory. The alpha coeﬃcients of the substructures varied between 0.16 and 0.59, while

those of the subﬁelds varied between .09 and .73. The inter-item correlations varied between 0 and .73. These

statistics did not comply with the reliability (α > 0.70; Nunnally & Bernstein (1994) and inter-item correlation

(0.15 < r < 0.50); Clark & Watson, 1995) guidelines. This is an indication that the substructures and subﬁelds

did not have the required internal consistency (Grobler, 2003).

Owing to the poor results that were obtained for the substructures and the subﬁelds which were the basis

of Fick’s (2001) integrity model, the data obtained from the second group (N = 1,457) was subjected to an

exploratory factor analysis. In this group most of the participants held the rank of inspector, and have been pro-

moted in the past ﬁve years. The majority of the participants had more than 10 years’ service in the SAPS and

more than six years’ service in their current police station. The age group that was best represented was 31–40

years. The types of work done by the participants in the police stations showed a wide distribution between

Grobler - 225

client service centre, crime prevention, administration and detective services, as well as the so-called “other”

work, for example labourers and cleaners. The group under investigation consisted of 87.29% South African

Police Act staﬀ and 12.28% Public Service Act staﬀ. With regard to the highest level of training, most of the

participants had a Grade 12 qualiﬁcation, in other words, the minimum requirement for appointment as a func-

tional or uniform member. Most participants were Black (79.33%), and the oﬃcial language best represented

was Xhosa (16.93%), Venda (15.67%), Zulu (14.22), Tswana (14.22%) and Afrikaans (13.97%).

First-order factor analysis was carried out on all 200 variables (items) using the SAS FACTOR

(N=1457). Eleven factors with eigenvalues greater than one (1) were identiﬁed. The scree plot, however,

showed a sharp levelling oﬀ to the right after the eighth factor, hence only eight factors were analysed with

regard to factor loadings. The loadings on all eight factors were unacceptable (loadings < 0.35), and only four

factors, which explained 53% of the variance, were included in the ﬁnal factor structure. The inter-correla-

tions of the four-factor structure of the adapted IP 200 revealed acceptable interfactor correlations, being 0.36,

-0.14, -0.41, 0.29, -0.11, 0.36. This meets the guidelines of Nunnally and Bernstein (1994). As a result of

these relatively strong correlations (Cohen, 1988), the four factors were rotated by means of the oblique meth-

od using a Promax rotation in order to minimise the variance of the distinguishing factors (StatSoft, 1995).

In total 68 of the 200 variables (items) loaded on the four factors, F

, F

and F

and consisted of 26,

23, 9 and 10 items respectively. A cut-oﬀ point of 0.35 for inclusion in the interpretation of a factor was used.

The factor designations were developed by inspecting the items that loaded on the four factors, and where

they were grouped in the integrity model of the IP 200, which is based on Fick’s (2001) model. Inspection

showed that the items of F

were grouped largely under job behaviour, attitudes, dysfunctional behaviour and

the manipulative abuse of power, as described in the IP200. We called this factor integrity limiting orientation.

showed strong similarities with the IP 200 substructures of socialisation, reliability, work ethic and value.

We called this factor moral conscientiousness and accountability. The items for F

described the integrity of

management and the organisation. Items of the original subﬁelds of the IP 200, such as participation, social

responsibility, lifestyle aspirations and creditworthiness, were found in F

. We called this factor factor organi-

sational/management integrity. Lastly, F

agrees with the original lie scale of the IP 200, and was called as such.

The squared multiple correlations (SMC), which varied from 0.84 to 0.92, are an indication that the factors are

internally consistent and that they are well deﬁned by the variables (items). The communalities of F

, F

and F

are high, although F

only has average communalities. Fifty-three per cent (53%) of the variance is explained

by the four factors, with F

being the strongest factor, which explains 19.84% of the total variance and 37.40%

of the covariance.

In Table 1 the descriptive statistics, Cronbach’s alpha coeﬃcient and interim correlations of the adapted

IP 200 are shown.

Table 1

Descriptive statistics, Cronbach’s Alpha Coeﬃcient and Interitem correlations for the Adapted IP 200 (N =

1457)

No. of

items

Factor M SD Mean

Skewness Kurtosis α

(average)

26 F

47.69 11.45 1.83 0.39 0.89 0.83 0.16

23 F

38.00 10.50 1.65 1.80 6.48 0.85 0.21

9 F

21.58 5.04 2.40 0.03 0.59 0.73 0.23

10 F

31.34 5.87 3.13 -0.85 0.57 0.85 0.36

Factor designations: F

: Integrity limiting orientation; F

Moral conscientiousness and accountability; F

Or-

ganisational/management integrity, and F

Lie scale.

Mean score presented on a 5 point scale

Grobler - 226

Apart from F

(moral conscientiousness and accountability), the scores of the factors are normally distrib-

uted, although the skewness and kurtosis scores are within the parameters of 2 and 7 respectively (Hair, Ander-

son, Tatham & Black, 1998).

The Cronbach’s alpha coeﬃcient of the factors in Table 1 are acceptable if the guideline of α > 0.70

(Nunnally & Bernstein, 1994) is applied, as well as the inter-item correlations (0.15 < r < 0.50, Clark & Wat-

son, 1995). It would thus appear that the factors have suﬃcient levels of internal consistency.

The mean and standard deviation should serve as guidelines when the adapted IP 200 is used. With (i)

integrity limiting orientation, lower scores should be interpreted positively as integrity-limiting orientation refers

to a tendency to justify deviant behaviour by means of rationalisation and other defence mechanisms, to dis-

play work behaviour that is characterised by a deviation from the rules and the unfair handling of other people,

the misuse of power and position to one’s own advantage, actions characterised by unethical behaviour, and

emotional withdrawal and isolation. These factors concentrate mainly on the work environment and the ratio-

nalisation of deviant behaviour, which mostly concerns the satisfaction of needs at the cost of the organisation,

junior employees and colleagues. With the second factor, (ii) moral conscientiousness and accountability, lower

scores should be interpreted positively. A low score indicates that value is attached to membership of a group;

behaviour is characterised by self-pride and self-conﬁdence; there is the courage of one’s own convictions and

a high degree of responsibility and accountability, especially with regard to the attainment of goals. This factor

does not focus primarily on the work environment, but rather on general values with regard to the community

and individual characteristics in general. With (iii) organisational/management integrity, a low score indicates

that the organisation provides for the employees’ lifestyle aspirations and meets its social responsibilities; that

the management is creditworthy, encourages participation and is willing to change. This factor therefore mea-

sures the individual’s perception of and attitude towards the organisation as well as management. Lower scores

on organisational/management integrity would be interpreted positively. The fourth factor does not form part of

the integrity model. On (iv) the lie scale, a low score indicates the degree to which the respondent has distorted

the results, consciously or unconsciously. A high score should be interpreted positively.

The results of the item bias analysis, carried out by means of a variance analysis on the 68 items of the

adapted IP 200, were satisfactory. The bias analysis was done with regard to the Black, White and Coloured

race groups. The analysis could not be applied to the Indian race group because the group comprised fewer

than 50 members, which is regarded as the minimum by Van de Vijver and Leung (1997). It is thus clear that

the adapted IP 200 has no race group bias (Grobler 2003).

Discussion

Integrity is an essential element of orderly co-existence and an imperative facet of professional policing,

speciﬁcally in terms of the social expectations of the public (Benson, 2010; Gottschalk, 2010; Hertzog, 2000;

Lee & Vaughn, 2010; Miller, 2010; Sellbom, et al, 2007; Schafer & Martinelli, 2008; Wolfe & Piquero, 2011).

Integrity-deviating behaviour often occurs in police organisations, although high integrity is expected of police

oﬃcials (Benson, 2010; Gottschalk, 2010; Huberts, et al, 2007; Lee & Vaughn, 2010; Schafer & Martinelli,

2008; Wright, 2010). According to Benson (2010) and Grobler (2003), this is not only an international phe-

nomenon, but is also found within the South African context. Arnold and Lampe (1999), Kaptein and Avelino

(2005), Kaptein and Van Reenen (2001), Miller (2010), Oﬃce of Police Integrity (2009), and Sellbom, et al

(2007) regard self-examination (integrity audit) as a ﬁrst step in addressing integrity-limiting behaviour in an

organisational context.

In order to carry out any integrity audit, whether on an organisational or an individual level, it is neces-

sary to have a standardised integrity instrument (Congress of the US, 1990; Kaptein & Van Reenen, 2001;

Murphy, 1995; Sellbom, et al, 2007). The value of the utilisation of the IP 200 Integrity Model and the IP 200

questionnaire was determined within the SAPS context. It seems that the constructs of the IP 200 Integrity

Model are well represented in literature. Finding literature on the dynamic interplay between the constructs

Grobler - 227

was diﬃcult and would have been over-ambitious and beyond the scope of this paper. This could also be seen

as one of the limitations of this paper which could be explored by future researchers.

The IP 200 was administered and evaluated to assess the factorial validity (model ﬁt) of the instrument

in the South African context. The psychometric properties of the IP 200 were tested with signiﬁcantly poor

results. It was found that the substructures and the subﬁelds did not have the necessary internal consistency.

This lead to an explanatory factor analysis, with the intention to reveal latent structure of integrity as it pertains

to the South African population. After all 200 items of the IP 200 had been subjected to an exploratory factor

analysis, four factors that explained 53% of the variance were included in the ﬁnal factor structure. These fac-

tors are internally consistent (acceptable Cronbach’s alpha values and interim correlations) and are well deﬁned

by the variables deduced from high square multiple correlations (SMC) and item communalities.

The empirical results of the study suggest an integrity model consisting of four factors, which contrasts

the muti-dimensional literature based Fick (IP 200, 1999, 2000) model. The SAPS intergrity model two

distinct areas, namely an individual and organisational facet. The individual integrity consists of two factors,

namely integrity-limiting orientation (F

) and moral conscientiousness and accountability (F

), which are anal-

ogous to the work that Trevinyo-Rodriguez (2007) has done. The relatively low weighted average was obtained

for F

and F

(1.83/5 and 1.65/5 in Table 1) and is an indication that the respondents are indeed accountable

and have the necessary courage of their own convictions to act according to ethical principles, and are able

to justify this. These results support the view of Hertzog (2000), Kempen (2001), Lersch and Mieczkowski

(1996), Sechrest and Burns (1992) as well as Tswete and Selebi (2000), that there are just individuals in an

organisation who are guilty of unethical and deviant behaviour, and that such individuals do not represent the

norm.

In contrast to the IP 200 Integrity Model (Fick, 2001), in which the corporate facet comprises two fac-

tors, the adapted model has only one factor. This factor is called organisational/management integrity (F

). The

weighted average score (2.40/5 in Table 1) indicates that the organisation provides averagely for the employees’

lifestyle aspirations and meets its social responsibilities only relatively, and that management’s creditworthiness

is not without repute. The individuals’ perception of and attitude towards the organisation as well as manage-

ment is thus lukewarm.

The fourth factor that was empirically obtained is concerned with the measuring instrument and is not

part of the integrity model. A high weighted average was obtained, which is an indication that the participants

answered the questionnaire honestly.

Item bias with regard to the diﬀerent race groups (excluding Indians owing to poor representation in the

population) was carried out, with the result that no signiﬁcant diﬀerences between the race groups were report-

ed. It is thus clear that the adapted IP 200 contains no uniform or non-uniform bias for the race groups in the

SAPS.

The relatively positive results of the individual integrity factors are an indication that the average police

oﬃcial has the necessary integrity, as expected from the public. Interventions on individual level should still

be institutionalised, such as integrity testing or screening during the recruitment and selection process, regular

integrity testing of serving police oﬃcials to determine the organisation’s (and sub-unit) integrity proﬁle. This

could also be used for targeted integrity interventions on individual as well as unit level.

If these results are regarded as a self-investigation of integrity, as suggested by many authors in the intro-

duction, it would seem that the focus should be on organisational and management integrity, owing to the nega-

tive score obtained on this factor (See Table 1), and its impact on the indivual factors. Benson (2010), Huberts,

et al (2007), Lee and Vaughn, (2010), Miller (2010), Oﬃce of Police Integrity (2009), Peterson (2004), Scha-

fer and Martinelli (2008) and Schafer (2010a & b) indicate that there is a relationship between police oﬃcials’

behaviour and their perception of the organisation. If the individual’s opinion is that the organisation’s values

and norms (as contained in the code of conduct) are not visible in management behaviour, deviant behaviour

will be manifested, which could clash with the organisation and with the attainment of goals. A positive atti-

tude towards the organisation, management and the work of police oﬃcials will result in the individual desist-

Grobler - 228

ing from deviant behaviour, especially if it negatively aﬀects the image of the organisation and goal attainment

(Simons, 1999).

Simons (1999) maintains that management integrity is an essential component when an organisation is in

the process of transformation; in other words, all management and organisational processes should be modelled

on clear moral and ethical principles. According to Hertzog (2000), Kaptein and Van Reenen (2001), as well as

Sechrest and Burns (1992), these principles should be communicated clearly and unambiguously to all mem-

bers of the organisation. According to Burger (1999) and Kaptein and Van Reenen (2001), the example set

by the management of the organisation is the most important component of organisational integrity; in other

words, management behaviour should be directed by, for example, a code of ethics. In studies by Van Aswegen

and Engelbrecht (2009) as well as Wallace, Hunt and Richards (1999) it was determined that there is a direct

relationship between the way in which management demonstrates its values (management integrity) and organ-

isational climate, including trust, the perception of leadership support, cooperation, conﬂict and role ambiguity

in a police environment.

This study has speciﬁc shortcomings that should be taken into account. Reference was already made to

the absence of a discussion on the dynamic interplay between the constructs of the IP 200 model. This was

diﬃcult and beyond the scope of this paper. This could be explored by future researchers. A further matter

that needs to be addressed in future research is the investigation of diﬀerences in factor structures across ethnic

groups.

The measurement is based on self-reporting, which was not veriﬁed by means of objective criteria. The

inclusion of the lie scale somewhat compensated for it, but future researchers are urged to ﬁnd objective exter-

nal criteria. In many respects this research is only a ﬁrst step in a long process, and objective criteria should be

included in future studies. Finally, the population comprised only participants stationed at 14 police stations,

which are not representative of the SAPS as a whole. However, this research covered a broad range of police

stations and the standardisation data collected could be used as a starting point for future investigations.

Recommendations

The SAPS has an urgent need for an integrity measure, especially in terms of an “integrity thermometer”,

as Kaptein and van Reenen (2001) call it, as part of an organisational development and integrity intervention.

The adapted IP 200 can be applied as such a tool and can be tentatively (experimentally) used as a selection

tool for police recruits, as well as an integrity barometer in identiﬁed units, sections or police stations in order

to develop integrity proﬁles for these.

One area in this study that has been identiﬁed as a problem in the 14 police stations are the negative

perceptions of management/organisational integrity. It is recommended that interventions should be developed

so that members have certainty with regard to expectations, that they see management as being consistent, that

goal setting is realistic and attainable, that members obtain support from the organisation and management, and

that ongoing feedback occurs through open communication and remuneration as well as through recognition.

These interventions would be in line with the seven-factor development model of Kaptein and Van Reenen

(2001), and the writings of Burger (1999), Hertzog (2000), Sechrest and Burns (1992), as well as Wallace, et

al (1999).

References

Arnold, V., & Lampe, J. C. (1999). Understanding the factors underlying ethical organizations: Enabling continuous

ethical improvement. Journal of Applied Business Research, 15, 1-20.

Babović, B. (2000). Police brutality or police torture. Policing: An International Journal of Police Strategies & Manage-

ment, 23, 97-104.

Bayram, R., Gursakal, N., & Bilgel, N. (2009). Counterproductive work behavior among white-collar employees: A

study from Turkey. International Journal of Selection and Assessment, 17, 180-188.

Benson, B. (2010). Police corruption: What is the plan? Acta Academia, 42, 188-208.

Burger, J. A. P. (1999). Maintaining police service integrity: Towards an integrated framework against abusive actions

Grobler - 229

such as corruption. Unpublished document, South African Police Service.

Clark, L. A., & Watson, D. (1995). Constructing validity: Basic issues in objective scale development. Psychological

Assessment, 7, 309-319.

Cleary, T. A., & Hilton, T. L. (1968). An investigation of item bias. Educational and Psychological Measurement, 28,

61-75.

Cohen, J. (1988). Statistical power analysis for behavioral science (Rev. ed.). Orlando, FL: Academic Press.

Congress of the U.S. (1990). The use of integrity tests for pre-employment screening. Washington, DC: U.S. Government

Printing Oﬃce.

De la Rey, R. P. (1978). Statistiese metodes in sielkundige navorsing. Pretoria, South Africa: University of Pretoria.

Department of Public Service and Administration. (2001). Public service anti-corruption strategy. Unpublished document.

Fick, L. (2001). IP 200: Premises-Assumptions. Unpublished report.

Gottschalk, P. (2010). Crime-based survey instrument for police integrity measurement. Policing: An International Jour-

nal of Police Strategies & Management, 33 91, 56-68.

Grobler, A. (2003). Die rol van organisasieklimaat en disposisionele veranderlikes op integriteit in die Suid-Afrikaan-

se Polisiediens (SAPD. Unpublished PhD Thesis: Potchefstoom University for Christian Higher Education (PU for

CHE).

Gruys, M. L., Stewart, S. M., Goodstein, J., Bing, M. N., & Wicks, A. C. (2008). Values enactment on organisations: A

multi-level examination. Journal of Management, 34, 806-842.

Haarr, R. N. (1997). “They’re making a bad name for the department.” Exploring the link between organizational

commitment and police occupational deviance in a police patrol bureau. Policing: An International Journal of Police

Strategies & Management, 20, 130-159.

Hair, F. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Multivariate data analysis (5

ed.). Upper Saddle

River, NJ: Prentice-Hall.

Hertzog, S. (2000). Deviant organizational messages among suspect police oﬃcers in Israel. Policing: An International

Journal of Police Strategies & Management, 23, 416-438.

Hkulalwa, K. (2001). Principle based leadership. Servamus, 94, 52-53.

Holtzhausen, E. (2001). Crime and corruption within the police ranks. Unpublished monthly internal newsletter:

Communication Services South African Police Service. (March 2001).

Huberts, L. G. J. C., Kaptein, M., & Lasthuizen, K. (2007). A study of the impact of three leadership styles on integrity

violations committed by police oﬃcers. Policing: An International Journal of Police Strategies & Management, 30,

587-607.

Integrity Proﬁle 200. (1999). Integrity Proﬁle 200: Interpreter. Unpublished technical report.

Integrity Proﬁle 200. (2000). Integrity Proﬁle 200: Users manual & various validity studies. Unpublished technical report.

Kaptein, M., & Avelino, S. (2005). Measuring corporate integrity: a survey based approach. Corporate Governance, 5,

45-54.

Kaptein, M. & Van Reenen, P. (2001). Integrity management in police organizations. Policing: An International Journal

of Police Strategies & Management, 24, 281-300.

Kempen, A. (2001). Legal and (il)legal ways to earn money (when in the SAPS). Servamus, 94, 68-69.

Lee, M., & Vaughn, M. S. (2010). Organisational factors that contribute to police deadly force liability. Journal of Crimi-

nal Justice, 38, 193-206.

Lersch, K. M., & Mieczkowski, T. (1996). Who are the problem-prone oﬃcers? An analysis of citizen complaints.

American Journal of Police, 15, 23-44.

Matei, A., Matei, L., & Savulescu, C. (2010). Public integrity, economic freedom and governance performance: A

comparative study of the EU member states and acceding countries. Theoretical and Applied Economics, 17, 21-52.

Meiring, D., Van de Vijver, F. J. R., Rothman, S., & Barrick, M. R. (2005). Construct, item, and method bias of

cognitive and personality measures in South Africa. South African Journal of Industrial Psychology, 31, 1-8.

Melamed, T., & Jackson, D. (1995). Psychometric instruments: Potential beneﬁts and practical use. Industrial and Com-

mercial Training, 27, 11-16.

Grobler - 230

Miller, S. (2010). Integrity systems and professionsl reporting in police organisations. Criminal Justice Ethics, 29, 241-

257.

Murphy, K. R. (1995). Integrity testing. In N. Brewer & C. Wilson (Eds.), Psychology and policing (pp. 205-228).

Hillsdale, NJ: Lawrence Erlbaum.

Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric theory (3

ed.). New York, NY: McGraw-Hill.

Oﬃce of Police Integrity. (2009). A guide to building workplace integrity: Indicators and Practise. Unpublished document.

Victoria Police.

Peterson, D. (2004). Perceived leader integrity and ethical intentions of subordinates. The Leadership & Organizational

Development Journal, 25, 7-23.

Republic of South Africa. (1997). White paper on transforming the public service delivery. Pretoria, South Africa:

Government Printer.

SAS Institute. (2000). The SAS system for Windows: Release 8.01. Cary, NC: SAS Institute Inc.

Schafer, J. A., & Martinelli, T. J. (2008). First-line supervisor’s perceptions of police integrity. The measurement of

police integrity revisited. Policing: An International Journal of Police Strategies & Management, 31, 306-323.

Schafer, J. A. (2010a). The ineﬀective police leader: Acts of commission and omission. Journal of Criminal Justice, 38,

737-746.

Schafer, J. A. (2010b). Eﬀective leaders and leadership in policing: traits, assessment, development and expansion. Polic-

ing: An International Journal of Police Strategies & Management, 33, 644-663.

Sechrest, D. K., & Burns, P. (1992). Police corruption-the Miami case. Criminal and Justice Behavior, 19, 294-313.

Sellbom, M., Fischler, G. L., & Ben-Porath, Y. S. (2007). Identify the MMPI-2 predictors of police oﬃcer integrity and

misconduct. Criminal Justice Behavior, 34, 985-1004.

Shaughnessy, J. J., & Zechmeister, E. B. (1997). Research methods in psychology (4

ed.). New York, NY: McGraw-Hill.

Simons, T. L. (1999). Behavioral integrity as a critical ingredient for transformational leadership. Journal of Organiza-

tional Change Management, 12, 30-45.

South African Police Service. (2002). Strategic plan for the South African police service: 2002-2005. Reproduced

document, South African Police Service.

StatSoft (1995). Statistica for Windows. [computer program manual]. Tulsa, OK: StatSoft.

Svensson, G., Wood, G., & Callaghan, M. (2010). Measurement and structural properties of organizational codes of ethics in

private and public Sweden. International Journal of Public Sector Management, 23, 549-566.

Trevinyo-Rodriguez, R. N. (2007). Integrity: A systems theory classiﬁcation. Journal of Management History, 13, 74-93.

Tswete, S. V., & Selebi, J. S. (2000). Addressing racism in the SAPS and the transformation of the Dog Units. South

African Police Service Media release, Pretoria, 11 November.

Van Aswegen, A. S., & Engelbrecht, A. S. (2009). The relationship between transformational leadership, integrity and an

ethical climate in organisations. SA Journal for Human Resource Management. 7, 221-229.

Van de Vijver, F. J. R., & Leung, K. (1997). Methods and data-analysis for cross-cultural research. Thousand Oaks, CA:

Sage.

Van Duyne, P. C., Stocco, E., Bajovic, V., Milenović, M., & Lojpur, E. E. (2010). Searching for corruption in Serbia.

Journal of Financial Crime, 17, 22-46.

Walkin, M. M. (2005). Professional integrity. Available at: www.usafa.af.mil/core-value/integrity.html, viewed on 2011-

03-03.

Wallace, J., Hunt, J., & Richards, C. (1999). The relationship between organisational culture, organisational climate and

managerial values. The International Journal of Public Sector Management, 12, 548-564.

Wolfe, S. E., & Piquero, A. R. (in press). Organizational justice and police misconduct. Criminal Justice and Behaviour.

Wright, B. (2010). Civilization the “blue code”? An examination of attitudes to misconduct in the police extended family.

International Journal of Police Science and Management, 12, 339-356.

Xenikou, A. & Furnham, A. (1996). A correlational and factor analytic study of four questionnaire measures of

organizational culture. Human Relations, 49, 349-371.